A communication method and apparatus

Abstract

A communication method and apparatus are disclosed to reduce the resource requirements for training customized models. The method includes: a first network device receiving first request information from a second network device, the first request information being used to request a model corresponding to a first analysis service for a first terminal, the first request information including type information of the first terminal; the first network device obtaining type information of a second terminal based on the type information of the first terminal; and the first network device sending information about the model to the second network device, the model information being determined based on the type information of the second terminal.

Description

Technical Field

[0001] This application relates to the field of mobile communication technology, and in particular to a communication method and apparatus. Background Technology

[0002] Current mobile communication technologies support network devices in using artificial intelligence (AI) to obtain terminal communication configurations or perform terminal location. To optimize configuration or location performance, customized training can be used to obtain a machine learning (ML) model serving the terminal, which is then used to obtain the terminal's configuration or location. However, the customization process of ML models involves extensive model training, resulting in high resource requirements.

[0003] Currently, reducing the amount of resources required for training ML models is an urgent problem to be solved. Summary of the Invention

[0004] This application provides a communication method and apparatus for reducing the amount of resources required for training customized models.

[0005] Firstly, a communication method is provided. This method can be implemented by a first network device or a component within the first network device, which may also be referred to as a communication apparatus or network element. The first network element is, for example, a network data analytics function (NWDAF). The components in this application may include, for example, at least one of a chip, a chip system, a processor, a transceiver, a processing unit, or a transceiver unit. Taking the first network element as the executing entity as an example, the method can be implemented through the following steps:

[0006] A first network device receives a first request message from a second network device. The first request message is used to request a model corresponding to a first analysis service for a first terminal. The first request message includes the type information of the first terminal. The first network device obtains the type information of the second terminal based on the type information of the first terminal. The first network device sends the model information to the second network device. The model information is determined based on the type information of the second terminal.

[0007] Based on the first aspect, after receiving a first request message for a model corresponding to a first analysis service requested by a first terminal, the first network device can determine the type information of the second terminal based on the type information of the first terminal, and provide the second network device with model information determined based on the type information of the second terminal. That is, the first network device can provide the second network device with model information for the first terminal based on the type information of the second terminal. Subsequently, the second network device can provide services to the first terminal and / or the second terminal based on this model. Therefore, it is not necessary to train separate models for the first terminal and the second terminal, which can reduce unnecessary model training iterations and lower the resource requirements for training customized models.

[0008] In one possible implementation, the model can be used to provide services to both the first terminal and the second terminal. The first terminal can be one or more terminals, and the second terminal can be one or more terminals. If the first terminal consists of multiple terminals, all of them have the same type information, which is the type information of the first terminal. Similarly, if the second terminal consists of multiple terminals, all of them have the same type information, which is the type information of the second terminal.

[0009] Optionally, the model can also serve terminals other than the first terminal that have type information of the first terminal and / or terminals other than the second terminal that have type information of the second terminal.

[0010] In this application, the type information of the first terminal can also be replaced with first type information. That is, the type information of one or more terminals, including the first terminal, is first type information. The type information of the second terminal can also be replaced with second type information. That is, the type information of one or more terminals, including the second terminal, is second type information. In other words, the above model can serve terminals with first type information and / or terminals with second type information. The terminal with first type information includes the first terminal, and optionally may also include one or more terminals other than the first terminal. The terminal with second type information includes the second terminal, and optionally may also include one or more terminals other than the second terminal.

[0011] In this application, "the first request information is used to request the model corresponding to the first analysis service for the first terminal" can also be replaced or interpreted as: the first request information is used to request the model corresponding to the first analysis service for the first terminal, or the first request information is used to request support or permission for the model corresponding to the first analysis service for the first terminal.

[0012] The first request information can also be alternatively described as: the first request information is used to request the type information of the second terminal corresponding to the type information of the first terminal.

[0013] As an example, the type information of the first terminal is its model allocation code, and the type information of the second terminal is its model allocation code. The terminal type information can also be information such as equipment manufacturer identification or brand identification. That is, the terminal type information can specifically be one or more of the following: model allocation code, equipment manufacturer identification, and brand identification.

[0014] In one possible implementation, the first request information includes the analysis identifier of the first analysis service, and obtaining the type information of the second terminal based on the type information of the first terminal includes: the first network device obtaining the type information of the second terminal based on the type information of the first terminal and the analysis identifier.

[0015] Based on this implementation, there is a correspondence between the analysis identifier of the first analysis service, the type information of the first terminal, and the type information of the second terminal. The first network device can further determine the type information of the second terminal based on the analysis identifier of the first analysis service and the type information of the first terminal, thus accurately determining the type information of the second terminal. For example, for different analysis services, the type information of the first terminal can correspond to different type information of the second terminal, thereby achieving flexible determination of the correspondence between terminal type information.

[0016] In one possible implementation, the hardware information of the first terminal is partially or entirely the same as the hardware information of the second terminal; and / or, the software information of the first terminal is partially or entirely the same as the software information of the second terminal.

[0017] Based on this implementation, the hardware and / or software information of the first terminal and the second terminal are partially or entirely the same, meaning that the same model can be used to serve multiple terminals with partially or entirely the same hardware and / or software information. Optionally, if there are multiple first terminals and multiple second terminals, the hardware information of any one of the multiple first terminals is partially or entirely the same as the hardware information of any one of the multiple second terminals; the software information of any one of the multiple first terminals is partially or entirely the same as the software information of any one of the multiple second terminals.

[0018] In one possible implementation, the model's input includes channel information between the first terminal and the access network device, and correspondingly, the model's output includes the communication configuration between the first terminal and the access network device. Similarly, the model's input includes channel information between the second terminal and the access network device, and correspondingly, the model's output includes the communication configuration between the second terminal and the access network device.

[0019] Based on this implementation, the model can be used to determine the communication configuration of terminals and / or access network devices. Since the hardware and / or condition information of the first terminal and the second terminal are partially or completely identical, if the channel information corresponding to the first terminal is the same as that corresponding to the second terminal, then the communication configuration of the first terminal and the second terminal are identical. Therefore, the same model can be used to serve both the first terminal and the second terminal.

[0020] In one possible implementation, the model's input includes channel information between the first terminal and the access network device, and correspondingly, the model's output includes the positioning configuration or location information of the first terminal. Similarly, the model's input includes channel information between the second terminal and the access network device, and correspondingly, the model's output includes the positioning configuration or location information of the second terminal. That is, the model can be used for terminal positioning. Since the hardware and / or conditional information of the first and second terminals are partially or completely identical, if the channel information corresponding to the first terminal is the same as that corresponding to the second terminal, then the positioning configuration of the first terminal and the positioning configuration of the second terminal are the same; or, if the location information of the first terminal and the location information of the second terminal are the same, then the same model can be used to serve both the first and second terminals.

[0021] In one possible implementation, the model's input includes channel information between the first terminal and the access network device, and correspondingly, the model's output includes the communication configuration of the first terminal and / or the communication configuration of the access network device. The communication configuration of the first terminal can be used for communication by the first terminal, and the communication configuration of the access network device can be used for communication between the access network device and the first terminal. Similarly, the model's input includes channel information between the second terminal and the access network device, and correspondingly, the model's output includes the communication configuration of the second terminal and / or the access network device. That is, the model can be used to determine the communication configuration of the terminal and / or the access network device. The communication configuration of the second terminal can be used for communication by the second terminal, and the communication configuration of the access network device can be used for communication between the access network device and the second terminal. Since the hardware and / or condition information of the first terminal and the second terminal are partially or completely identical, if the channel information corresponding to the first terminal is the same as the channel information corresponding to the second terminal, then the configuration information of the first terminal and the configuration information of the second terminal are the same. Therefore, the same model can be used to serve both the first terminal and the second terminal. It can also be understood that the access network device can use the same communication configuration to communicate with both the first terminal and the second terminal. It can also be understood that access network devices can use the same communication configuration to provide services to the first terminal and / or the second terminal.

[0022] Unless otherwise specified below, "communication configuration" can refer to the communication configuration between the terminal and the access network device. This communication configuration may include configurations configured on the terminal side, and / or, may include communication configurations configured on the access network device side. The terminal may use the communication configurations configured on the terminal side to communicate, and the access network device may use the communication configurations configured on the access network device side to communicate with the terminal or provide services to the terminal.

[0023] In one possible implementation, obtaining the type information of the second terminal based on the type information of the first terminal includes: a first network device sending the type information of the first terminal to a third network device; the first network device receiving the type information of the second terminal from the third network device, wherein the type information of the second terminal is determined based on the type information of the first terminal.

[0024] Based on this implementation, the first network device can send the type information of the first terminal to the third network device to obtain the type information of the second terminal from the third network device. For example, the third network device can store or configure the correspondence between the type information of the first terminal and the type information of the second terminal.

[0025] In one possible implementation, obtaining the type information of the second terminal based on the type information of the first terminal and the analysis identifier of the first analysis service includes: a first network device sending the type information of the first terminal and the analysis identifier to a third network device; the first network device receiving the type information of the second terminal from the third network device, wherein the type information of the second terminal is determined based on the type information of the first terminal and the analysis identifier.

[0026] Based on this implementation, the first network device can send the type information and analysis identifier of the first terminal to the third network device, in order to obtain the type information of the second terminal from the third network device. For example, the third network device can store or configure the correspondence between the type information of the first terminal, the analysis identifier of the first analysis service, and the type information of the second terminal. Therefore, the type information of the second terminal corresponding to the type information of the first terminal and the analysis identifier can be obtained from the third network device. The model of the analysis service can be applied to both the first terminal and / or the second terminal.

[0027] Optionally, the model can also be applied to terminals other than the first terminal that have type information of the first terminal and / or terminals other than the second terminal that have type information of the second terminal.

[0028] In one possible implementation, the information of the model is determined based on the type information of the second terminal, including: the information of the model includes the type information of the second terminal.

[0029] Based on this implementation, the model information sent by the first network device to the second network device may include the type information of the second terminal, indicating that the model can also be applied to the second terminal. For example, after receiving the first request information, the first network device can generate a model for the first terminal, which can serve the second terminal. Therefore, it is not necessary to repeatedly generate a model for the second terminal, thereby reducing the amount of resources required for model training. The model information may also include the model or model parameters. The second network device can then serve the second terminal based on the type information of the second terminal and the model.

[0030] In one possible implementation, the model information is determined based on the type information of the second terminal, including: the model is generated in response to second request information carrying the type information of the second terminal, or in other words, the model is generated based on the second request information, which is used to request the model corresponding to the first analysis service for the second terminal. After obtaining the type information of the second terminal based on the type information of the first terminal, the method further includes: the first network device determining the model based on the type information of the second terminal.

[0031] Based on this implementation, the first network device can generate a model for the second terminal after receiving the second request information, and after receiving the first request information and determining the type information of the second terminal based on the type information of the first terminal, use the model generated for the second terminal as a model that serves at least the first terminal and / or the second terminal. Therefore, it is not necessary to generate a separate model for the first terminal, thereby reducing the amount of resources required for model training.

[0032] The phrase “the second request information is used to request the model corresponding to the first analysis service for the second terminal” can also be replaced with or interpreted as: the second request information is used to request the model corresponding to the first analysis service for the second terminal, or it can be replaced with or interpreted as: the second request information is used to request support or permission for the model corresponding to the first analysis service for the second terminal.

[0033] In one possible implementation, the method further includes: a first network device receiving the second request information; the first network device may also generate the model in response to the second request message; the first network device may also obtain the correspondence between the type information of the first terminal and the type information of the second terminal from a third network device. Obtaining the type information of the second terminal based on the type information of the first terminal includes: the first network device obtaining the type information of the second terminal based on the type information of the first terminal and the correspondence.

[0034] Based on this implementation, the first network device can obtain the correspondence between the type information of the first terminal and the type information of the second terminal from the third network device after receiving the second request information. After receiving the first request information, the first network device can determine the type information of the second terminal corresponding to the type information of the first terminal based on this correspondence, thus achieving flexible determination of the type information of the second terminal. Alternatively, after receiving the second request information, the first network device can also send the type information of the first terminal to the third network device and receive the type information of the second terminal from the third network device, thereby obtaining the correspondence between the type information of the first terminal and the type information of the second terminal. Optionally, the first network device can also send the analysis identifier of the first analysis service to the third network device, wherein the type information of the first terminal is determined based on the type information of the second terminal and the analysis identifier. Optionally, the first network device can also store the correspondence between the type information of the first terminal and the type information of the second terminal.

[0035] In one possible implementation, the information of the model is determined based on the type information of the second terminal, including: the model is generated based on the training data of the second terminal, and the training data of the second terminal is obtained based on the type information of the second terminal.

[0036] For example, after determining the type information of the second terminal based on the type information of the first terminal, the first network device can generate the model based on the training data of the first terminal and the training data of the second terminal. Based on this implementation, the first network device can generate the model based on at least the training data of the second terminal after obtaining the type information of the second terminal based on the type information of the first terminal, enabling flexible determination of training data. Compared to generating a model based solely on the training data of the first terminal, it can support generating a model based on more training data, thereby improving model accuracy. Furthermore, the model can serve at least the first terminal and / or the second terminal. Therefore, the same model can be generated for terminals with different types of information, avoiding duplicate model generation and reducing the resource requirements for customized model training.

[0037] Secondly, a communication method is provided. This method can be implemented by a second network device or a component within the second network device, which may also be referred to as a communication apparatus or network element. The second network device can be an access network device serving a first terminal.

[0038] Taking the second network element as the executing entity as an example, the method can be implemented through the following steps: the second network device sends a first request information to the first network device. The first request information is used to request a model corresponding to the first analysis service for the first terminal. The first request information includes the type information of the first terminal. The second network device receives the information of the model from the first network device. The information of the model is determined according to the type information of the second terminal.

[0039] In one possible implementation, the second network device may also receive type information of the first terminal from the fourth network device, and obtain the analysis identifier of the first analysis service based on the type information of the first terminal, the analysis identifier being used to indicate that the first terminal is allowed to customize the model.

[0040] Based on this implementation, the second network device can receive the type information of the first terminal from the fourth network device. Additionally, the second network device can also obtain the type information of the first terminal based on information from the fourth network device. The fourth network device is, for example, an access and mobility management function (AMF) serving the terminal. The analysis identifier of the first analysis service can represent a model in which the first terminal is allowed or authorized to customize the first analysis service. Therefore, after determining that the first terminal is allowed or authorized to customize the model of the first analysis service, the second network device can send a first request message to the first network device.

[0041] In this application, the model that is allowed or authorized to customize the first analysis service can also be replaced with or interpreted as: the model that is allowed or authorized to customize the first analysis service, or the service provided by the customized model that is allowed or authorized to obtain the first analysis service, or the service provided by the customized model that is allowed or authorized to obtain the first analysis service, or the first analysis service that is allowed or authorized to obtain, or the first analysis service that is authorized or allowed to be provided to the first terminal or the terminals in the first terminal set, or the service provided by the customized model that uses the first analysis service.

[0042] In one possible implementation, the information of the model is determined based on the type information of the second terminal, including: the information of the model includes the type information of the second terminal.

[0043] In one possible implementation, the second network device can provide the first analysis service to the second terminal based on the type information of the second terminal carried in the model information. Therefore, after receiving a service request or registration request from the second terminal, the first analysis service can be provided to the second terminal based on the model without needing to request the model from the first network device again. Optionally, the service request or registration request may carry an analysis identifier for the first analysis service. Furthermore, the second network device may also provide the first analysis service to the second terminal based on the model after determining that the second terminal is allowed or authorized to customize the first analysis service. For an explanation of the model for which the second terminal is allowed or authorized to customize the first analysis service, please refer to the explanation of the model for which the first terminal is allowed or authorized to customize the first analysis service, which will not be repeated here.

[0044] Optionally, after receiving the type information of the second terminal and the identifier of the first analysis service from the fourth network device, the second terminal may determine whether it is allowed or authorized to customize the model of the first analysis service. Alternatively, after receiving the type information of the second terminal from the fourth network device, the second network device may send a fifth request message to the third network device. The fifth request message is used to request the analysis identifier of the analysis service corresponding to the second terminal, or the fifth request message is used to request the analysis identifier of the analysis service serving the second terminal, or the fifth request message can be used to request the analysis identifier of the analysis service to which the model customized by the second terminal belongs. The fifth request message includes the identifier and / or type information of the second terminal. The second network device may also receive the analysis identifier from the third network device, which is determined based on the identifier and / or type information of the second terminal.

[0045] Based on this implementation, the second network device can provide a first analysis service to the second terminal through the model, based on the type information of the second terminal carried in the model's information. That is, if the model's information includes the type information of the second terminal, the first analysis service can be provided to the second terminal through the model.

[0046] In one possible implementation, the information of the model is determined based on the type information of the second terminal, including: the model is generated based on the training data of the second terminal, and the training data of the second terminal is obtained based on the type information of the second terminal.

[0047] In this application, the training data for the terminal refers to data related to the terminal that can be used to train or generate models. The data for the terminal can originate from the terminal itself or from network devices, without specific limitations.

