Model processing method, network function, storage medium and product

Abstract

The invention discloses a model processing method. The method comprises the steps that a first network function receives a first request sent by a consumer; wherein the first request is used for requesting or subscribing an analysis result of the model; the first request comprises an analysis identifier or an identifier of the first model; the first network function determines a second network function based on the analysis identifier or the identifier of the first model; the first network function sends a second request to a second network function; wherein the second request is used for requesting or subscribing an analysis result of the first model; the second network function receives a second request; the second network function obtains an analysis result of the first model; the second network function sends an analysis result to the first network function; and the first network function receives the analysis result sent by the second network function. The invention further discloses a first network function, a second network function, a computer readable storage medium and a computer program product.

Description

Technical Field

[0001] The present application relates to but is not limited to the field of communications, and in particular to a model processing method, a first network function, a second network function, a computer-readable storage medium, and a computer program product. Background Art

[0002] Federated Learning is an emerging basic technology of artificial intelligence. Its design goal is to carry out efficient machine learning among multiple participants or multiple computing nodes under the premise of ensuring information security during big data exchange, protecting terminal data and personal data privacy, and ensuring legality and compliance. According to the different data distribution, Federated Learning can be divided into Horizontal Federated Learning and Vertical Federated Learning (VFL). In the related technology, the Network Data Analytical Function (NWDAF) in VFL assumes the function of the vertical federated learning server (VFL server), that is, participating in and leading the VFL model training and reasoning, and aggregating results.

[0003] However, with the in-depth application of machine learning and artificial intelligence technologies in the future, data may be widely distributed in various application functions (AF). Considering the privacy protection, resource consumption, latency and other issues brought about by data transmission, it is impossible to transfer all data to NWDAF for reasoning of the federated learning model. Summary of the invention

[0004] The embodiments of the present application provide a model processing method, a first network function, a second network function, a computer-readable storage medium, and a computer program product, and provide a method for NWDAF to control AF to implement vertical federated reasoning.

[0005] In a first aspect, an embodiment of the present application provides a model processing method, which is applied to a first network function, including:

[0006] Receive a first request sent by a consumer; wherein the first request is used to request or subscribe to an analysis result of a model; the first request includes an analysis identifier or an identifier of a first model;

[0007] Determining a second network function based on the analysis identifier or the identifier of the first model;

[0008] Sending a second request to the second network function; wherein the second request is used to request or subscribe to the analysis result of the first model;

[0009] An analysis result sent by the second network function is received.

[0010] In a second aspect, an embodiment of the present application provides a model processing method, which is applied to a second network function, including:

[0011] Receiving a second request sent by the first network function; wherein the second request is used to request or subscribe to an analysis result of the first model;

[0012] Obtaining analysis results of the first model;

[0013] The analysis result is sent to the first network function.

[0014] According to a third aspect, a first network function is provided, wherein the first network function includes:

[0015] A first receiving module, configured to receive a first request sent by a consumer; wherein the first request is used to request or subscribe to an analysis result of a model; and the first request includes an analysis identifier or an identifier of a first model;

[0016] A first processing module, configured to determine a second network function based on the analysis identifier or the identifier of the first model;

[0017] A first sending module, configured to send a second request to a second network function; wherein the second request is used to request or subscribe to an analysis result of the first model;

[0018] The first receiving module is further used to receive the analysis result sent by the second network function.

[0019] In a fourth aspect, a second network function is provided, where the second network function includes:

[0020] A second receiving module, configured to receive a second request sent by the first network function; wherein the second request is used to request or subscribe to an analysis result of the first model;

[0021] A second processing module, used for obtaining the analysis result of the first model;

[0022] The second sending module is used to send the analysis result to the first network function.

[0023] A fifth aspect, a first function, wherein the first function includes:

[0024] A first memory, used to store executable instructions;

[0025] The first processor is used to implement the above-mentioned model processing method when executing the executable instructions stored in the first memory.

[0026] A sixth aspect, a second function, wherein the second function includes:

[0027] A second memory, used for storing executable instructions;

[0028] The second processor is used to implement the above-mentioned model processing method when executing the executable instructions stored in the second memory.

[0029] In the seventh aspect, an embodiment of the present application provides a chip for implementing the above-mentioned model processing method; the chip includes: a processor for calling and running a computer program from a memory, so that a device equipped with the chip executes the above-mentioned model processing method.

[0030] In an eighth aspect, an embodiment of the present application provides a computer-readable storage medium for storing a computer program, which enables a computer to execute the above-mentioned communication method.

[0031] In a ninth aspect, an embodiment of the present application provides a computer program product, comprising computer program instructions, which enable a computer to execute the above-mentioned model processing method.

[0032] In a tenth aspect, an embodiment of the present application provides a computer program which, when executed on a computer, enables the computer to execute the above-mentioned model processing method.

[0033] Through the above solution, the first network function, namely the network data analysis function, one of which is implemented as NWDAF, offloads the reasoning, aggregation and training capabilities of the vertical federated model to the second network function, namely the application function as a VFL server, one of which is implemented as AF as a VFL server, and implements the process of vertical federated reasoning through AF, so that NWDAF can maximize the use of external vertical federated reasoning capabilities. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] Figure 1 is a schematic diagram of a network architecture applicable to embodiments of the present application;

[0035] Figure 2 Schematic diagram of the process of the model processing method provided in the embodiment of the present application Figure 1 ;

[0036] Figure 3 Schematic diagram of the process of the model processing method provided in the embodiment of the present application Figure 2 ;

[0037] Figure 4 Schematic diagram of the process of the model processing method provided in the embodiment of the present application Figure 3 ;

[0038] Figure 5 Schematic diagram of the process of the model processing method provided in the embodiment of the present application Figure 4 ;

[0039] Figure 6 A schematic block diagram of a first network function provided in an embodiment of the present application;

[0040] Figure 7 A schematic block diagram of a second network function provided in an embodiment of the present application;

[0041] Figure 8 A schematic structural diagram of a communication device provided in an embodiment of the present application;

[0042] Fig. 9 A schematic structural diagram of a chip provided in an embodiment of the present application. DETAILED DESCRIPTION

[0043] The following will describe the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0044] In the present application, "first", "second" and various numerical references are used to distinguish for the convenience of description and are not used to limit the scope of the embodiments of the present application. For example, different messages are distinguished, rather than used to describe a specific order or sequence. It should be understood that the objects described in this way can be interchanged where appropriate so as to be able to describe solutions other than the embodiments of the present application.

[0045] The technical solutions of the embodiments of the present application can be applied to various communication systems, such as: 5G system or new radio (NR), long term evolution (LTE) system, LTE frequency division duplex (FDD) system, LTE time division duplex (TDD), etc. The technical solutions provided in the present application can also be applied to future communication systems, such as the sixth generation mobile communication system. The technical solutions of the embodiments of the present application can also be applied to device to device (D2D) communication, vehicle-to-everything (V2X) communication, machine to machine (M2M) communication, machine type communication (MTC), and Internet of Things (IoT) communication system or other communication systems.

[0046] For the convenience of description, the 5G network will be used as an example in the embodiments of the present application.

[0047] Figure 1 1 is a schematic diagram of a network architecture 100 applicable to an embodiment of the present application. Figure 1 As shown, the network architecture may specifically include three parts, namely, the terminal equipment part, the data network (DN) and the operator network public land mobile network (PLMN) part. The functions of the network elements of each part are briefly described below.

[0048] The terminal equipment part may include a terminal device 110, which may also be referred to as a user equipment (UE). The terminal device 110 in the present application is a device with wireless transceiver functions, which may communicate with one or more core network (CN) devices via an access network device (or may also be referred to as an access device) in a radio access network (RAN) 130. The terminal device 110 may also be referred to as an access terminal, a terminal, a user unit, a user station, a mobile station, a mobile station, a remote station, a remote terminal, a mobile device, a user terminal, a user agent or a user device, etc. The terminal device 110 may be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; it may also be deployed on the water (e.g., a ship, etc.); it may also be deployed in the air (e.g., on an airplane, a balloon, and a satellite, etc.). The terminal device 110 may be a cellular phone, a cordless phone, a session initiation protocol (SIP) phone, a smart phone, a mobile phone, a wireless local loop (WLL) station, a personal digital assistant (PDA), etc. Alternatively, the terminal device 110 may also be a handheld device with wireless communication function, a computing device or other device connected to a wireless modem, a vehicle-mounted device, a wearable device, a drone device or a terminal in the Internet of Things or the Internet of Vehicles, a terminal in any form in a 5G network and future networks, a relay user device or a terminal in a future evolved 6G network, etc. Among them, the relay user device may be, for example, a 5G residential gateway (RG). For example, the terminal device 110 can be a virtual reality (VR) terminal, an augmented reality (AR) terminal, a wireless terminal in industrial control, a wireless terminal in unmanned driving, a wireless terminal in remote medical, a wireless terminal in a smart grid, a wireless terminal in transportation safety, a wireless terminal in a smart city, a wireless terminal in a smart home, etc. The terminal device here refers to a 3GPP terminal. The embodiments of the present application do not limit the type or type of terminal devices. For ease of explanation, the present application will be described later using UE as an example to refer to a terminal device.

[0049] The operator network PLMN part may include but is not limited to a (radio) access network ((R)AN) 120 and a core network (CN) part.