[0048] In one possible implementation, obtaining the analysis identifier based on the type information of the first terminal includes: a second network device sending third request information to a third network device, the third request information being used to request an analysis identifier corresponding to the analysis service of the first terminal, or the third request information being used to request an analysis identifier for an analysis service serving the first terminal, or the third request information being used to request an analysis identifier for an analysis service to which a model that the first terminal is allowed or authorized to customize belongs, the third request information including the identifier and / or type information of the first terminal; the second network device may also receive the analysis identifier from the third network device, the analysis identifier being determined based on the identifier and / or type information of the first terminal.

[0049] Based on this implementation, the second network device can obtain the analysis identifier of the first analysis service from the third network device according to the type information of the first terminal. Specifically, the second network device can determine, based on the analysis identifier obtained from the third network device, whether the first terminal is allowed or authorized to obtain the first analysis service; or, in other words, whether the second network device is allowed or authorized to customize the model of the first analysis service, or whether the customized model provided by the first terminal is allowed or authorized to obtain the first analysis service. Therefore, requesting a model for a terminal that is not allowed or authorized to customize a model can be avoided.

[0050] For example, the second network device can obtain the analysis identifier of the analysis service to which the model customized by the first terminal is permitted or authorized belongs from the third network device. If the analysis identifier of the analysis service to which the model customized by the first terminal is permitted or authorized belongs includes the analysis identifier of the first analysis service, it can be determined that the first terminal is permitted or authorized to obtain the first analysis service. In this application, the analysis identifier of the analysis service to which the model customized by the first terminal is permitted or authorized belongs can also be replaced or interpreted as: the analysis identifier of the analysis service to which the first terminal is permitted or authorized, the analysis identifier of the analysis service provided by the model customized by the first terminal is permitted or authorized, the analysis identifier of the analysis service to which the type information of the first terminal is permitted to obtain, the analysis identifier of the analysis service to which the customized model to which the type information of the first terminal is permitted to obtain belongs, and the analysis identifier of the analysis service provided by the customized model to which the type information of the first terminal is permitted to obtain.

[0051] In this implementation, the third network device can also be replaced by a fifth network device. The fifth network device can be the same as or different from the third network device.

[0052] In one possible implementation, the second network device may also receive type information of the first terminal and analysis identifier of the first analysis service from the fourth network device.

[0053] Based on this implementation, the second network device can determine, according to the analysis identifier of the first analysis service from the fourth network device, whether the first terminal is allowed or authorized to customize the model of the first analysis service, or whether the first terminal is allowed or authorized to obtain the service provided by the customized model of the first analysis service. Therefore, after determining that the first terminal is allowed or authorized to customize the model of the first analysis service, the second network device can send a first request message to the first network device.

[0054] In one possible implementation, the hardware information of the first terminal is partially or entirely the same as the hardware information of the second terminal; and / or, the software information of the first terminal is partially or entirely the same as the software information of the second terminal.

[0055] In one possible implementation, the second network device may also receive first instruction information before sending the first request information to the first network device. The first instruction information may be used to instruct the first terminal to provide a first analysis service according to a customized model. Optionally, the first instruction information may include an analysis identifier for the first analysis service, or other information.

[0056] Based on this implementation, the second network device can send a first request message to the first network device after receiving the first instruction information. The first instruction information can be sent by the first terminal. For example, the terminal can request a customized model service through the first instruction information. The instruction to provide a first analysis service to the first terminal based on the customized model can also be replaced with or interpreted as: requesting an AI service, requesting a first analysis service, requesting a high-performance analysis service (high performance, e.g., higher accuracy, lower energy consumption, or higher service response speed), requesting a high-performance first analysis service, requesting a more accurate analysis service, requesting a lower energy consumption analysis service, requesting a higher service response speed analysis service, requesting a first analysis service to be provided to the first terminal based on the customized model, requesting a customized model, indicating a model to obtain the first analysis service based on the training data of the first terminal, or requesting a customized first analysis service, etc.

[0057] The beneficial effects of the methods shown in the second aspect above and their various possible implementations can be found in the description of the first aspect and the corresponding implementations, and will not be repeated here.

[0058] Thirdly, a communication method is provided. This method can be implemented by a third network device or a component within a third network device, which may also be referred to as a communication apparatus or network element. The third network device can be a storage network element; for example, it can be a unified data management (UDM), unified data repository (UDR), analytics data repository functional (ADRF), or equipment identity check (EIR) network element or function.

[0059] Taking a third network device as an example, the method may include the following steps: the third network device receives type information of a first terminal from a first network device; the third network device sends type information of a second terminal to the first network device according to the correspondence between the type information of the first terminal and the type information of a second terminal.

[0060] In one possible implementation, the third network device may also receive an analysis identifier from the first analysis service of the first network device. The step of sending the type information of the second terminal to the first network device according to the correspondence between the type information of the first terminal and the type information of the second terminal includes: sending the type information of the second terminal to the first network device according to the analysis identifier and the correspondence between the type information of the first terminal and the type information of the second terminal.

[0061] In one possible implementation, the third network device may further receive third request information, which is used to request an analysis identifier corresponding to the analysis service of the first terminal, or the third request information is used to request an analysis identifier of the analysis service serving the first terminal, and / or the third request information may be used to request an analysis identifier of the analysis service to which the model customized by the first terminal is permitted or authorized, the third request information including the identifier and / or type information of the first terminal; the third network device may also send the analysis identifier, which is determined based on the identifier and / or type information of the first terminal.

[0062] In one possible implementation, the third network device may also obtain indication information of the mapping relationship. For example, the third network device may configure or receive the mapping relationship from other network devices.

[0063] In one possible implementation, the hardware information of the first terminal is partially or entirely the same as the hardware information of the second terminal; and / or, the software information of the first terminal is partially or entirely the same as the software information of the second terminal.

[0064] The beneficial effects of the methods shown in the third aspect above and their various possible implementations can be found in the description of the first aspect and the corresponding implementations, and will not be repeated here.

[0065] Fourthly, a communication method is provided. This method can be implemented by a second network device or a component within the second network device, which may also be referred to as a communication apparatus or network element. Taking the second network element as the executing entity as an example, the method can be implemented through the following steps: The second network device sends fourth request information, which requests a correspondence between multiple types of terminal information. The second network device receives the correspondence, which includes a correspondence between the type information of a first terminal and the type information of a second terminal. The second network device can also generate a model for a first analysis service based on this correspondence. This model can serve at least the first terminal and / or the second terminal. Alternatively, the second network device generates a model for a first analysis service, which serves at least the first terminal and the second terminal.

[0066] Based on this implementation, the second network device can request the type information of the corresponding terminal and obtain a model that serves one or more terminals based on the type information of the terminal. In other words, a model that can serve multiple terminals can be obtained through one model training, without the need to train a model separately for each terminal. This can reduce the number of model training sessions and reduce the amount of resources required for customized model training.

[0067] In one possible implementation, the second network device may generate a model for the first analysis service after receiving a first request message carrying type information of the first terminal, and / or after receiving a second request message carrying type information of the second terminal.

[0068] In one possible implementation, the fourth request information may include the analysis identifier of the first analysis service and / or the type information of the first terminal.

[0069] For example, the second network device can flexibly request terminal type information based on analysis service needs and / or terminal needs, thus flexibly determining the range of terminals for model services.

[0070] In one possible implementation, the hardware information of the first terminal is partially or entirely the same as the hardware information of the second terminal; and / or, the software information of the first terminal is partially or entirely the same as the software information of the second terminal.

[0071] The beneficial effects of the methods shown in the fourth aspect above and their various possible implementations can be found in the description of the first aspect and the corresponding implementations, and will not be repeated here.

[0072] Fifthly, a communication method is provided. This method can be implemented by a sixth network device or a component within a sixth network device, which may also be referred to as a communication apparatus or network element. The sixth network device can be a storage network element, for example, it can be a UDM, UDR, ADRF, or EIR. The sixth network device can be the same as or different from the third network device. The sixth network device can also be an AMF or RAN, etc. The sixth network device can be the same as or different from the third network device.

[0073] Taking a sixth network device as the executing entity as an example, the sixth network device can receive a fourth request message from the second network device. This fourth request message is used to request the correspondence between multiple types of terminal information. The sixth network device can also send the correspondence to the second network device, whereby the correspondence includes the correspondence between the type information of the first terminal and the type information of the second terminal.

[0074] The sixth network device can store the correspondence and, upon receiving the fourth request information, send the correspondence to the second network device. The hardware and / or software information of the multiple terminals corresponding to the type information is partially or completely identical. For example, the multiple terminals corresponding to the type information include a first terminal and a second terminal.

[0075] In one possible implementation, the correspondence may specifically include the correspondence between the analysis identifier of the first analysis service, the type information of the first terminal, and the type information of the second terminal.

[0076] The beneficial effects of the methods shown in the fifth aspect above and their various possible implementations can be found in the description of the fourth aspect and the corresponding implementations, and will not be repeated here.

[0077] Sixthly, a communication method is provided. This method can be implemented by a fourth network device or a component within a fourth network device, which may also be referred to as a communication apparatus or network element. The fourth network element is, for example, an AMF (Automatic Communication Component). The components in this application may include, for example, at least one of a chip, a chip system, a processor, a transceiver, a processing unit, or a transceiver unit. Taking the fourth network element as the executing entity as an example, the method can be implemented through the following steps:

[0078] The fourth network device sends a third request message to the fifth network device. This third request message is used to request an analysis identifier corresponding to the analysis service of the first terminal; alternatively, it can be used to request an analysis identifier for an analysis service serving the first terminal; or it can be used to request an analysis identifier for an analysis service to which a model customized by the first terminal is permitted or authorized. The third request message includes the identifier and / or type information of the first terminal. The fourth network device may also receive the analysis identifier from the fifth network device, which is determined based on the identifier and / or type information of the first terminal. The fourth network device may also send the analysis identifier of the first analysis service and the type information of the first terminal to the second network device.

[0079] The analysis identifier can represent a model in which the first terminal is allowed or authorized to customize the first analysis service.

[0080] Based on this implementation, the fourth network device can obtain the analysis identifier of the first analysis service from the fifth network device according to the type information of the first terminal. Specifically, the fourth network device can determine, based on the analysis identifier obtained from the fifth network device, whether the first terminal is allowed or authorized to obtain the first analysis service. In other words, the fourth network device can determine, based on the analysis identifier, whether the first terminal is allowed or authorized to customize the model of the first analysis service, or whether the customized model provided by the first analysis service allows or authorizes the first terminal to obtain the service. Subsequently, it can send the analysis identifier of the first analysis service and the type information of the first terminal to the second network device to notify the second network device that the first terminal is allowed or authorized to obtain the first analysis service. Therefore, it avoids requesting models for terminals that are not allowed or authorized to customize models.

[0081] The fifth network device may be the same as or different from the third network device.

[0082] Seventhly, a communication method is provided. This method can be implemented by a second network device or a component within the second network device, which may also be referred to as a communication apparatus or network element. Taking the second network element as the executing entity as an example, the method can be implemented through the following steps: The second network device receives type information of the first terminal from a fourth network device. Optionally, the second network device can also obtain an analysis identifier of the first analysis service based on the type information of the first terminal, or the second network device can also receive the analysis identifier from the fourth network device. The analysis identifier of the first analysis service may represent a model in which the first terminal is allowed or authorized to customize the first analysis service.

[0083] Eighthly, a communication device is provided. The device can implement the method described in any possible implementation of any of the first to seventh aspects. The device possesses the functions of the first, second, or third network device described above. The device is, for example, a terminal, a functional module within a terminal, a network device, or a functional module within a network device.

[0084] In one optional implementation, the device may include modules corresponding one-to-one with the methods / operations / steps / actions performed in any possible implementation of any of the first to seventh aspects. These modules may be hardware circuits, software, or a combination of hardware circuits and software. In another optional implementation, the device includes a processing unit (sometimes also called a processing module) and a communication unit (sometimes also called a transceiver module, communication module, etc.). The transceiver unit is capable of both sending and receiving functions. When the transceiver unit performs the sending function, it may be called a sending unit (sometimes also called a sending module); when the transceiver unit performs the receiving function, it may be called a receiving unit (sometimes also called a receiving module). The sending unit and the receiving unit may be the same functional module, which is called the transceiver unit and can perform both sending and receiving functions; or, the sending unit and the receiving unit may be different functional modules, with the transceiver unit being a collective term for these functional modules.

[0085] For example, when the apparatus is used to perform the method described in any one of the first to seventh aspects, the apparatus may include a communication unit and a processing unit.

[0086] In a ninth aspect, embodiments of this application also provide a communication device, including a processor for executing a computer program (or computer-executable instructions) stored in a memory, such that when the computer program (or computer-executable instructions) is executed, the device performs the method as described in any possible implementation of any of the first to seventh aspects.

[0087] In one possible implementation, the processor and memory are integrated together;

[0088] In another possible implementation, the memory is located outside the communication device.

[0089] The communication device also includes a communication interface for communicating with other devices, such as sending or receiving data and / or signals. Exemplarily, the communication interface may be a transceiver, circuit, bus, module, or other type of communication interface.

[0090] A tenth aspect provides a computer-readable storage medium for storing a computer program or instructions that, when executed, enable the implementation of the method described in any possible implementation of any of the first to seventh aspects, and the method shown in any possible implementation of the method described in any of the seventh aspects.

[0091] Eleventhly, a computer program product containing instructions is provided, which, when run on a computer, enables the method described in any possible implementation of any of the first to seventh aspects to be implemented.

[0092] In a twelfth aspect, embodiments of this application also provide a communication device for performing the method described in any possible implementation of any of the first to seventh aspects.

[0093] In a thirteenth aspect, a chip system is provided, comprising logic circuitry (or, as understood, a processor, which may include logic circuitry, etc.), and further comprising input / output interfaces. The input / output interfaces can be used to input messages or to output messages. The input / output interfaces can be the same interface, i.e., the same interface can implement both sending and receiving functions; or, the input / output interface includes an input interface and an output interface, the input interface being used to implement the receiving function, i.e., to receive messages; and the output interface being used to implement the sending function, i.e., to send messages. The logic circuitry can be used to perform operations other than the sending and receiving functions in any possible implementation of any of the first to seventh aspects described above; the logic circuitry can also be used to transmit messages to the input / output interfaces or to receive messages from other communication devices from the input / output interfaces. The chip system can be used to implement the methods described in any possible implementation of any of the first to seventh aspects described above. The chip system can be composed of chips or can include chips and other discrete devices.

[0094] Optionally, the chip system may also include a memory, which can be used to store instructions, and the logic circuits can call the instructions stored in the memory to implement the corresponding functions.

[0095] In a fourteenth aspect, a communication method is provided, which may include the method implemented by a first network device as shown in the first aspect and any possible implementation thereof, and the method implemented by a second network device as shown in the second aspect and any possible implementation thereof. Optionally, the method may further include the method implemented by a third network device as shown in the third aspect and any possible implementation thereof.

[0096] Alternatively, the method may include the method implemented by the second network device as shown in the fourth aspect and any possible implementation thereof, and the method implemented by the sixth network device as shown in the fifth aspect and any possible implementation thereof.

[0097] Alternatively, the method may include the method implemented by the fourth network device as shown in the sixth aspect and any possible implementation thereof, and the method implemented by the second network device as shown in the seventh aspect and any possible implementation thereof.

[0098] In a fifteenth aspect, a communication system is provided, which may include a first network device and a second network device, wherein the first network device can be used to implement the first aspect and any possible implementation thereof, and the second network device can be used to implement the second aspect and any possible implementation thereof. The communication system may also include a third network device for implementing the third aspect and any possible implementation thereof.

[0099] Alternatively, the system may include a second network device and a sixth network device, wherein the second network device may be used to implement the fourth aspect and any possible implementation thereof, and the sixth network device may be used to implement the fifth aspect and any possible implementation thereof.

[0100] Alternatively, the system may include a second network device and a fourth network device, wherein the second network device may be used to implement the seventh aspect and any possible implementation thereof, and the fourth network device may be used to implement the sixth aspect and any possible implementation thereof.

[0101] The technical effects brought about by aspects nine through sixteen above can be found in the descriptions of the beneficial effects of the corresponding solutions in aspects one through seven above, and will not be repeated here. Attached Figure Description

[0102] Figure 1 This application provides a schematic diagram of the architecture of a wireless communication system.

[0103] Figure 2 This is a schematic diagram of another wireless communication system architecture provided in an embodiment of this application;

[0104] Figure 3 A schematic diagram illustrating the customization process of an ML model provided in an embodiment of this application;

[0105] Figure 4 A flowchart illustrating a communication method provided in an embodiment of this application;

[0106] Figure 5 This is a schematic diagram of the structure of a TAC provided in an embodiment of this application;

[0107] Figure 6 A flowchart illustrating a communication method provided in an embodiment of this application;

[0108] Figure 7 A flowchart illustrating another communication method provided in an embodiment of this application;

[0109] Figure 8 A flowchart illustrating another communication method provided in an embodiment of this application;

[0110] Figure 9 A flowchart illustrating another communication method provided in an embodiment of this application;

[0111] Figure 10 A flowchart illustrating another communication method provided in an embodiment of this application;

[0112] Figure 11 This is a schematic diagram of the structure of a communication device provided in an embodiment of this application;

[0113] Figure 12 This is a schematic diagram of another communication device provided in an embodiment of this application. Detailed Implementation

[0114] This application can be applied to mobile communication systems or mobile network systems. The network architecture and service scenarios described in the embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. As those skilled in the art will know, with the evolution of network architecture and the emergence of new service scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0115] For example, this application can be applied to 4G, 5G mobile communication systems, or future-oriented evolution systems, without specific limitations.

[0116] As an example, Figure 1 This is a schematic diagram of a 5G network architecture based on a service-oriented architecture. Figure 1The 5G network architecture shown may include terminal devices, access network devices, and core network (CN) devices. Terminal devices access the data network (DN) through access network devices and core network devices. The core network devices include various network functions (NFs) or network elements, such as some or all of the following network elements: UDM network element, UDR network element, NEF network element (not shown in the figure), application function (AF) network element, policy control function (PCF) network element, AMF network element, session management function (SMF) network element, user plane function (UPF) network element, NWDAF network element, network repository function (NRF) network element (not shown in the figure), and location management function (LMF) network element (not shown in the figure).

[0117] Access network equipment can be radio access network (RAN) equipment. Examples include: base stations, evolved NodeBs (eNodeBs), transmission reception points (TRPs), next-generation NodeBs (gNBs) in 5G mobile communication systems, base stations in future mobile communication systems, or access nodes in wireless fidelity (WiFi) systems. It can also be modules or units that perform some of the functions of a base station; for example, it can be a central unit (CU) or a distributed unit (DU). RAN equipment can be macro base stations, micro base stations, indoor stations, relay nodes, or donor nodes. The embodiments of this application do not limit the specific technologies or equipment forms used in the RAN equipment.

[0118] Terminal devices can be user equipment (UE), mobile stations, mobile terminals, etc. They can be widely used in various scenarios, such as device-to-device (D2D), vehicle-to-everything (V2X) communication, machine-type communication (MTC), the Internet of Things (IoT), virtual reality, augmented reality, industrial control, autonomous driving, telemedicine, smart grids, smart furniture, smart offices, smart wearables, smart transportation, and smart cities. Terminal devices can be mobile phones, tablets, computers with wireless transceiver capabilities, wearable devices, vehicles, urban air mobility vehicles (such as drones and helicopters), ships, robots, robotic arms, smart home devices, etc. In the following text, unless otherwise specified, a terminal will be used as an example of a terminal device.

[0119] Access network devices and terminals can be fixed in location or mobile. They can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; they can also be deployed on water; and they can be deployed in the air on aircraft, balloons, and artificial satellites. The embodiments of this application do not limit the application scenarios of the access network devices and terminals.

[0120] The core network equipment involved in this application is described below:

[0121] The AMF (Agency Flow Management) network element includes functions such as mobility management and access authentication / authorization. It is also responsible for transmitting user policies between the terminal and the PCF (Programmable Flow Function). The SMF (Sustainable Flow Management) network element includes functions such as session management, execution of PCF-issued control policies, UPF (User Flow Filter) selection, and terminal Internet Protocol (IP) address allocation. The UPF (User Flow Filter) network element, as the interface with the data network, includes functions such as user plane data forwarding, session / flow-level billing statistics, and bandwidth limiting. The UDM (User Flow Management) network element includes functions such as managing subscription data and user access authorization. The UDR (User Flow Filter) network element includes functions such as storing and retrieving subscription data, policy data, and application data. The NEF (Network Flow Filter) network element is used to support the opening of capabilities and events. The AF (Application Flow Filter) network element transmits application-side requirements to the network side, such as quality of service (QoS) requirements or user state event subscriptions. AF can be a third-party functional entity or an application server deployed by the operator. The PCF (Programmable Flow Function) network element includes policy control functions such as billing at the session and service flow levels, QoS bandwidth assurance and mobility management, and terminal policy decisions. NRF network elements can be used to provide network element discovery functionality, providing network element information corresponding to the network element type based on requests from other network elements. NRF network elements also provide network element management services, such as network element registration, updates, deregistration, and network element status subscription and push.

[0122] NWDAF (Network Data Center) elements are primarily used to collect data (including one or more of the following: terminal data, access network equipment data, core network data, and third-party application equipment data). This data can be data from the terminal itself, the access network equipment, the core network element, or the third-party application equipment, or data from the terminal on that specific access network equipment, core network element, or third-party application equipment. The collected data is then analyzed, and the results are output for use by the network, network management equipment, and applications in decision-making. NWDAFs can utilize machine learning models for data analysis. 3GPP separates the training and inference functions of NWDAFs. An NWDAF can support only model training, only data inference, or both. An NWDAF supporting model training can also be called a training NWDAF, or an NWDAF supporting model training logical function (MTLF) (abbreviated as MTLF). A training NWDAF can train a model based on the acquired data to obtain a trained model. An NWDAF that supports data inference can also be called an inference NWDAF, or an NWDAF that supports analytics logical function (AnLF) (abbreviated as AnLF). An inference NWDAF can input input data into a trained model to obtain analysis results or inference data. In this embodiment, a training NWDAF refers to an NWDAF that at least supports model training. As a possible implementation, a training NWDAF can also support data inference. An inference NWDAF refers to an NWDAF that at least supports data inference. As a possible implementation, an inference NWDAF can also support model training. If an NWDAF supports both model training and data inference, it can be called a training NWDAF, an inference NWDAF, a training-inference NWDAF, or simply an NWDAF. In this embodiment, an NWDAF can be a separate network element or can be co-located with other network elements, such as being placed in a PCF network element or an AMF network element.

[0123] An ADRF network element, or simply ADRF, is used to store model-related data. This data can be generated by the AnLF. The ADRF can provide model-related data to the MTLF upon request. The ADRF... Figure 1 or Figure 2 Not shown in the image.

[0124] The LMF (Local Position Controller) network element is used to manage the location information of terminals. It can calculate or verify the terminal's location and / or estimate the terminal's velocity, and provide the accuracy of the estimate. The LMF network element can receive location requests from the AMF (Agency Position Controller) network element via the Nlmf interface. The granularity of the terminal's location calculated by the LMF network element can be one or more of longitude, latitude, altitude, cell location, and Global Positioning System (GPS).

[0125] A Domain Provider (DN) is a network located outside the carrier's network. A carrier's network can connect to multiple DNs, and various services can be deployed on a DN, providing data and / or voice services to terminals. For example, a DN might be the private network of a smart factory. Sensors installed in the workshop can act as terminals, and a control server for these sensors is deployed within the DN. The control server provides services to the sensors. Sensors can communicate with the control server, receive instructions from it, and transmit the collected sensor data back to the control server accordingly. Another example is a DN serving as an internal office network for a company. Employees' mobile phones or computers can act as terminals, accessing information and data resources on the company's internal office network.

[0126] It is understood that the above network elements are examples of one implementation method. This application does not exclude the possibility that in future wireless communication systems, there may be network elements or devices with the above network element functions that have other names or other forms.

[0127] Figure 1 Nudr, Npcf, Namf, Nudm, Nsmf, Naf, and Nnwdaf are the service interfaces provided by UDR, PCF, AMF, UDM, SMF, AF, and NWDAF, respectively, used to call the corresponding service operations. N1, N2, N3, N4, and N6 are interface sequence numbers, and the meanings of these interface sequence numbers are as follows:

[0128] 1) N1: The interface between the AMF network element and the terminal, which can be used to transmit non-access stratum (NAS) signaling (such as QoS rules from the AMF network element) to the terminal.

[0129] 2) N2: The interface between the AMF network element and the access network equipment, which can be used to transmit radio bearer control information from the core network side to the access network equipment.

[0130] 3) N3: The interface between the access network equipment and the UPF network element, mainly used to transmit uplink and downlink user plane data between the access network equipment and the UPF network element.

[0131] 4) N4: The interface between SMF network elements and UPF network elements. It can be used to transmit information between the control plane and the user plane, including the distribution of forwarding rules, QoS rules, traffic statistics rules, etc. from the control plane to the user plane, as well as the reporting of information from the user plane.

[0132] 5) N6: The interface between the UPF network element and the DN, used to transmit uplink and downlink user data streams between the UP network element F and the DN.

[0133] In addition, Figure 1 In the architecture shown, Nadrf can serve as a service interface for ADRF.

[0134] Figure 2 This is a schematic diagram of a 5G network architecture based on a point-to-point interface. For a description of the functions of the network elements, please refer to [reference needed]. Figure 1 The functions of the corresponding network elements will not be described in detail here. Figure 2 and Figure 1 The main difference is: Figure 1 The interfaces between the various control plane network elements are service-oriented interfaces. Figure 2 The interfaces between the various control plane network elements are point-to-point interfaces.

[0135] exist Figure 2 In the architecture shown, the interface names and functions between the various network elements in the core network are as follows:

[0136] 1) N5: The interface between AF network element and PCF network element, which can be used for application service request distribution and network event reporting.

[0137] 2) N7: The interface between PCF network elements and SMF network elements, which can be used to issue protocol data unit (PDU) session granularity and service data flow granularity control strategies.

[0138] 3) N8: The interface between the AMF network element and the UDM network element. It can be used by the AMF network element to obtain access and mobility management related subscription data and authentication data from the UDM network element, as well as by the AMF network element to register terminal current mobility management related information with the UDM network element.

[0139] 4) N9: User plane interface between UPF network elements, used to transmit uplink and downlink user data streams between UPF network elements.

[0140] 5) N10: The interface between the SMF network element and the UDM network element. It can be used by the SMF network element to obtain session management-related subscription data from the UDM network element, and by the SMF network element to register terminal current session-related information with the UDM network element.

[0141] 6) N11: The interface between SMF network elements and AMF network elements. It can be used to transmit PDU session tunnel information between access network devices and UPF network elements, transmit control messages sent to terminals, and transmit radio resource control information sent to access network devices.

[0142] 7) N15: The interface between PCF network elements and AMF network elements, which can be used to issue terminal policies and access control related policies.

[0143] 8) N23: The interface between the PCF network element and the NWDAF network element. The NWDAF network element can collect data from the PCF network element through this interface. It should be noted that the NWDAF network element can also have interfaces with other devices (such as AMF network elements, UPF network elements, access network devices, terminals, etc.), which are not fully shown in the figure.

[0144] 9) N35: The interface between UDM network elements and UDR network elements, which can be used by UDM network elements to obtain user subscription data information from UDR network elements.

[0145] 10) N36: The interface between PCF network elements and UDR network elements, which can be used by PCF network elements to obtain policy-related contract data and application data related information from UDR network elements.

[0146] It is understood that the aforementioned network element or function can be a network component in a hardware device, a software function running on dedicated hardware, or a virtualization function instantiated on a platform (e.g., a cloud platform). As one possible implementation method, the aforementioned network element or function can be implemented by a single device, multiple devices working together, or a functional module within a single device; this application does not specifically limit this.

[0147] As one implementation method, the data analysis network element in the embodiments of this application can be the aforementioned NWDAF network element, or it can be a network element that has the functions of the aforementioned NWDAF network element in future communications. For ease of explanation, the following description uses an NWDAF network element as the data analysis network element, and the NWDAF network element will be abbreviated as NWDAF.

[0148] In this application, AnLF can be used to perform inference, derive analytical information, and publish analytical services. Here, analysis can refer to deriving statistical information and / or predictions based on requests from analytical consumers. MTLF can be used to train machine learning models and publish new training services, for example, providing a trained ML model to AnLF. In this application, for the process of acquiring model-related data, AnLF can act as a data producer network element, denoted as NFp; MTLF can act as a data consumer network element, denoted as NFc. Unless otherwise specified, the model in this application can be an AI ML model or an ML model, or it can be other types of AI models.

[0149] Currently, AnLF can request models from MTLF via the MLModelProvision_Subscribe service or messages. These models can be trained by MTLF using model-related data (e.g., samples). Furthermore, AnLF can specify a dataset using a dataset tag, which can contain model-related data. Therefore, a dataset tag can be used to label a dataset, i.e., to label model-related data. A dataset tag can also be understood as an index of the dataset or the data itself. MTLF can act as a consumer network element for model-related data, acquiring data stored in a data storage provider by a data producer network element. The data producer network element can be AnLF, and the data storage provider network element can be an analytics data repository functional (ADRF) or other network elements used for storing data. In this application, related data can be understood as data used for longitudinal federated learning tasks, such as input data, training data, inference data, or sample data.

[0150] Specifically, model-related data can include any data from ADRF, including model input data, model output data (e.g., prediction results or predicted data), model true values, and prediction bias data. Model input data is used to train the model. Model output data includes, for example, prediction results obtained from input data, or model prediction results corresponding to input data. Model true values ​​represent the true values ​​of the output data corresponding to the model input data, also known as the true results of the input data. Prediction bias data identifies the deviation between predicted data and true values. Model output data and model true values ​​can be used for model accuracy monitoring and retraining and / or re-providing relevant data. In other words, the model can be obtained from model training data. When obtaining the model through initial training, it can be trained using model input data and model true values; that is, the training data includes model input data and / or model true values. Furthermore, the model can also be obtained by retraining based on a historical model using model-related data; the historical model is an existing model. Prediction data can be the model's output results.

[0151] In this application, a first network device can provide a model to a second network device to support the second network device in providing services to the terminal based on the model, or to support the second network device in providing the model to the terminal or other network devices. This model can be a model customized by the first network device for the terminal. It is understood that providing services to the terminal through a customized model can improve the terminal's service experience.

[0152] like Figure 3The diagram illustrates a model customization process to support customized model services. An NWDAF service consumer, acting as a model user, can subscribe to, modify, or unsubscribe from one or more training ML models associated with one or more analysis IDs by invoking the Nnwdaf_MLModelProvision_Subscribe / Nnwdaf_MLModelProvision_Unsubscribe service operations. In this application, subscription can refer to model customization. The NWDAF consumer can provide the analysis ID to the NWDAF containing the MTLF. Upon receiving an ML model subscription, the NWDAF containing the MTLF can determine whether an existing trained ML model is available for subscription, or whether further training of the existing trained ML model is required. If the NWDAF containing the MTLF determines that further training is needed, it can initiate data collection from NF (e.g., AMF, DCCF, or ADRF), UE applications (via AF), or OAM to generate an ML model based on the data. An NWDAF containing MTLF can provide a retrained ML model to an NWDAF service consumer by calling the Nnwdaf_MLModelProvision_Notify service operation. For example, after obtaining the model, the NWDAF service consumer can provide the model to a terminal, or it can provide the model to a terminal or other network devices that provide services to terminals.

[0153] An analysis identifier can be used to indicate an analysis service (or analysis business) and / or a model. The analysis service is associated with a model; that is, the model can be used to perform the analysis service. Any analysis identifier or analysis service can correspond to one or more models.

[0154] Customizing a model requires training it based on data, which demands significant computational resources. When the ML model serves multiple devices, customizing a separate model for each device would consume a large amount of training resources. Therefore, reducing the resource requirements for training ML models serving multiple devices is a pressing technical problem.

[0155] This application provides a communication method for reducing unnecessary custom model training and lowering the resource requirements for customized model training. This method can be executed by a first network device and a second network device. The actions of the first network device can also be implemented by modules or chips within the first network device, and the actions of the second network device can also be implemented by modules or chips within the second network device. The first network device can be a core network element or a core network device. For example, the first network element can be an NWDAF. The second network device can be an access network device or a core network device, without specific limitation.

[0156] Furthermore, this communication method can also be executed by a third network device and / or a fourth network device. The operation of the third network device can also be replaced by a module or chip within the third network device, and the operation of the fourth network device can also be replaced by a module or chip within the fourth network device. The third network device is, for example, a storage network element or storage device. In this application, the storage network element can be a UDR, UDM, or EIR network element or function, or other network elements with data storage capabilities. In the following description, a UDR is used as an example of a storage network element. The fourth network device is, for example, an AMF serving the terminal.

[0157] like Figure 4 As shown, after receiving a request from a second network device for a model to provide a first analysis service for a first terminal, the first network device can determine the type information of the second terminal based on the type information of the first terminal contained in the request information. Furthermore, the model provided by the first network device to the second network device can be determined based on the type information of the second terminal; that is, the first network device can provide the second network device with information about a model that provides services to the first terminal based on the type information of the second terminal. This model can serve at least the first terminal and / or the second terminal, meaning that the same model can serve both the first and / or the second terminal, avoiding redundant model training for both terminals and reducing the resources required for model training. In addition, this model can also serve other terminals besides the first and second terminals that possess the type information of the first terminal and / or the type information of the second terminal.

[0158] It can be assumed that terminals with the same hardware and / or software information can use the same model. Therefore, to avoid retraining or determining the model and to avoid wasting training resources, the same model can be used to serve multiple terminals with the same hardware and / or software information. It can be assumed that in this application, the type information of the terminal is used to determine whether the hardware and / or software information between terminals is the same. Depending on the needs, the "type information" of the terminal can also be replaced with the "hardware and / or software information" of the terminal.

[0159] As an example, the second network device could be Figure 3 In the context of NWDAF service consumers, the first network device can be... Figure 3 The NWDAF in the context. That is, it can be based on the first network device. Figure 3 The process provides a customized model to the second network device.

[0160] Unless otherwise specified, "model" in this application may include a model for determining the communication configuration between the terminal and the access network device, or a model for assisting in determining the terminal's location-related configuration or for determining the terminal's location, and may also include other models; this application does not specifically limit this. Models for assisting in determining the terminal's location-related configuration may include, for example, models for extracting features from channel response information, and models for determining the terminal's location may be models for outputting the terminal's location information. Unless otherwise specified, "communication configuration" in this application may refer to the communication configuration between the terminal and the access network device, which may include configurations configured on the terminal side and / or, include communication configurations configured on the access network device side. The terminal may use the communication configuration configured on the terminal side to communicate, and the access network device may use the communication configuration configured on the access network device side to communicate with the terminal or provide services to the terminal.

[0161] In this application, the model used to determine the communication configuration between the terminal and the access network device can also be called a communication configuration model. The second network device can obtain the communication configuration between the terminal and the access network device through this model to optimize communication performance. That is, the second network device can use the communication configuration of the terminal adopted by the model. The channel information between the terminal and the access network device can be used as input data for the model, and correspondingly, the output data of the model can be the communication configuration such as beam parameters between the terminal and the access network device. The second network device (or other network device) can also configure the terminal and / or access network device according to the communication configuration output by the model, so that the terminal and / or access network device communicate according to the configuration. The second network device can refer to the access network device or a core network device such as an AMF (Advanced Network Components), without specific limitation.

[0162] Furthermore, positioning technology can specifically include AI positioning, which utilizes machine learning algorithms to enhance positioning accuracy. In AI positioning, the model can be deployed on the LMF side or the base station side, without specific limitations. This model can include a model used to assist in determining the terminal's positioning-related configuration or a model used to determine the terminal's location. The model used to assist in determining the terminal's positioning-related configuration can be used to determine the terminal's positioning-related configuration. Positioning-related configuration, also known as positioning configuration, includes, for example, features of the terminal's channel information (or channel response). The input to this model can be the terminal's channel information, and the output can be positioning configuration features such as channel information. Whether uplink or downlink positioning, channel information can be obtained through reference signals such as sounding reference signals (SRS) or positioning reference signals (PRS) sent by access network equipment.

[0163] Furthermore, the model used to determine the terminal's location can be used to output the model's location information. The input to this model can be channel information or features of the channel information, and the output of the model can be the terminal's location information.

[0164] In AI positioning, the second network device can... Figure 3 The process yields a model customized for the terminal. For example, a network device can serve as... Figure 3 The NWDAF service consumer in the system obtains the model from the NWDAF. This model may include a model for assisting in determining the terminal's location-related configuration or a model for determining the terminal's location.

[0165] In this application, the terminal's channel information can describe the terminal's channel, for example, it can be used to describe the channel between the terminal and the access network equipment. For example, the channel information includes measurement results obtained from measuring the reference signal between the terminal and the access network equipment. That is, the measurement here can include uplink channel measurements and / or downlink channel measurements. It is understood that in the first application service, the terminal's channel information can be used as input to the model.

[0166] For example, in downlink channel measurement, the reference signal can be a channel state information reference signal (CSI-RS). Similarly, in uplink channel measurement, the reference signal can be an SRS. The measurement results of the reference signal can include at least one of the following: reference signal receiving power (RSRP), reference signal receiving quality (RSRQ), reference signal strength indicator (RSSI), or signal to interference plus noise ratio (SINR).

[0167] The channel measurement results in the examples above are illustrative of channel information and should not be construed as limiting the scope of channel information. This application does not limit other parameters that can be used to characterize the channel quality of a terminal as channel information.

[0168] It can be assumed that if the first analysis service or model is used to determine the communication configuration of a terminal, the input of the model may include the terminal's channel information, and the output of the model may include the terminal's communication configuration. Specifically, the communication configuration may include the terminal's transmit and / or receive configurations, such as beam-related configurations, resource-related configurations, or power-related configurations, etc., without specific limitations. Similarly, if the first analysis service or model is used for the terminal's positioning, the input of the model may include channel information, and the output of the model may include positioning-related configurations or location information.

[0169] It can be assumed that if the first analysis service or model is used to determine the communication configuration of the terminal and / or the communication configuration of the access network equipment (when serving the terminal), then the input of the model may include channel information of the terminal and / or the access network equipment, and the output of the model may include the communication configuration of the terminal and / or the access network equipment. The communication configuration may specifically include the transmission configuration and / or reception configuration of the terminal and / or the access network equipment, such as beam-related configuration, resource-related configuration, or power-related configuration, etc., without specific limitations. Similarly, if the first analysis service or model is used for the positioning of the terminal, the input of the model may include channel information, and the output of the model may include positioning-related configuration or location information.

[0170] The following is combined Figure 4 The flowchart shown describes the communication method provided in this application. Figure 4 As shown, the method may include the following steps:

[0171] S101: The second network device sends a first request message to the first network device. The first request message is used to request the model corresponding to the first analysis service for the first terminal.

[0172] The phrase "the first request information is used to request a model corresponding to the first analysis service for the first terminal" can also be replaced with or interpreted as: the first request information is used to request a model corresponding to the first analysis service serving the first terminal, or the first request information is used to request a model corresponding to the first analysis service serving (or applicable to) the first terminal. The first terminal may include one or more terminals. If the first terminal includes multiple terminals, these multiple terminals may have the same type information, which is the type information of the first terminal.

[0173] Optionally, the first request information may include the type information of the first terminal. In this application, the type information of the first terminal may also be replaced with first type information. Unless otherwise specified, the type information of the first terminal or the first type information may be the type information of one or more terminals, wherein the one or more terminals include the first terminal, and optionally may also include one or more terminals other than the first terminal.

[0174] For ease of explanation, the one or more terminals with the first type of information may be referred to as the first terminal set. The terminals in the first terminal set include the first terminal, and optionally may also include one or more terminals other than the first terminal.

[0175] In this application, "information" can also be replaced with "message". "Information" can be carried in fields or information elements within a message. Alternatively, the function of information can be reflected through the message itself; for example, the message's name and / or format can indicate the function of the information.

[0176] In this application, the type information of the terminal can be used to indicate the type of the terminal, or it can be considered to indicate or represent the hardware and / or software information of the terminal. For example, the type information of the terminal can be one or more of the following: the terminal's type allocation code (TAC), vendor ID, or brand identifier.

[0177] For example, the hardware information of a terminal may include at least one of the following: baseband hardware module information (such as baseband chip model), radio frequency (RF) hardware module information (such as RF module), or hardware parameters affecting power consumption. The software information of a terminal may include at least one of the following: baseband software configuration, RF software configuration, or software parameters affecting power consumption. Alternatively, the type of terminal can be used to indicate at least one of the following: baseband hardware module information, baseband software configuration, RF hardware module information, RF software configuration, hardware parameters affecting power consumption, or software parameters affecting power consumption. Therefore, a first terminal can be considered a collective term for terminals having one or more of the same software and / or hardware features. For example, a first terminal may include multiple terminals with partially or completely identical hardware information and / or multiple terminals with partially or completely identical software information.

[0178] It can be considered that the hardware and / or software information of the aforementioned terminals affects their communication performance. Specifically, communication performance can refer to the terminal's signal processing capability, which indicates the minimum received power of the signal the terminal can receive or recover. That is, terminals with different hardware and / or different software information have different minimum received power values ​​for the signals they can receive or recover. For example, terminal 1's signal processing capability can support it in receiving a signal with power #1 and successfully recovering the information carried by that signal, while due to differences in hardware and / or software information, terminal 2's signal processing capability cannot support it in recovering the information carried by a signal with power #1.

[0179] In this application, the hardware information of the first terminal is partially or entirely the same as the hardware information of the second terminal, and / or the software information of the first terminal is partially or entirely the same as the software information of the second terminal. The hardware information of the terminal here may include all the hardware information of the terminal, or it may include at least one hardware parameter such as baseband hardware module information, radio frequency hardware module information, or hardware parameters affecting power consumption. That is, all the hardware information of the first terminal and the second terminal is the same, or at least one of the baseband, radio frequency, or power consumption-affecting hardware parameters is the same for the first terminal and the second terminal.

[0180] In addition, the terminal's software information may include all of the terminal's software information, or at least one software parameter such as baseband software configuration, radio frequency software configuration, or software parameters affecting power consumption. That is, the first terminal and the second terminal have the same all of their software information, or at least one of their baseband software configuration, radio frequency software configuration, or software parameters affecting power consumption is the same.

[0181] Since the hardware and / or condition information of the first terminal and the second terminal are partially or completely the same, and the channel information corresponding to the first terminal is the same as that corresponding to the second terminal, the communication configuration of the first terminal and the communication configuration of the second terminal are the same. If the first analysis service is used to determine the antenna configuration of the terminal, the same model can be used to determine the communication configuration of the first terminal and the communication configuration of the second terminal respectively.

[0182] Similarly, if the channel information corresponding to the first terminal is the same as that corresponding to the second terminal, then the positioning configuration of the first terminal is the same as that of the second terminal. Therefore, if the first analysis service is to determine the positioning configuration of the terminals, the same model can be used to determine the positioning configuration of the first terminal and the positioning configuration of the second terminal respectively. Alternatively, if the channel information corresponding to the first terminal is the same as that corresponding to the second terminal, then the location information of the first terminal is the same as that of the second terminal. Therefore, if the first analysis service is to determine the location information of the terminals, the same model can be used to determine the location information of the first terminal and the location information of the second terminal respectively.

[0183] Taking the model corresponding to the first analysis service, which can be used to determine the communication configuration between the terminal and the access network device, as an example, the input data of this model includes the terminal's channel information, and the output data includes the communication configuration between the terminal and the access network device. Since the hardware and / or condition information of the first terminal and the second terminal are partially or completely identical, this model can be used to determine the communication configuration of the first terminal and / or to determine the communication configuration of the second terminal. Specifically, if the channel information between the first terminal and the access network device is used as the input data of this model, the corresponding communication configuration of the first terminal can be used as the output data of this model. Similarly, if the channel information between the second terminal and the access network device is used as the input data of this model, the corresponding communication configuration of the second terminal can be used as the output data of this model.

[0184] As an example of type information, TAC can be the first N bits of the International Mobile Equipment Identity (IMEI), where N is a positive integer greater than 1. For example... Figure 5 As shown, the terminal's TAC is the first 8 bits of the IMEI, i.e., N=8. Figure 5This is merely one example of a TAC and should not be construed as limiting the definition of a TAC. A vendor identifier can be used to represent the vendor of the terminal. Furthermore, a TAC can also be a portion of bits from an International Mobile Equipment Identity Software Version (IMEISV) or a Permanent Equipment Identifier (PEI). A brand identifier can be an identifier set by the vendor for the terminal. For example, a brand identifier can be a product identifier for the terminal or an internal identifier of the manufacturer, without specific limitations. Generally, terminals with the same brand identifier have the same type, some or all of the same hardware information, and / or some or all of the same software information.

[0185] Optionally, during or after a terminal's registration with the network, a fourth network device serving the terminal can obtain the terminal's type information and send it to a second network device. Taking the type information as TAC as an example, the fourth network device can be an AMF (Application Function Manager). The AMF can obtain the first terminal's IMEI during the first terminal's registration process and then obtain the first terminal's TAC based on the IMEI. The AMF can also proactively provide the first terminal's TAC to the RAN (Radio Area Network) serving the first terminal. Alternatively, the AMF can send the first terminal's TAC to the RAN after receiving a request for the first terminal's type information from the RAN serving the first terminal. For example, the TAC can be included in the first terminal's mask IMEISV or PEI sent by the AMF to the RAN, allowing the RAN to determine the first terminal's TAC based on the first terminal's mask IMEISV or PEI.

[0186] In S101, the model corresponding to the first analysis service can refer to a model used to perform the first analysis service. As an example, the first analysis service may include a service for determining the communication configuration between the terminal and the access network device; that is, the model corresponding to the first analysis service can be used to determine the communication configuration between the terminal and the access network device, such as beam transmission or transmission resource transmission. For example, the model may specifically be a beam management model or a resource configuration model.

[0187] In S101, the first request information can be a model acquisition request, or it can have other names; there are no specific requirements. The first request information carries the type information of the first terminal, which can be interpreted or replaced as: requesting the above-mentioned model for the first terminal or a terminal with the type information of the first terminal, or requesting the above-mentioned model applicable to the first terminal or a terminal with the type information of the first terminal. The terminal with the type information of the first terminal includes the first terminal, and may also include other terminals besides the first terminal.

[0188] S102: The first network device obtains the type information of the second terminal based on the type information of the first terminal.

[0189] The type information of the second terminal can also be replaced with second type information. S102 can also be replaced with obtaining the second type information based on the type information or the first type information of the first terminal. Here, the second terminal can be considered as one or more terminals.

[0190] A second terminal can be considered a collective term for terminals that have one or more of the same software and / or hardware features. For example, a second terminal may include multiple terminals with partially or completely identical hardware information, and / or multiple terminals with partially or completely identical software information. Unless otherwise specified, the type information or second type information of the second terminal may be the type information of one or more terminals, wherein the one or more terminals include the second terminal, and optionally may also include one or more terminals other than the second terminal.

[0191] For ease of explanation, the term "second terminal set" may be used to refer to one or more terminals with the second type of information. The second terminal set includes the second terminals, and optionally may also include one or more terminals other than the second terminals.

[0192] In this scenario, the type information of the first terminal differs from that of the second terminal. For example, the TAC of the first terminal differs from that of the second terminal. Furthermore, as described in S101, the hardware information of the first terminal is partially or entirely the same as that of the second terminal, and / or the software information of the first terminal is partially or entirely the same as that of the second terminal. Therefore, the same model of the first analysis service can be used to serve both the first and second terminals. As one implementation of the first network device obtaining the type information of the second terminal based on the type information of the first terminal, the first network device can obtain the type information of the second terminal from a third network device based on the type information of the first terminal. For example, the first network device can send the type information of the first terminal to the third network device and receive type information of one or more terminals from the third network device, including the type information of the second terminal. The type information of the one or more terminals (such as the type information of the second terminal) can be determined based on the type information of the first terminal. The third network device can store or configure a correspondence between the type information of the first terminal and the type information of one or more terminals (which can be called a first correspondence) to determine the type information of the one or more terminals corresponding to the type information of the first terminal. Alternatively, the first correspondence can be considered to include the correspondence between the type information of the first terminal and the type information of the second terminal. The same model can serve multiple terminals corresponding to the type information, and these multiple terminals may include the first terminal, as well as one or more terminals including the second terminal mentioned above.

[0193] Optionally, the first network device may obtain the type information of the second terminal from the third network device after receiving the first request information. Alternatively, the first network device may obtain the correspondence between the type information of the first terminal and the type information of the second terminal from the third network device before receiving the first request information, and then determine the second type information corresponding to the type information of the first terminal based on the locally stored correspondence. The method of obtaining the correspondence between the type information of the first terminal and the type information of the second terminal from the third network device can be, for example, the first network device sending the type information of the first terminal to the third network device and receiving the type information of the second terminal from the third network device; or, for example, the first network device sending the type information of the second terminal to the third network device and receiving the type information of the first terminal from the third network device.

[0194] Taking a first terminal with type information TAC1 and a second terminal with type information TAC2 as an example, the UDR can store a list of TACs, which contains the correspondence between TAC1 and TAC2. This list can also contain the correspondence between TACs other than TAC1 and TAC2.

[0195] An exemplary list of TACs is shown in Table 1. In Table 1, the multiple TACs in each row correspond to each other.

[0196] Table 1

[0197]

[0198]

[0199] Table 1 above is an example of a TAC list and should not be construed as a limitation on the TAC list in this application.

[0200] It can be assumed that the TAC list stored in the UDR contains one or more rows as shown in Table 1. For example, when the UDR receives TAC1 from the first terminal of the first network device, it can determine the TAC2 corresponding to TAC1 according to Table 1 and send TAC2 to the first network device. This TAC2 can then be used as the type information of the second terminal. As another example, if the type information of the first terminal is TAC3, the third network device can determine TAC4 and TAC5 corresponding to TAC3 and send TAC4 and TAC5 to the first network device. One or more of TAC4 and TAC5 can be used as the TAC of the second terminal.

[0201] As shown in Table 2, the TAC list may also contain the correspondence between analysis identifiers and multiple TACs, indicating that the analysis service corresponding to the analysis identifier can be applied to the terminals corresponding to the multiple TACs. Specifically, the same model of the analysis service can be used to serve the terminals corresponding to multiple TACs with corresponding relationships in the TAC list. That is, the first correspondence may include the correspondence between the analysis identifier, the type information of the first terminal, and the type information of the second terminal. In other words, there is a correspondence between the analysis identifier (or analysis service), the type information of the first terminal, and the type information of the second terminal. Any analysis identifier can represent an analysis service. Any analysis service and / or analysis identifier corresponds to one or more models. Furthermore, any analysis identifier can correspond to one or more TACs.

[0202] For example, each row shown in Table 2 can serve as an analysis identifier and its correspondence with multiple TACs.

[0203] Table 2

[0204] Analysis Identifier TAC Analysis Identifier #1 TAC1, TAC2 Analysis identifier #2 TAC3, TAC4, TAC5 Analysis identifier #3 TAC1, TAC6 …… ……

[0205] Table 2 above is an example of a list of TACs containing the correspondence between analysis identifiers and TACs, and should not be construed as a limitation on the TAC list in this application.

[0206] When the UDR receives TAC1 from the first terminal of the first network device, it can determine TAC2 and TAC6 corresponding to TAC1 according to Table 2, and send TAC2 and TAC6 to the first network device. TAC2 and / or TAC6 can be used as the type information of the second terminal.

[0207] Optionally, when the first correspondence includes a correspondence between multiple analysis identifiers corresponding to the type information of the first terminal and the type information of one or more terminals, the third network device can determine the multiple analysis identifiers corresponding to the type information of the first terminal and the type information of one or more terminals based on the first correspondence, and send the correspondence between the multiple analysis identifiers corresponding to the type information of the first terminal and the type information of one or more terminals to the first network device, or send the type information of the first terminal and its corresponding multiple analysis identifiers and the type information of one or more terminals. Correspondingly, the first network device can, based on the analysis identifier of the first analysis service and the multiple analysis identifiers and the type information of one or more second terminals sent by the third network device, search for the type information of the terminal corresponding to the analysis identifier of the first analysis service, and obtain the type information of one or more terminals. Part or all of the type information of these one or more terminals can be used as the type information of the second terminals.

[0208] Taking Table 2 as an example, the first network device can send the TAC1 of the first terminal to the UDR. Since TAC1 corresponds to analysis identifier #1 and analysis identifier #2 in the correspondence shown in Table 2, the UDR can send the correspondence between analysis identifier #1 and TAC2, as well as the correspondence between analysis identifier #3 and TAC6, to the first network device. If the first analysis identifier is analysis identifier #1, the first network device can determine that the type information of the second terminal is TAC2 instead of TAC6 based on the correspondence between analysis identifier #1 and TAC2.

[0209] Optionally, the first request information received by the first network device may also include an analysis identifier for the first analysis service. This first request can be used to request the model of the analysis service corresponding to the analysis identifier. Accordingly, the first network device can obtain the type information of the second terminal based on the analysis identifier and the type information of the first terminal. For example, the first network device can send the type information of the first terminal and the analysis identifier to the UDR. The UDR can determine the type information of the second terminal based on the type information of the first terminal and the analysis identifier. As shown in Table 2, if the first request information includes the TAC1 and analysis identifier #1 of the first terminal, the first network device can send the TAC1 and analysis identifier #1 of the first terminal to the UDR. The UDR can determine the TAC2 corresponding to TAC1 and analysis identifier #1 based on TAC1 and analysis identifier #1, and send TAC2 to the first network device. This TAC2 can be used as the type information of the second terminal.

[0210] As another implementation of the first network device determining the type information of the second terminal based on the type information of the first terminal, the first network device can determine the type information of the second terminal corresponding to the type information of the first terminal based on a locally stored first correspondence. The first correspondence can be obtained by the first network device from a third network device, or it can be configured in the third network device. For example, if the first network device obtains the type information of the second terminal from the third network device based on the type information of the first terminal during previous communication, or obtains the type information of the first terminal from the third network device based on the type information of the second terminal, then the first network device can store the first correspondence.

[0211] It is understandable that the above first correspondence can be sent by the AF to the third network device. Taking the type information as TAC as an example, one implementation is that the AF can send a TAC list to the third network device. The TAC list contains the correspondence between TACs, i.e., the first correspondence. For example, the TAC list contains the correspondence between the TACs of the first terminal and the TACs of the second terminal, indicating that the type of the first terminal is the same as the type of the second terminal, the hardware information of the first terminal is partially or completely the same as the hardware information of the second terminal, and / or, the software information of the first terminal is partially or completely the same as the software information of the second terminal. That is, in this implementation, the TAC list can be used to indicate the correspondence between TACs of terminals with the same type, terminals with partially or completely the same hardware information, and / or terminals with partially or completely the same software information.

[0212] In another implementation, the TAC list also corresponds to the analysis identifier of the analysis service. Alternatively, the TAC list may also contain a correspondence between the analysis identifier and multiple TACs, indicating that the analysis service model can serve UEs with these multiple TACs. A correspondence exists between these multiple TACs. For example, the AF can send a service subscription request to a third network device to request permission or authorization for terminals in the first terminal set and terminals in the second terminal set to customize the same analysis service model. This request may carry a first correspondence. The request may also have other names, such as subscription request or registration request, without specific limitation. For example, the first correspondence may include a correspondence between the TACs of the first terminal and the TACs of the second terminal, or it may include a correspondence between the analysis identifier of the first analysis service, the TACs of the first terminal, and the TACs of the second terminal. Accordingly, the third network device can store the first correspondence. The third network device can also send response information to the AF to indicate acceptance of the request.

[0213] Unless otherwise specified in this application, AF can be used for communication between equipment vendors and the network. For example, AF can be used for subscription information of services provided by equipment vendors to the network. Specifically, AF can be an over-the-top (OTT) system or OTT server of the equipment vendor for terminals in the first terminal set and / or terminals in the second terminal set. As described in this application, the equipment vendor can determine that the model of the first analysis service can serve terminals in both the first and second terminal sets based on one or more of the types, hardware information, and software information of the terminals in the first and second terminal sets, and send the first correspondence relationship to the third network device through AF. As an example, AF sends the first correspondence relationship to the third network device when the types of terminals in the first and second terminal sets are the same, the hardware information of terminals in the first and second terminal sets is partially the same as the hardware information of terminals in the second terminal set, and / or the software information of terminals in the first and second terminal sets is partially or completely the same as the software information of terminals in the second terminal set.

[0214] S103: The first network device sends the model information to the second network device. The model information is determined based on the type information of the second terminal.

[0215] In S103, the model information may include information used to indicate the model. For example, the model information may include the model itself, or model parameters such as the model's weights and / or gradients, or the model's identifier.

[0216] In S103, the model can serve a first set of terminals and / or a second set of terminals. For example, the model can serve terminals in the first set, such as the first terminal, and / or the model can serve terminals in the second set, such as the second terminal. Alternatively, the model can serve at least the first and / or the second terminals. It can also be said that the model can serve one or more terminals whose type information is the same as that of the first terminal, and / or serve one or more terminals whose type information is the same as that of the second terminal.

[0217] Optionally, the model may be generated based on training data from terminals in the first set of terminals and / or training data from terminals in the second set of terminals. For example, the model may be generated based at least on training data from the first set of terminals and / or training data from the second set of terminals.

[0218] Taking the generation of the model based on training data from a first terminal and / or a second terminal as an example, when the first network device determines that the model serves the first terminal and / or the second terminal, the first network device can collect training data from both the first and second terminals, and train the model based on the training data from both terminals to obtain the model. Compared to training the model based solely on the training data from either the first or second terminal, this increases the amount of training data, which helps improve model accuracy. Furthermore, to improve the efficiency of model training, the first network device can also train the model based on either the training data from the first or second terminal.

[0219] The training data can be used to generate a model for the first analysis service. For example, the training data may include the model's input data and / or actual output data. For instance, the input data may include communication information, and the output data may include communication configuration, location configuration, or location information.

[0220] In one possible implementation, the first network device can generate the model even if it cannot find the model corresponding to the analysis identifier of the first analysis service (or the model of the first analysis service). For example, the first network device can search for the corresponding model based on the analysis identifier. If it cannot find or does not exist the model corresponding to the analysis identifier, the first network device can trigger the generation of the model.

[0221] Specifically, the first network device can query the correspondence between the analysis identifier and the model (or model information) of the first analysis service stored locally (which can be referred to as the second correspondence) based on the analysis identifier, and / or query the second correspondence in other network devices. If the second correspondence includes the correspondence between the analysis identifier and the model of the first analysis service, the first network device can determine that a model of the first analysis service has been found, i.e., a model corresponding to the analysis identifier exists. Furthermore, if the second correspondence contains a correspondence between the analysis identifier and the model, the first network device can determine that a model of the first analysis service has been found, i.e., a model corresponding to the analysis identifier exists.

[0222] Optionally, the first network device and / or other network devices may store the correspondence between the analysis identifier and the model of the analysis service, which may include a second correspondence. For example, after generating the model of the analysis service, the first network device and / or other network devices may store the correspondence between the analysis identifier and the model.

[0223] In one possible implementation, "the model information is determined based on the type information of the second terminal" can be understood as: the model is generated at least based on the type information of the second terminal, or the model is generated at least based on second type information. For example, the model information may be generated by the first network device after receiving a request (referred to as second request information) for a model requesting a first analysis service for the second terminal, wherein the request may carry the type information of the second terminal. Alternatively, the model may be considered as a model of the first analysis service customized for the second terminal by the first network device after receiving the second request information.

[0224] Optionally, the model can be generated based on training data from terminals in the first terminal set and training data from terminals in the second terminal set, or the model can be generated based on training data from terminals in the second terminal set. The terminals in the first terminal set can be determined based on the type information of the first terminal, and the terminals in the second terminal set can be determined based on the type information of the second terminal. In this implementation, the model can be trained by the first network device, or the first network device can request or instruct other network devices or nodes to train the model; this application does not specifically limit this.

[0225] As an example of this implementation, the first network device can trigger the generation of the model after receiving the second request information. The second request information may carry the type information of the second terminal. This second request information is similar to the description of the first request information, except that the type information of the first terminal in the first request information is replaced with the type information of the second terminal. For example, after receiving the second request information, the first network device can trigger the training of the model to generate a model serving the second terminal. This model can be trained using training data from terminals in the first terminal set and / or training data from terminals in the second terminal set. In S102, after determining the type information of the second terminal based on the type information of the first terminal, the first network device can also determine the model information based on the type information of the second terminal, that is, determine that the model can also serve the first terminal.

[0226] For example, after generating the model, the first network device can store the correspondence between the model's information (such as at least one of the model, model parameters, or model identifier) ​​and the type information of the second terminal. After receiving the first request information in S101, the second network device can obtain the corresponding type information of the second terminal according to the type information of the first terminal in S102, and then determine the model's information based on the model's information and the second terminal's type information.

[0227] For example, after generating a model that serves the second terminal, the first network device can store the correspondence between the information of the model and the type information of the second terminal. After determining the type information of the corresponding second terminal based on the type information of the first terminal in S102, the type information of the first terminal is added to the above correspondence. That is, the first network device can store the correspondence between the information of the model, the type information of the first terminal and the type information of the second terminal, and determine the information of the model corresponding to the type information of the first terminal based on the correspondence.

[0228] For example, after generating a model serving the second terminal, the first network device can store the correspondence between the analysis identifier of the first analysis service, the information of the model, and the type information of the second terminal. After receiving a first request message carrying the analysis identifier of the first analysis service in S101, the first network device can determine whether a corresponding model exists based on the analysis identifier, and determine whether the model can still serve the first terminal based on the type information of the terminal served by the corresponding model. For instance, after receiving the first request message carrying the analysis identifier of the first analysis service in S101, the first network device can query the above correspondence based on the analysis identifier to determine whether there is information about the model corresponding to the analysis identifier and the type information of the second terminal. The first network device can also determine one or more type information corresponding to the type information of the second terminal. For example, the first network device can also send the type information of the second terminal to a third network device and receive one or more type information of terminals corresponding to the type information of the second terminal. It is understood that if the type information of one or more terminals corresponding to the type information of the second terminal includes the type information of the first terminal, it can be determined that the model serves the first terminal and / or the second terminal, that is, the first network device can send the information of the model to the second network device; if the type information of one or more terminals corresponding to the type information of the second terminal does not include the type information of the first terminal, it can be determined that the model does not serve the first terminal.

[0229] Figure 9 The illustrated process can serve as an example of this implementation method. Among them, Figure 9 In the illustrated process, after receiving a model retrieval request from UE1, the NWDAF generates a model. Upon receiving a model retrieval request from UE2, the NWDAF can provide the model information to the RAN serving UE2. Therefore, UE1 can be considered an example of a second terminal, and UE2 can be considered an example of a first terminal. Furthermore, Figure 9 The process should not be interpreted as a limiting process for this implementation method.

[0230] In another possible implementation, "the model information is determined based on the type information of the second terminal" can be understood as follows: the model information may include information indicating the model and the type information of the second terminal. Here, the type information of the second terminal can also be replaced with second type information. Specifically, the type information of the second terminal in the model information can be used to indicate that the model at least serves the second terminal or terminals in the set of second terminals. Furthermore, the model information may also serve the first terminal, or it can be considered that the model may be a customized model provided by the first network device to the second network device after receiving the first request information in S101.

[0231] As an example, the model's information may include the type information of the terminals served by the model, such as a list of terminal type information for the model service. The terminal type information for the model service may include the type information of a second terminal. Optionally, the terminal type information for the model service may also include the type information of a first terminal. Taking TAC as an example, the type information list may be a TAC list. In one possible embodiment, based on S103, the second network device can obtain the model or its parameters, as well as the TAC list, and can use the model to provide services to terminals with TACs in the TAC list (such as terminals in the first terminal set and / or terminals in the second terminal set).

[0232] In this implementation, the second network device can at least serve the second terminal through this model. The method by which the second network device serves the second terminal can be referred to... Figure 8 The process shown is, that is Figure 8 The process shown can serve as an example of how a second network device, after obtaining information about the model, provides services to a second terminal. Figure 8 UE1 can be considered as an example of a first terminal, and UE2 can be considered as an example of a second terminal. Furthermore, Figure 8 The process should not be interpreted as a restrictive process by which the second network device obtains information about the model and then provides services to the second terminal.

[0233] Optionally, the second network device may subsequently provide the first analysis service to the second terminal based on the model after receiving a service request or registration request from the second terminal, without needing to request the model from the first network device again. The service request can be used to request analysis services, such as requesting the first analysis service. Optionally, the service request or registration request may carry an analysis identifier for the first analysis service. Furthermore, the second network device may also provide the first analysis service to the second terminal based on the model after determining that the second terminal is allowed or authorized to customize the model of the first analysis service. The method for determining whether the second terminal is allowed or authorized to customize the model of the first analysis service can be found in the description of the method for determining whether the first terminal is allowed or authorized to customize the model of the first analysis service, and will not be repeated here.

[0234] For example, after receiving the type information of the second terminal and the identifier of the first analysis service from the fourth network device, the second terminal can determine that it is allowed or authorized to customize the model of the first analysis service. Alternatively, after receiving the type information of the second terminal from the fourth network device, the second network device can send a fifth request message to the third network device. The fifth request message can be used to request the analysis identifier of the analysis service corresponding to the second terminal, or it can be used to request the analysis identifier of the analysis service serving the second terminal, or it can be used to request the analysis identifier of the analysis service to which the model to which the second terminal is allowed or authorized to customize belongs. The fifth request message includes the identifier and / or type information of the second terminal. The second network device can also receive the analysis identifier from the third network device, which is determined based on the identifier and / or type information of the second terminal.

[0235] In another possible implementation, "the model information is determined based on the type information of the second terminal" can be understood or replaced as: the model is generated at least based on the training data of the second terminal, wherein the training data of the second terminal can be obtained based on the type information of the second terminal. The training data of the second terminal can also be replaced with the training data of terminals in the second terminal set. For example, the model can be obtained by training the model using the training data of terminals in the first terminal set and the training data of terminals in the second terminal set, or the model can be obtained by training the model using the training data of terminals in the second terminal set. This implementation may not require the first network device to receive the second request information. The terminals in the first terminal set include at least the first terminal, and the terminals in the second terminal set include at least the second terminal.

[0236] For example, in S102, the first network device obtains the type information of the second terminal based on the type information of the first terminal. This allows the model representing the first analysis service to serve both the first and / or second terminals. Therefore, the model can be generated at least based on the training data of the second terminal. For instance, the first network device can determine the second terminal based on its type information, collect its training data, and further train the model using both the training data from the first and second terminals to obtain the model. Alternatively, the first network device can determine the second terminal based on its type information, collect its training data, and further obtain the model based on the training data from the second terminal. In this implementation, the second network device can provide services to both the first and / or second terminals based at least on the model.

[0237] In addition, in this implementation, the model can be trained by the first network device, or the first network device can request or instruct other network devices or nodes to train the model; this application does not specifically limit this.

[0238] Figure 10 The illustrated process can serve as an example of this implementation method. Among them, Figure 10 In the illustrated process, after receiving the model retrieval request corresponding to UE1, NWDAF obtains other TACs corresponding to TAC1 from UDR. These other TACs may include TAC2 of UE2. NWDAF can further generate a model based on the training data of UE1 and UE2. This model can serve both UE1 and / or UE2. Therefore, UE1 can be considered as an example of a first terminal, and UE2 can be considered as an example of a second terminal. Furthermore, Figure 10 The process should not be interpreted as a limiting process for this implementation method.

[0239] In one possible embodiment, before executing S101, the second network device may further determine whether the first terminal is allowed or authorized to customize the model of the first analysis service. In this application, determining whether the first terminal is allowed or authorized to customize the model of the first analysis service can also be interpreted as or replaced by: determining whether terminals in the first terminal set are allowed or authorized to customize the model of the first analysis service, or determining whether the first terminal is allowed or authorized to obtain the service provided by the customized model of the first analysis service, or determining whether terminals in the first terminal set are allowed or authorized to obtain the service provided by the customized model of the first analysis service, or determining whether terminals in the first terminal set are allowed or authorized to obtain the first analysis service, or determining whether the second network device is authorized or authorized to provide the first analysis service to the first terminal or terminals in the first terminal set, or determining whether the second network device is authorized or authorized to provide services using the customized model of the first analysis service. If the first terminal is allowed or authorized to customize the model of the first analysis service, the second network device may execute S101, i.e., send first request information. The second network device can determine whether the first terminal is allowed or authorized to customize the model of the first analysis service based on the type information of the first terminal.

[0240] As an example, the second network device can request confirmation from the fifth network device, based on the type information of the first terminal, whether the first terminal is allowed or authorized to customize the model of the first analysis service. The fifth network device can be a UDR, UDM, EIR, or other network device, without specific limitation. The fifth network device and the third network device can be the same network device, network element, or node, or they can be different network devices, network elements, or different nodes. Specifically, the second network device can send a third request message to the fifth network device. The third request message can include the identifier of the first terminal and / or the type information of the first terminal. The third request message is used to request the analysis identifier corresponding to the analysis service of the first terminal, or it can be used to request the analysis identifier of the analysis service serving the first terminal, or it can be used to request the analysis identifier of the analysis service to which the model to which the first terminal is allowed or authorized to customize belongs.

[0241] In this application, the analysis identifier of the analysis service to which the model to which the terminal is allowed or authorized to be customized can also be replaced or interpreted as: the analysis identifier of the analysis service to which the terminal is allowed or authorized to obtain, the analysis identifier of the analysis service provided by the model to which the terminal is allowed or authorized to customize, the analysis identifier of the analysis service to which the terminal's type information is allowed to obtain, the analysis identifier of the analysis service to which the customized model to which the terminal's type information is allowed to obtain belongs, and the analysis identifier of the analysis service provided by the customized model to which the terminal's type information is allowed to obtain.

[0242] It can be assumed that the analysis identifier of the analysis service to which the model customized by the first terminal belongs includes the analysis identifier of the first analysis service.

[0243] If the second network device receives an analysis identifier from the fifth network device's first analysis service, it means that the first terminal is allowed or authorized to customize the model of the first analysis service. Alternatively, the analysis identifier of the first analysis service can be considered as indicating that the first terminal is allowed or authorized to customize the model of the first analysis service. For example, the second network device could receive an analysis identifier from the fifth network device for the analysis service to which the model the first terminal is allowed or authorized to customize belongs, including the analysis identifier of the first analysis service. Conversely, if the second network device does not receive an analysis identifier from the fifth network device's first analysis service, or receives a rejection response, it means that the first terminal is not allowed or authorized to customize the model of the first analysis service.

[0244] The following text is passed Figure 7The illustrated process describes how the second network device requests confirmation from the fifth network device, based on the type information of the first terminal, whether the first terminal is allowed or authorized to customize the model of the first analysis service, and how the model is obtained when the first terminal is allowed or authorized to customize the model of the first analysis service. That is... Figure 7 The corresponding steps can serve as an example of how the second network device requests confirmation from the fifth network device, based on the type information of the first terminal, whether the first terminal is allowed or authorized to customize the model of the first analysis service.

[0245] It can be assumed that the fifth network device can store the correspondence between the terminal's identification and / or type information and the analysis identifier of the analysis service to which the model that the terminal is allowed or authorized to customize belongs (which can be called the third correspondence), in order to identify whether the terminal is allowed or authorized to customize a model of a certain analysis service based on the terminal's identification and / or type information.

[0246] Furthermore, the fifth network device can also determine the type information of the first terminal based on its identifier, or determine the identifier of the first terminal based on its type information. Therefore, upon receiving the identifier of the first terminal, it can determine the type information of the first terminal based on that identifier, and then query the correspondence between the terminal's type information and the analysis identifier of the analysis service to which the model that the terminal is allowed or authorized to customize belongs, thereby determining the analysis identifier of the analysis service to which the model that the terminal is allowed or authorized to customize belongs. Alternatively, upon receiving the type information of the first terminal, the fifth network device can determine the identifier of the first terminal based on that type information, and then query the correspondence between the terminal's identifier and the analysis identifier of the analysis service to which the model that the terminal is allowed or authorized to customize belongs, thereby determining the analysis identifier of the analysis service to which the model that the terminal is allowed or authorized to customize belongs.

[0247] The third correspondence can be sent by the AF to the fifth network device. For example, the AF can request the fifth network device to allow or authorize the first terminal to obtain a model for the first analytics service, and this request can carry the type information of the first terminal. Accordingly, the fifth network device can store or obtain the third correspondence if it allows or authorizes the first terminal to obtain a model for the first analytics service. The fifth network device can also send a response message to the AF to indicate acceptance of the request. For example, the request provided by the AF can be used to apply for model customization for the first terminal for the first analytics service. The fifth network device can accept the request if the model customization conditions are met, such as proof of payment for the model customization service provided by the AF. Here, the AF can be the OTT system or OTT server of the equipment vendor of the first terminal and / or the second terminal.

[0248] Additionally, before executing S101, the second network device may also send an analysis identifier of the first analysis service to the fifth network device to request terminal type information permitted or authorized by the first analysis service. For example, the fifth network device stores the correspondence between the analysis identifier of the first analysis service and the terminal type information, and sends the terminal type information permitted or authorized by the first analysis service to the second network device according to the request of the second network device. If the correspondence includes a third correspondence, it means that the first terminal is permitted or authorized to customize the model of the first analysis service; conversely, if the correspondence does not include a third correspondence, it means that the first terminal is not permitted or authorized to customize the model of the first analysis service. It can be understood that after receiving the terminal type information permitted or authorized by the first analysis service, the second network device determines that the first terminal is permitted or authorized to customize the model of the first analysis service if the type information includes the first terminal's type information, and determines that the first terminal is not permitted or authorized to customize the model of the first analysis service if the type information does not include the first terminal's type information.

[0249] Additionally, before executing S101, the second network device may also send the type information of the first terminal to the fifth network device to request the analysis identifier of the analysis service that the type information of the first terminal is allowed or authorized to be customized. For example, the fifth network device stores the correspondence between the type information of the first terminal and the analysis identifier of the analysis service that the first terminal is allowed or authorized to customize, and sends the analysis identifier of the analysis service that the type information of the first terminal is allowed or authorized to be customized to the second network device according to the request of the second network device. If the correspondence includes a third correspondence, it represents a model where the first terminal is allowed or authorized to customize the first analysis service; conversely, if the correspondence does not include a third correspondence, it represents a model where the first terminal is not allowed or authorized to customize the first analysis service. It can be understood that after the second network device receives the analysis identifier of the analysis service that the first terminal is allowed or authorized to customize, if the analysis identifier of the analysis service includes the analysis identifier of the first analysis service, it can determine that the first terminal is allowed or authorized to customize the first analysis service; if the analysis identifier of the analysis service does not include the analysis identifier of the first analysis service, it can determine that the first terminal is not allowed or authorized to customize the first analysis service.

[0250] As one possible implementation, this third correspondence can be combined with the first correspondence. That is, the correspondence between the terminal type information and the analysis identifier of the analysis service can be carried in the TAC list, and this correspondence can serve as both the first and third correspondences. Alternatively, the third correspondence can also be independent of the first correspondence.

[0251] As another example, the second network device can determine whether the first terminal is allowed or authorized to customize the model of the first analysis service based on the type information of the first terminal and the analysis identifier information of the first analysis service from the fourth network device. That is, the analysis identifier of the first analysis service can indicate whether the first terminal is allowed or authorized to customize the model of the first analysis service. The fourth network device can determine whether the first terminal is allowed or authorized to customize the model of the first analysis service. The fourth network device can request confirmation from the fifth network device based on the identifier and / or type information of the first terminal regarding whether the first terminal is allowed or authorized to customize the model of the first analysis service. For example, the fourth network device can send a third request message to the fifth network device, which requests the analysis identifier of the analysis service corresponding to the first terminal, or requests the analysis identifier of the analysis service to which the allowed or authorized customized model of the first terminal belongs, and / or the third request message can be used to request the analysis identifier of the analysis service to which the allowed or authorized customized model of the first terminal belongs. For details, please refer to the method by which the second network device requests confirmation from the fifth network device regarding whether the first terminal is allowed or authorized to customize the model of the first analysis service, which will not be repeated here.

[0252] It is understood that, when the first terminal is allowed or authorized to customize the model of the first analysis service, the fourth network device can send the type information of the first terminal and the analysis identifier of the first analysis service to the second network device during or after the first terminal's registration process, to indicate that the first terminal is allowed or authorized to customize the model of the first analysis service. In other words, the analysis identifier is used to indicate that the first terminal is allowed or authorized to customize the model of the first analysis service. Conversely, if the first terminal is not allowed or authorized to customize the model of the first analysis service, the fourth network device does not need to send the type information of the first terminal and / or the analysis identifier of the first analysis service to the second network device; furthermore, if the first terminal is not allowed or authorized to customize the model of the first analysis service, the fourth network device can also send indication information to the second network device to indicate that the first terminal is not allowed or authorized to customize the model of the first analysis service.

[0253] In one possible implementation, the second network device may receive first instruction information before executing S101. This first instruction information can be used to instruct the first terminal to provide a first analysis service based on a customized model. Therefore, the second network device can send a first request message to the first network device after receiving the first instruction information. Alternatively, the first terminal can instruct or trigger the second network device to send the first request message via the first instruction information.

[0254] The first instruction information can be sent by the first terminal to the second network device during or after registration. Alternatively, the first terminal can send the first instruction information to the fourth network device during or after registration, and then the fourth network device can send the first instruction information to the second network device.

[0255] This term is used to instruct the provision of a first analysis service to a first terminal based on a customized model. It can also be replaced or interpreted as: used to request AI services, used to request a first analysis service, used to request high-performance analysis services (high performance, such as higher accuracy, lower energy consumption, or higher service response speed), used to request analysis services with higher accuracy, used to request analysis services with lower energy consumption, used to request analysis services with higher service response speed, used to request the provision of a first analysis service to the first terminal based on a customized model, used to request a customized model, used to instruct the model to obtain the first analysis service based on the training data of the first terminal, or used to request a customized first analysis service, etc.

[0256] The first instruction information can be a field, information element, or message containing specific content that can be used to provide a first analysis service to a first terminal based on a customized model. Alternatively, the message name and / or format can indicate the provision of the first analysis service to the first terminal based on a customized model.

[0257] As an example, the first instruction information may also include an analysis identifier of the first analysis service, used to indicate the analysis service and / or model. Alternatively, the first instruction information may be sent together with the analysis identifier of the first analysis service. For example, the second network device receives the first instruction information and the analysis identifier from the first terminal before performing S101.

[0258] The following text is passed Figure 6 The illustrated process describes how the fourth network device requests confirmation from the fifth network device, based on the type information of the first terminal, whether the first terminal is allowed or authorized to customize the model of the first analysis service, and how the model is obtained when the first terminal is allowed or authorized to customize the model of the first analysis service. That is... Figure 6 The corresponding steps can serve as an example of how the fourth network device requests confirmation from the fifth network device, based on the type information of the first terminal, whether the first terminal is allowed or authorized to customize the model of the first analysis service.

[0259] The following describes the flow of the communication method provided in this application, taking UE1 as the first terminal, UE2 as the second terminal, NWDAF as the first network device, RAN as the second network device, UDR as the third and fifth network devices, and AMF as the fourth network device, with reference to the accompanying drawings.

[0260] like Figure 6As shown, in the implementation of the model where the fourth network device determines whether the first terminal is allowed or authorized to customize the first analysis service, the communication method provided in this application may include the following steps:

[0261] S201: The OTT sends a service subscription request at the TAC level to the UDR. This request may include the analytics identifier of the analytics service and a list of TACs, used to request, subscribe to, or register for a model of analytics service corresponding to the analytics identifier for a terminal that has a TAC in the TAC list.

[0262] In this context, the analysis identifier corresponds to one or more TACs in the TAC list. The correspondence between the analysis identifier and the TACs in the TAC list can be considered as the aforementioned third correspondence, or as the first correspondence including the analysis identifier. Furthermore, the correspondence between multiple TACs in the TAC list can be considered as the aforementioned first correspondence.

[0263] Furthermore, any analysis identifier can correspond to one or more models, meaning that any analysis service can be executed by one or more models. These one or more models can serve terminals that have the TAC corresponding to that analysis identifier in the TAC list.

[0264] Taking analysis identifier #1 in Table 1 as an example, the service subscription request may include the correspondence between analysis identifier #1, TAC1 and TAC2, which is used to request, subscribe to or register for a model corresponding to analysis identifier #1 for a terminal with TAC1 and TAC2.

[0265] Optionally, the UDR sends a response message to the OTT. The response message may indicate that the UDR accepts the service subscription request.

[0266] It is understandable that S201 can serve as an example of a third network device obtaining a first or third correspondence.

[0267] S202: Optional, the RAN serving UE1 sends an analysis service acquisition request to the AMF to request the analysis identifier of the analysis service serving UE1, and / or to request permission or authorization for the analysis identifier of the analysis service to which the model customized by UE1 belongs.

[0268] The RAN can trigger the sending of an analysis service acquisition request after UE1 registers or during the registration process. This analysis service acquisition request can also have other names and is not specifically limited. For example, after the terminal registers, the RAN sends an analysis service acquisition request to the AMF by default to request the analysis identifier of the analysis service serving that terminal.

[0269] Optionally, the analysis service acquisition request may include first indication information, which can be used to instruct UE1 to provide a first analysis service based on or through a customized model. Alternatively, the first indication information can be used to instruct the model for obtaining the first analysis service based on UE1's training data. The RAN may receive the first indication information from UE1 after UE1 registers or during the UE1 registration process and include this indication in the analysis service acquisition request. The first indication information may also be referred to as an AI optimization indication or a model optimization indication, etc.

[0270] S203: The AMF sends an Analysis Service Acquisition Request to the UDR to request the analysis identifier of the analysis service to which the model customized by UE1 is permitted or authorized belongs. This Analysis Service Acquisition Request may also have other names, without specific limitations.

[0271] The analysis service retrieval request may carry the identifier of UE1 and / or TAC1. The identifier of UE1 may be, for example, UE1's SUPI or other identifiers.

[0272] For example, the AMF can execute S203 after receiving the analysis service acquisition request in S202. Alternatively, the AMF can execute S203 after receiving the first instruction information or analysis identifier of the first analysis service from UE1 during or after the UE1 registration process.

[0273] S204: UDR queries the TAC list based on UE1's TAC1 to obtain the analysis identifier of the analysis service corresponding to TAC1.

[0274] Taking Table 1 as an example, UDR can determine the analysis identifiers corresponding to TAC1, including analysis identifiers #1 and #3, based on the correspondence between analysis identifiers #1 and #3 in Table 1 and TAC1. Analysis identifiers #1 and / or #3 can be used as the analysis identifiers for the first analysis service.

[0275] It is understandable that if the TAC list contains the analysis identifier of the analysis service corresponding to TAC1, it means that UE1 is allowed or authorized to customize the model of that analysis service.

[0276] S205: The UDR sends a response message to the AMF, which includes the analysis identifier corresponding to TAC1. This analysis identifier can be the analysis identifier of the analysis service to which the model customized by UE1 is permitted or authorized belongs.

[0277] Accordingly, AMF can determine whether UE1 is allowed or authorized to customize the model of the analysis service corresponding to the analysis identifier based on the analysis identifier corresponding to TAC1 from UDR.

[0278] S203 and S204 can also be replaced by: The AMF queries the locally stored mapping between TAC1 and analysis identifiers based on UE1's TAC1 to obtain the analysis identifier of the analysis service corresponding to TAC1. The locally stored mapping between TAC1 and analysis identifiers can be obtained by the AMF from the UDR or other network devices. For example, during the previous query of the analysis identifier corresponding to the terminal's TAC, the AMF obtained the mapping between TAC1 and analysis identifiers from the UDR.

[0279] S206: AMF sends TAC1 to the RAN serving UE1.

[0280] Optionally, if the AMF obtains the analysis identifier corresponding to TAC1 from the UDR, the AMF may also send the analysis identifier corresponding to TAC1 to the RAN.

[0281] Optionally, TAC1 and / or the analysis identifier corresponding to TAC1 can be carried in the N2 message sent by the AMF to the RAN.

[0282] S206 can also be replaced by the AMF sending the masked IMEISV, IMEISV, masked PEI, PEI, masked IMEI, or IMEI of UE1 to the RAN, and the RAN determines TAC1 based on the masked IMEISV, PEI, or IMEI. It can be understood that the RAN can determine the required TAC based on the obtained XX, that is, ensure that the TAC is not masked, and whether other locations of the identifier are masked is not limited.

[0283] S207: The RAN sends a model retrieval request to the NWDAF. The model retrieval request can be used to request the model corresponding to TAC1, or in other words, the model request can be used to request the NWDAF to train the model corresponding to TAC1.

[0284] The model request may include TAC1, and optionally, it may also include the analysis identifier corresponding to the model. This analysis identifier is included in the analysis identifier of the analysis service to which the model that UE1 is allowed or authorized to customize belongs.

[0285] If the RAN receives the analysis identifier corresponding to TAC1 from the AMF, then the analysis identifier corresponding to TAC1 may include the analysis identifier corresponding to the model. As shown in Table 1, the analysis identifier corresponding to TAC1 includes analysis identifier #1 and analysis identifier #3, and the analysis identifier corresponding to the model may include one or more of analysis identifier #1 and analysis identifier #3. Analysis identifier #1 and / or analysis identifier #3 can be used as examples of the analysis identifier corresponding to the first analysis service.

[0286] Specifically, the RAN can determine the model serving UE1 or the analysis identifier corresponding to the model, and include the analysis identifier corresponding to the model in the model request. For example, the RAN can determine whether the requested UE1 is model #1 or model #2, and include analysis identifier #1 in the model request.

[0287] S207 can be seen as an example of one implementation of S101. The model retrieval request can be seen as an example of the first request information in S101.

[0288] TAC1 can serve as an example of the type information of the first terminal in S101. UEs with TAC1 include, but are not limited to, UE1.

[0289] S208: The NWDAF sends a TAC query request to the UDR to request other TACs corresponding to TAC1. This request may include TAC1. Optionally, the request may also include the analysis identifier corresponding to TAC1, to request other TACs corresponding to that analysis identifier and TAC1.

[0290] The TAC query request can also have other names, without specific restrictions.

[0291] S209: UDR queries the TAC list based on TAC1 to determine the other TACs corresponding to TAC1.

[0292] Taking Table 1 as an example, other TACs corresponding to TAC1 include TAC2 and TAC6. TAC2 and / or TAC6 can serve as examples of the type information of the second terminal in S102.

[0293] Optionally, if the TAC query request also includes the analysis identifier corresponding to TAC1, the UDR can query the TAC list based on TAC1 and the analysis identifier to determine other TACs corresponding to TAC1. Taking Table 1 as an example, if the analysis identifier corresponding to TAC1 is TAC#1, then the other TACs corresponding to TAC1 and analysis identifier #1 are TAC2. TAC2 can serve as an example of the type information of the second terminal in S102.

[0294] Optionally, without executing S203 to S205, before determining other TACs corresponding to TAC1, the UDR may also refer to S204 to query the TAC list based on TAC1 to determine if an analysis identifier corresponding to TAC1 exists, thus determining that TAC1 is allowed or authorized to customize the model of the analysis service. It can be understood that in S203 to S205, the UDR determines whether a corresponding analysis service exists for TAC1 based on the request sent by the AMF, while in S209, the UDR determines whether a corresponding analysis service exists for TAC1 based on the request from the NWDAF. The actions of determining whether an analysis identifier corresponding to TAC1 exists in S203 to S205 and S209 can be performed selectively.

[0295] S210: UDR sends a list of TACs to NWDAF, including TAC1 and the corresponding other TACs.

[0296] For example, UDR can send a list of TACs to NWDAF, which includes TAC1 and other TACs corresponding to TAC1.

[0297] Optionally, UDR can also send the analysis identifier corresponding to the TAC list to NWDAF.

[0298] For example, the UDR can send the analysis identifier #1 and the corresponding TAC list to the NWDAF, which includes TAC1 and TAC2, and / or the UDR can send the analysis identifier #2 and the corresponding TAC list to the NWDAF, which includes TAC1 and TAC6.

[0299] S208 to S210 can serve as examples of one implementation of S102.

[0300] S211: NWDAF sends model information to RAN.

[0301] The model information may include the model itself or its parameters, and may also include a list of TACs. This model information can serve as an example of the model information in S103. The model information includes a list of TACs, which may include TAC2, meaning the model information is determined based on TAC2.

[0302] The model or its parameters can be used by the RAN to determine the model as the terminal service.

[0303] The TAC list can be used to represent the terminals served by the model. For example, the TAC list includes TAC1 for UE1 and TAC2 for UE2. The NWDAF can also send TAC1 and TAC2 to the RAN to indicate that the model can serve UEs with TAC1 and / or UEs with TAC2. Here, UEs with TAC1 include UE1, and UEs with TAC2 include UE2. For example, the model can serve at least UE1 and / or UE2.

[0304] In this context, a UE with TAC1 can be considered as an example of a first set of terminals, and a UE with TAC2 can be considered as an example of a second set of terminals. UE2 can be one or more UEs with TAC2. For example, UE2 is a UE with TAC2 serving this RAN.

[0305] As an example of obtaining the model in S211, NWDAF can train the model based on the training data of the terminal corresponding to the TAC in the TAC list in S210 to obtain the model. For example, if the TAC list includes TAC1 of UE1 and TAC2 of UE2, NWDAF can collect the training data of the UE with TAC1 and / or the UE with TAC2, and train the model based on the training data of the UE with TAC1 and / or the UE with TAC2 to obtain the model.

[0306] In addition, NWDAF can store the correspondence between the analysis identifier corresponding to the model and the TAC in the TAC list.

[0307] As another example of obtaining the S211 model, NWDAF can look up the already generated model based on the TAC in the TAC list. For example, if NWDAF has previously generated a model applicable to a UE with TAC2 (such as UE2) and stored the correspondence between the model or the model's parameters and TAC2, then before S211, NWDAF can query the correspondence based on TAC2 in the TAC list to obtain the model or model's parameters corresponding to TAC2.

[0308] S211 can serve as an example of one implementation of S103.

[0309] S212: The RAN provides services to one or more terminals that have TACs in the TAC list, according to the model. The terminal may include UE1.

[0310] For example, taking the first analysis service as an example of providing communication configuration between the terminal and the access network equipment, if the TAC list contains TAC1 and TAC2, the RAN can determine the communication configuration of UE1 and / or the communication configuration of UE2 based on this model.

[0311] Based on the above Figure 6 The illustrated process allows the RAN to request a model for the first analysis service from the NWDAF for UE1. The NWDAF can determine the TAC list based on UE1's TAC1 and provide the RAN with model information, which includes the TAC list, which may include TAC1 and TAC2. The RAN can then provide the first analysis service to terminals in the TAC list based on the model, without needing to retrain the model for each terminal. Before requesting the model for the first analysis service from the NWDAF, the RAN can obtain UE1's TAC1 from the AMF. Optionally, the RAN can also obtain the analysis identifier for the first analysis service from the AMF, thus determining whether UE1 is allowed or authorized to customize the model for the first analysis service.

[0312] like Figure 7 The diagram illustrates how a model for determining whether a first terminal is allowed or authorized to customize a first analysis service is implemented by a second network device. Figure 6 The difference in the process shown is that, Figure 6 In S203, the action of sending an analysis service acquisition request to the UDR is replaced by being performed by the RAN. See [link / reference]. Figure 7 S303 in the middle. Figure 7 The process may include the following steps:

[0313] S301: The OTT sends a service subscription request at the TAC level to the UDR. This request may include the analytics identifier of the analytics service and a list of TACs, used to request, subscribe to, or register for a model of analytics service corresponding to the analytics identifier for a terminal that has a TAC in the TAC list.

[0314] S301 can be found in the description of S201.

[0315] S302: The RAN serving UE1 obtains UE1's TAC1.

[0316] The RAN can obtain the TAC1 from the AMF during or after the UE1 registration process. For example, after registering during the registration process, the RAN requests the TAC1 of UE1 from the AMF. Alternatively, after UE1 registers, the AMF provides the RAN with the UE1's masked IMEISV, PEI, or IMEI, and the RAN determines the TAC1 based on the masked IMEISV, PEI, or IMEI.

[0317] This UE1 can serve as an example of a first terminal.

[0318] S303: The RAN sends an Analysis Service Acquisition Request to the UDR to request the analysis identifier of the analysis service serving UE1, and / or to request permission or authorization for the analysis identifier of the analysis service to which the model customized by UE1 belongs. This request may include TAC1.

[0319] S303 can be referenced from S203, the difference being that the actions performed by the AMF in S203 are replaced by those performed by the RAN in S203.

[0320] The RAN can trigger the sending of an analysis service acquisition request after UE1 registers or during the registration process. This analysis service acquisition request can also have other names and is not specifically limited. For example, after the terminal registers, the RAN sends an analysis service acquisition request to the UDR by default to request the analysis identifier of the analysis service serving UE1, and / or, request the analysis identifier of the analysis service to which the customized model of UE1 belongs. Alternatively, after obtaining the first indication information or the analysis identifier of the first analysis service from UE1, the RAN executes S302 and / or S303.

[0321] S304: UDR queries the TAC list based on UE1's TAC1 to obtain the analysis identifier of the analysis service corresponding to TAC1.

[0322] S304 can be referenced from S204.

[0323] S305: The UDR sends a response message to the RAN, which contains the analysis identifier corresponding to TAC1.

[0324] S305 can be referenced from S205.

[0325] S306: The RAN sends a model retrieval request to the NWDAF. The model retrieval request can be used to request the model corresponding to TAC1, or in other words, the model request can be used to request the NWDAF to train the model corresponding to TAC1.

[0326] S306 can be referenced from the explanation in S207.

[0327] S306 can be used as an example of one implementation of S101. TAC1 can be used as an example of the type information of the first terminal in S101.

[0328] S307: NWDAF sends a TAC query request to UDR to request the query of other TACs corresponding to TAC1, such as TAC2.

[0329] S308: UDR queries the TAC list based on TAC1 to determine the other TACs corresponding to TAC1.

[0330] S309: UDR sends a list of TACs to NWDAF, which includes TAC1 and the other corresponding TACs.

[0331] Optionally, UDR can also send the analysis identifier corresponding to the TAC list to NWDAF.

[0332] S310: NWDAF sends model information to RAN.

[0333] S311: The RAN can provide services to one or more terminals that have TACs in the TAC list, based on the model. The terminal may include UE1.

[0334] S308 to S311 can be referred to S209 to S212 respectively.

[0335] Based on the above Figure 7 The illustrated process allows the RAN to request a model for the first analysis service from the NWDAF for UE1. The NWDAF can determine the TAC list based on UE1's TAC1 and provide the model information to the RAN. This model information may include the TAC list, which can include TAC1 and TAC2. The RAN can then provide the first analysis service to UEs with TAC1 and TAC2 listed in the TAC list, without needing to request separate model training requests for each UE. Before requesting the model for the first analysis service from the NWDAF, the RAN can confirm with the UDR that UE1 is permitted or authorized to customize the model for the first analysis service.

[0336] like Figure 8 The diagram illustrates an exemplary process by which the RAN provides services to UE1 and UE2 according to a model. UE1's type information is TAC1, and UE2's type information is TAC2. Figure 8 UE1 can be used as an example of a first terminal, and UE2 can be used as an example of a second terminal.

[0337] like Figure 8 As shown, the process may include the following steps:

[0338] S400: The RAN obtains information about the model and a list of TACs, which includes TAC1 and TAC2.

[0339] The RAN can receive the model information and TAC list after requesting the first analysis service model for UE1 (such as executing S207 or S306).

[0340] The method by which RAN obtains this TAC list can be found in [reference]. Figure 6 S211 in Figure 7The description of S310, S103 or other implementations in this application is not specifically limited.

[0341] Optionally, the RAN can store the mapping between the analysis identifier of the first analysis service, model information, and the TAC list.

[0342] S401: The RAN receives a registration request or analysis service request from UE2.

[0343] The registration request can be used to request the registration of UE2, and the analysis service request can be used to request the analysis service for UE2. Optionally, the request may include the analysis identifier of the first analysis service, such as analysis identifier #1.

[0344] S402: RAN obtains TAC2 of UE2.

[0345] For example, referring to S302, the RAN can obtain TAC2 from the AMF during or after UE2's registration process and store the UE2 identifier and the corresponding relationship of TAC2. After receiving the request from UE2 in S401, the RAN can look up TAC2 based on the UE2 identifier.

[0346] S403: RAN determines that TAC2 is included in the TAC list.

[0347] Optionally, the RAN can query the correspondence between the analysis identifier, model information and TAC list of the first analysis service based on the analysis identifier and TAC2 carried in the request in S401, in order to determine whether there is a correspondence between the analysis identifier and TAC2.

[0348] S404: RAN serves UE2 according to the model corresponding to TAC2.

[0349] If the TAC list contains TAC2, the RAN can provide services to UE2 based on the model information corresponding to the TAC list. Specifically, if TAC1 corresponds to TAC2 in the TAC list, and information about the model corresponding to TAC1 exists, then that model information can be used as the model information for serving the UE with TAC2. UEs with TAC2 include, for example, UE2. Therefore, the RAN can determine the communication configuration of UE2 based on this model.

[0350] Based on the above Figure 8The process shown allows the RAN to store the mapping between the model information and the TAC list after receiving the model information and the TAC list. Subsequently, when the RAN receives a registration or service request from a terminal that has TACs in the TAC list, the RAN can query the model serving that terminal based on this mapping and provide analysis services to the terminal based on the model, without needing to request the model again.

[0351] like Figure 9 In the illustrated process, after receiving the model requesting the first analysis service for UE1, the NWDAF obtains the TAC list corresponding to the analysis identifier of the first analysis service and generates the model for the first analysis service. The NWDAF can also send model information to the RAN. This model information includes the model itself or its parameters. Furthermore, after receiving the model requesting the first analysis service for UE2, the NWDAF provides the model to UE2 based on the TAC list, without needing to retrain the model for UE2. UE1 can be considered as an example of a second terminal, and UE2 can be considered as an example of a first terminal.

[0352] like Figure 9 As shown, the process may include the following steps:

[0353] S501: The OTT sends a service subscription request at the TAC level to the UDR.

[0354] S501 can be found in the description of S201.

[0355] S502: The RAN serving UE1 sends a model acquisition request to the NWDAF, which includes UE1's TAC1, to request a model for UE1.

[0356] S502 can be referenced from the descriptions in S207 or S306.

[0357] Before S502, you can also execute S202 to S206, or S302 to S305.

[0358] Figure 9 In the process, UE1 serves as the second terminal, and TAC1 serves as an example of the type information of the second terminal.

[0359] S503: Optional, NWDAF sends a TAC query request to UDR to request the query of other TACs corresponding to TAC1.

[0360] S504: Optional, UDR queries the TAC list based on TAC1 to determine the other TACs corresponding to TAC1, including TAC2.

[0361] S505: Optional, the UDR sends a list of TACs to the NWDAF, which may include TAC1 and other TACs corresponding to TAC1, such as TAC2.

[0362] The above S503 to S505 can be found in S307 to S309 respectively, and will not be repeated here.

[0363] S503 to S505 can also be replaced by: NWDAF determining the TAC list based on the correspondence between TACs stored locally. This TAC list includes TAC1 and other TACs corresponding to TAC1, such as TAC2.

[0364] Accordingly, NWDAF can store a list of TACs, that is, it stores the correspondence between multiple TACs of the model service of the first analysis service. These multiple TACs can include TAC1 and TAC2.

[0365] S506: NWDAF sends model information to the RAN serving UE1.

[0366] Optionally, NWDAF may also send a list of TACs to the RAN, which includes TAC1 and optionally TAC2.

[0367] The model information includes the model or model parameters. This model can be obtained by NWDAF based on UE1's TAC1-triggered model training, meaning the model is obtained based on UE1's type information. Optionally, the model information may also include the model's identifier.

[0368] Optionally, NWDAF may generate the model if it is determined that the analysis identifier corresponding to the first analysis service does not have information on a corresponding model.

[0369] Understandably, unlike S310, the information for the model in S506 does not include a TAC list.

[0370] The RAN serving UE1 can use this model to serve UE1.

[0371] S507: The RAN serving UE2 sends a model acquisition request to the NWDAF, which includes UE2's TAC2, to request a model for UE2.

[0372] It is understandable that the RAN serving UE1 can be the same as or different from the RAN serving UE2.

[0373] S508: NWDAF determines that TAC2 is included in the TAC list.

[0374] The TAC list can be obtained in S504, or it can be received by NWDAF after receiving the model acquisition request in S507, sending TAC2 to UDR, and then receiving it from UDR.

[0375] S509: NWDAF sends the model information to the RAN serving UE2.

[0376] This includes model information such as the model itself, its parameters, or its identifier. The model identifier can be used to indicate the model, and this identifier is the same as or related to the model identifier contained in the model information in S506.

[0377] The RAN can provide services to UE2 based on the information from the model.

[0378] Based on the above Figure 9 The illustrated process allows the NWDAF to generate a model and store the correspondence between the model and UE2 after receiving the model request information for UE2. Upon receiving the model request information for UE1, the NWDAF can determine whether a correspondence exists between TAC1 and TAC2 based on the TAC list. If a correspondence exists between TAC1 and TAC2, as shown in S508, the NWDAF determines that TAC2 is included in the TAC list. The NWDAF can then provide the RAN with the model information corresponding to UE1, enabling the RAN to provide services to UE1 using this model, without needing to train a model for UE2, thus reducing model training overhead.

[0379] like Figure 10 In the illustrated process, after receiving the model requesting the first analysis service for UE1, NWDAF obtains the TAC list corresponding to the analysis identifier of the first analysis service. This TAC list includes TAC1 for UE1 and TAC2 for UE2. NWDAF can then generate model M1 to serve UE1 based on the training data of UE1 and UE2. For example... Figure 9 As shown, the process includes the following steps:

[0380] S601: The OTT sends a service subscription request at the TAC level to the UDR. This request may include the analytics identifier for the analytics service and TAC1. Additionally, the request may include a list of TACs corresponding to TAC1, including TAC2.

[0381] Wherein, TAC1 is the TAC of the terminal requesting, subscribing to, or registering a customized sub-model. TAC2 corresponds to TAC1, where TAC2 is the TAC of UE2. UE1 can be considered as an example of a first terminal, and UE2 can be considered as an example of a second terminal. It can be assumed that UE1 and UE2 have the same type, that UE1's hardware information is partially or completely the same as UE2's hardware information, and / or that UE1's software information is partially or completely the same as UE2's software information.

[0382] Figure 10 During the process, the model can serve UE1 and / or UE2. That is, the model does not necessarily have to serve UE1 and UE2. For example, if UE2 has not subscribed to the model customization service, the model does not serve UE2 but serves UE1. The model can be trained based on the training data of UE2 to improve the model accuracy.

[0383] S602: The RAN serving UE1 sends a model retrieval request to the NWDAF, which includes UE1's TAC1, to request a model for UE1 or a model applicable to UE1. It can be understood that the RAN requests a model for TAC1 or a model applicable to TAC1.

[0384] For S602, please refer to the descriptions in 207 or 306.

[0385] Before S602, S202 to S206, or S302 to S305, can also be executed.

[0386] S603: NWDAF sends a TAC query request to UDR to request other TACs corresponding to TAC1. Optionally, the query request may also include an analysis identifier.

[0387] For details regarding S603, please refer to the description in S208.

[0388] S604: UDR queries the TAC list based on TAC1 to determine the other TACs corresponding to TAC1, including TAC2.

[0389] For details regarding S604, please refer to the description in S209.

[0390] S605: UDR sends a list of TACs to NWDAF, including TAC1 and the corresponding other TACs.

[0391] S605 can refer to the description in S210.

[0392] S606: NWDAF generates a model based on the training data of UE1 and UE2.

[0393] In S606, NWDAF can collect training data for UE1 and UE2 based on TAC1 and TAC2 in the TAC list, and generate a model based on this training data. Compared to generating a model based solely on the training data of UE1, S606 can improve the accuracy of the model, thus improving the accuracy of the communication configuration generated by the model, and consequently improving the communication performance when communicating according to this communication configuration.

[0394] In S606, taking TAC1 and TAC2 as examples in the TAC list, if other TACs are also included in the TAC list, NWDAF can also collect training data of UEs with other TACs to generate the model.

[0395] S606 can be found in the description of S210.

[0396] S607: NWDAF sends model information to RAN.

[0397] S607 can be found in the description of S211.

[0398] S608: RAN provides services to UE1 according to the model.

[0399] For example, taking the first analysis service as an example of providing communication configuration between the terminal and the access network equipment, the RAN can determine the communication configuration of UE1 based on this model.

[0400] Based on the above Figure 10 The process shown allows NWDAF to generate a model based on the training data of UE1 and UE2 after receiving the model request information corresponding to UE1. This can expand the range of model training data and thus improve model accuracy.

[0401] This application also provides a communication method in which a second network device obtains a correspondence between terminal type information, which may be referred to as a fourth correspondence. This correspondence may include a correspondence between the type information of a first terminal and the type information of a second terminal. The second network device may also generate a model based on this correspondence, which can serve the first terminal and / or the second terminal. In other words, the second network device can generate a model that serves the first terminal and the second terminal as a first analysis service. The method may include the following steps:

[0402] S701: The second network device sends a fourth request message to the sixth network device. This fourth request message can be used to request the correspondence between multiple types of information of the terminal. The sixth network device can be an AMF, NWDAF, or UDR, etc.

[0403] This multiple type information can include multiple type information for multiple terminals. For example, the correspondence can include the correspondence between the type information of a first terminal and the type information of a second terminal. The terminal type information can be found in [reference needed]. Figure 4 The process details will not be repeated here.

[0404] S702: The second network device receives a fourth correspondence from the sixth network device. The fourth correspondence includes the correspondence between the type information of the first terminal and the type information of the second terminal.

[0405] Taking the sixth network device as a UDR as an example, the UDR can store a first correspondence, which is used to provide a fourth correspondence to the first network device after receiving the fourth request information. The fourth correspondence may include some or all of the correspondences in the first correspondence. For example, the first correspondence may include the correspondence between multiple types of terminal information, including the type information of the first terminal and the type information of the second terminal.

[0406] The first correspondence can be referred to Figure 4 The explanation in the document is as follows. Additionally, if the sixth network device is an AMF or NWDAF, the first mapping can be obtained from the UDR by the AMF or NWDAF.

[0407] As an example, if the fourth request information in S701 also includes an analysis identifier for the analysis service, and the first correspondence includes a correspondence between the analysis identifier and multiple type information, then in S702, the fourth correspondence may include type information of multiple terminals corresponding to the analysis identifier, such as a first terminal and a second terminal. The analysis service model can serve these multiple terminals. Alternatively, the multiple terminals may have the same type, and / or their hardware information may be partially or entirely the same, and / or their software information may be partially or entirely the same.

[0408] As another example, if the fourth request information in S701 also includes the type information of the first terminal, then in S702, the fourth correspondence may include the type information of the first terminal and the type information of the second terminal corresponding to the type information of the first terminal. This means that the first terminal and the second terminal can serve each other using the same model. In other words, the first terminal and the second terminal have the same type, and / or, the hardware information of the first terminal and the second terminal is partially or entirely the same, and / or, the software information of the first terminal and the second terminal is partially or entirely the same.

[0409] Furthermore, if the fourth request information includes an analysis identifier and the type information of the first terminal, and the first correspondence includes a correspondence between the analysis identifier and multiple types of information, then in S702, the fourth correspondence may include the type information of the second terminal corresponding to the analysis identifier and the type information of the first terminal.

[0410] S703: The second network device can obtain a model for the first analysis service based on the type information of the first terminal and / or the type information of the second terminal. This model can serve terminals in the first terminal set and / or terminals in the second terminal set.

[0411] Based on the process shown in S701 to S703, the second network device can request multiple types of information that have a corresponding relationship, and customize a model for the terminal with the multiple types of information according to the corresponding relationship. Therefore, it is not necessary to customize a model for each type of information terminal, which can reduce the amount of resources required for model training.

[0412] It is understood that, in order to achieve the functions in the above embodiments, each network device may include corresponding hardware structures and / or software modules for performing the above functions. Those skilled in the art should readily recognize that, based on the units and method steps described in conjunction with the embodiments disclosed in this application, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed by hardware or by computer software driving hardware depends on the specific application scenario and design constraints of the technical solution.

[0413] Figure 11 and Figure 12 The diagram illustrates the possible communication devices provided in the embodiments of this application. These communication devices can be used to implement the functions of any one of the first to sixth network devices in the above method embodiments, and thus can also achieve the beneficial effects of the above method embodiments.

[0414] like Figure 11 As shown, the communication device 1100 includes a processing unit 1110 and a transceiver unit 1120. The communication device 1100 is used to implement the above-mentioned... Figure 4 , Figures 6 to 10 The function of one or more of the first to sixth network devices in any of the illustrated method flows.

[0415] Wherein, when the communication device 1100 is used to realize Figure 4In the method embodiment shown, the first network device functions as follows: the transceiver unit 1120 is used to obtain first request information, as detailed in S101. The processing unit 1110 can be used to obtain the type information of the second terminal based on the type information of the first terminal, as detailed in S102. The transceiver unit 1120 can also be used to send model information, as detailed in S103.

[0416] For a more detailed description of the aforementioned processing unit 1110 and transceiver unit 1120, please refer to [reference needed]. Figure 4 , Figures 6 to 10 The relevant descriptions in any of the illustrated method embodiments are as follows. For example, the technical terms involved in the above operations of processing unit 1110 and transceiver unit 1120 can be found in the descriptions in the above method embodiments, and will not be repeated here.

[0417] like Figure 12 As shown, the communication device 1200 includes a processor 1210 and an interface circuit 1220. The processor 1210 and the interface circuit 1220 are coupled to each other. It is understood that the interface circuit 1220 can be a transceiver or an input / output interface. Optionally, the communication device 1200 may also include a memory 1230 for storing instructions executed by the processor 1210, or storing input data required by the processor 1210 to execute instructions, or storing data generated after the processor 1210 executes instructions.

[0418] When the communication device 1200 is used to implement Figure 4 , Figures 6 to 10 In any of the methods shown, the processor 1210 is used to implement the functions of the processing unit 1110, and the interface circuit 1220 is used to implement the functions of the transceiver unit 1120.

[0419] When the aforementioned communication device is a chip applied in the first to sixth network devices, the chip implements the functions of one or more of the first to sixth network devices in the above method embodiments. The chip can receive information sent to the device by other network elements or devices through other modules in the device (such as radio frequency modules, antennas, or communication interfaces), or the chip can send information to other modules in the device (such as radio frequency modules, antennas, or interfaces). This information is sent by the device to other network elements or devices.

[0420] It is understood that the processor in the embodiments of this application may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. A general-purpose processor may be a microprocessor or any conventional processor.

[0421] The method steps in the embodiments of this application can be implemented in hardware or in software instructions executable by a processor. The software instructions can consist of corresponding software modules, which can be stored in random access memory, flash memory, read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, registers, hard disk, portable hard disk, compact disc read-only memory (CD-ROM), or any other form of storage medium well known in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. The storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. Furthermore, the ASIC can reside in one or more of the first to sixth network devices. The processor and storage medium can also exist as discrete components in one or more of the first to sixth network devices.

[0422] This application also provides a communication system, including one or more devices among the first to sixth network devices for implementing the above method embodiments.

[0423] This application also provides a computer-readable storage medium for storing computer programs or instructions that, when run, enable the methods described in the above method embodiments to be implemented.

[0424] This application also provides a computer program product that, when run on a computer, enables the method shown in the method embodiment to be implemented.

[0425] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the embodiments of this application are performed entirely or partially. The computer can be a general-purpose computer, a special-purpose computer, a computer network, a network device, a user equipment, or other programmable device. The computer program or instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer program or instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; it can also be an optical medium, such as a digital video optical disc; or it can be a semiconductor medium, such as a solid-state drive. The computer-readable storage medium may be a volatile or non-volatile storage medium, or may include both types of storage media.

[0426] In the description of this application, terms such as "first" and "second" are used only for the purpose of distinguishing descriptions and should not be construed as indicating or implying relative importance, nor should they be construed as indicating or implying order. It should be noted that the order in which "first," "second," etc., appear is not limited in this application; for example, "second" may appear first, followed by "first," and this application does not impose such a limitation.

[0427] In the description of this application, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can mean: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple. In the description of this application, " / " means "or," for example, a / b means a or b.

[0428] In the various embodiments of this application, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions of different embodiments are consistent and can be referenced by each other. The technical features of different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0429] It is understood that the various numerical designations used in the embodiments of this application are merely for descriptive convenience and are not intended to limit the scope of the embodiments of this application. The order of the process numbers described above does not imply the order of execution; the execution order of each process should be determined by its function and internal logic.

Claims

1. A communication method, characterized in that, Applied to the first network device, including: Receive a first request message from a second network device. The first request message is used to request a model corresponding to the first analysis service for the first terminal. The first request message includes the type information of the first terminal. The type information of the second terminal is obtained based on the type information of the first terminal; The model information is sent to the second network device, and the model information is determined based on the type information of the second terminal.

2. The method as described in claim 1, characterized in that, The hardware information of the first terminal is partially or entirely the same as the hardware information of the second terminal; and / or, The software information of the first terminal is partially or completely the same as the software information of the second terminal.

3. The method as described in claim 1 or 2, characterized in that, The step of obtaining the type information of the second terminal based on the type information of the first terminal includes: Send the type information of the first terminal to the third network device; The system receives type information of the second terminal from the third network device, wherein the type information of the second terminal is determined based on the type information of the first terminal.

4. The method as described in claim 1 or 2, characterized in that, The first request information includes the analysis identifier of the first analysis service, and obtaining the type information of the second terminal based on the type information of the first terminal includes: The type information of the second terminal is obtained based on the type information of the first terminal and the analysis identifier.

5. The method as described in claim 4, characterized in that, Obtaining the type information of the second terminal based on the type information of the first terminal and the analysis identifier includes: Send the type information of the first terminal and the analysis identifier to the third network device; The system receives type information of the second terminal from the third network device, the type information of the second terminal being determined based on the type information of the first terminal and the analysis identifier.

6. The method according to any one of claims 1-5, characterized in that, The information of the model is determined based on the type information of the second terminal, including: The information in the model includes the type information of the second terminal.

7. The method according to any one of claims 1-5, characterized in that, The information of the model is determined based on the type information of the second terminal, including: The model is generated in response to a second request information carrying type information of the second terminal, wherein the second request information is used to request the model corresponding to the first analysis service for the second terminal. After obtaining the type information of the second terminal based on the type information of the first terminal, the process further includes: The model is determined based on the type information of the second terminal.

8. The method as described in claim 7, characterized in that, The method further includes: Receive the second request information; In response to the second request message, the model is generated; Obtain the correspondence between the type information of the first terminal and the type information of the second terminal from the third network device; The step of obtaining the type information of the second terminal based on the type information of the first terminal includes: The type information of the second terminal is obtained based on the type information of the first terminal and the corresponding relationship.

9. The method according to any one of claims 1-5, characterized in that, The information of the model is determined based on the type information of the second terminal, including: The model is generated based on the training data of the second terminal, which is obtained based on the type information of the second terminal.

10. A communication method, characterized in that, Applied to second network devices, including: Send a first request message to a first network device. The first request message is used to request a model corresponding to the first analysis service for the first terminal. The first request message includes the type information of the first terminal. The system receives information about the model from the first network device. The information about the model is determined based on the type information of the second terminal, which is obtained based on the type of the first terminal.

11. The method as described in claim 10, characterized in that, The information of the model is determined based on the type information of the second terminal, including: The information in the model includes the type information of the second terminal.

12. The method as described in claim 11, characterized in that, The method further includes: The first analysis service is provided to the second terminal according to the model.

13. The method according to any one of claims 10-12, characterized in that, The information of the model is determined based on the type information of the second terminal, including: The model is generated based on the training data of the second terminal, which is obtained based on the type information of the second terminal.

14. The method according to any one of claims 10-13, characterized in that, Before sending the first request information to the first network device, the method further includes: Receive type information of the first terminal from the fourth network device; The analysis identifier of the first analysis service is obtained based on the type information of the first terminal. The analysis identifier is used to indicate that the first terminal is allowed to customize the model.

15. The method as described in claim 14, characterized in that, Obtaining the analysis identifier based on the type information of the first terminal includes: Send a third request message to a third network device. The third request message is used to request an analysis identifier corresponding to the analysis service of the first terminal. The third request message includes the identifier and / or type information of the first terminal. The analysis identifier is received from the third network device, the analysis identifier being determined based on the identifier and / or type information of the first terminal.

16. The method according to any one of claims 10-13, characterized in that, Before sending the first request information to the first network device, the method further includes: The system receives type information of the first terminal from a fourth network device and an analysis identifier from the first analysis service, the analysis identifier indicating that the first terminal is allowed to customize the model.

17. The method according to any one of claims 10-16, characterized in that, The hardware information of the first terminal is partially or entirely the same as the hardware information of the second terminal; and / or, The software information of the first terminal is partially or completely the same as the software information of the second terminal.

18. The method according to any one of claims 10-17, characterized in that, Before sending the first request information to the first network device, the method further includes: Receive first instruction information, which is used to instruct the first terminal to provide the first analysis service according to the customized model.

19. A communication method, characterized in that, Applied to third-party network devices, including: Receive type information of the first terminal from the first network device; Based on the correspondence between the type information of the first terminal and the type information of the second terminal, the type information of the second terminal is sent to the first network device.

20. The method as described in claim 19, characterized in that, The method further includes: Receive the analysis identifier from the first analysis service of the first network device; The step of sending the type information of the second terminal to the first network device according to the correspondence between the type information of the first terminal and the type information of the second terminal includes: Based on the correspondence between the analysis identifier, the type information of the first terminal, and the type information of the second terminal, the type information of the second terminal is sent to the first network device.

21. The method as described in claim 19 or 20, characterized in that, The method further includes: Receive a third request message, the third request message being used to request an analysis identifier corresponding to the analysis service of the first terminal, the third request message including the identifier and / or type information of the first terminal; The analysis identifier is sent, which is determined based on the identifier and / or type information of the first terminal.

22. The method according to any one of claims 19-21, characterized in that, The hardware information of the first terminal is partially or entirely the same as the hardware information of the second terminal; and / or, The software information of the first terminal is partially or completely the same as the software information of the second terminal.

23. A communication device, characterized in that, It includes units or modules for performing the method as described in any one of claims 1-9, or units or modules for performing the method as described in any one of claims 10-18, or units or modules for performing the method as described in any one of claims 19-22.

24. A communication device, characterized in that, Includes a processor for executing computer programs or instructions to implement the method as described in any one of claims 1-9, or the method as described in any one of claims 10-18, or the method as described in any one of claims 19-22.

25. A computer-readable storage medium, characterized in that, The storage medium stores a computer program or instructions, which, when executed by a communication device, implement the method as described in any one of claims 1-9, or the method as described in any one of claims 10-18, or the method as described in any one of claims 19-22.

26. A computer program product, characterized in that, When the computer program product is executed by a computer, the computer executes the method as described in any one of claims 1-9, or executes the method as described in any one of claims 10-18, or executes the method as described in any one of claims 19-22.

Images