[0050] (R)AN 120 can be regarded as a sub-network of the operator network, and is an implementation system between the service nodes in the operator network and the terminal device 110. To access the operator network, the terminal device 110 first passes through (R)AN 120, and then can connect to the service nodes of the operator network through (R)AN 120. The access network device (RAN device) in the embodiment of the present application is a device that provides wireless communication functions for the terminal device 110, and can also be called a network device. The RAN device includes but is not limited to: the next generation node base station (gNB) in the 5G system, the evolved node B (eNB) in the long term evolution (LTE), the radio network controller (RNC), the node B (NB), the base station controller (BSC), the base transceiver station (BTS), the home base station (for example, home evolved node B, or home node B, HNB), the base band unit (BBU), the transmission point (TRP), the transmitting point (TP), the small base station device, the mobile switching center, or the network device in the future network. In systems using different wireless access technologies, the names of the devices with access network device functions may be different. For the convenience of description, in all embodiments of the present application, the above-mentioned devices that provide wireless communication functions for the terminal device 110 are collectively referred to as access network devices or RAN or AN for short. It should be understood that this document does not limit the specific type of access network equipment.

[0051] The CN part may include but is not limited to the following network functions (NF): user plane function (UPF), access and mobility management function (AMF), session management function (SMF), NWDAF, unified data management function (UDM), and network function repository function (NRF).

[0052] The data network DN 130, which may also be referred to as a packet data network (PDN), is usually a network located outside the operator network, such as a third-party network. Of course, in some implementations, the DN may also be deployed by the operator, that is, the DN is part of the PLMN. This application does not limit whether the DN belongs to the PLMN. The operator network PLMN may access multiple data network DNs 130, and multiple services may be deployed on the data network DN 130, which may provide services such as data and / or voice to the terminal device 110. For example, the data network DN 130 may be a private network of a smart factory, and the sensors installed in the workshop of the smart factory may be the terminal devices 110. The control server of the sensors is deployed in the data network DN 130, and the control server may provide services for the sensors. The sensors may communicate with the control server, obtain instructions from the control server, and transmit the collected sensor data to the control server according to the instructions. For another example, the data network DN 130 may be an internal office network of a company, and the mobile phones or computers of the company's employees may be the terminal devices 110, and the employees' mobile phones or computers may access information, data resources, etc. on the company's internal office network. The terminal device 110 can establish a connection with the operator network through an interface (such as N1, etc.) provided by the operator network, and use data and / or voice services provided by the operator network. The terminal device 110 can also access the data network DN 130 through the operator network, and use the operator services deployed on the data network DN 130, and / or services provided by a third party.

[0053] The following is a brief description of the NF functions included in CN.

[0054] UPF is a gateway provided by the operator and is the gateway for communication between the operator network and the data network DN 130. UPF includes user plane related functions such as packet routing and transmission, packet detection, service usage reporting, quality of service (QoS) processing, legal monitoring, uplink packet detection, and downlink packet storage.

[0055] AMF is a control plane network function provided by the operator network, responsible for access control and mobility management of the terminal device 110 accessing the operator network, including, for example, mobility status management, allocation of user temporary identity, authentication and authorization of users, and other functions.

[0056] SMF is a control plane network function provided by the operator network, responsible for managing the protocol data unit (PDU) session of the terminal device 110. A PDU session is a channel for transmitting PDUs, and the terminal device needs to transmit PDUs to and from the data network DN 130 through the PDU session. The SMF is responsible for establishing, maintaining, and deleting PDU sessions. SMF includes session management (such as session establishment, modification, and release, including tunnel maintenance between the user plane function UPF and (R)AN 120), UPF selection and control, service and session continuity (SSC) mode selection, roaming, and other session-related functions.

[0057] NRF is a control plane function provided by the operator, which can be used to maintain real-time information of network functions and services in the network. For example, it supports network service discovery, maintains the services supported by the NF profile of the NF instance, supports service discovery of the service communication proxy (SCP), maintains the SCP profile of the SCP instance, sends notifications about newly registered, unregistered, and updated NFs and SCPs, and maintains the health status of NF and SCP operations.

[0058] NWDAF is a control plane function provided by operators. Its main function is to collect data from NF, external application function (AF) and operation, administration and maintenance (OAM) system, and provide NWDAF service registration, data opening and data analysis for NF and AF.

[0059] Specifically, NWDAF can include the following logical functions:

[0060] Analytical Logical Function (AnLF): A logical function in NWDAF that is used to perform inference, derive analytical information (i.e., derive statistics and / or predictions based on analytical consumer requests), and expose analytical services.

[0061] Model Training Logical Function (MTLF): A logical function in NWDAF that trains a machine learning (ML) model and exposes new training services, e.g., serving trained ML models.

[0062] It should be understood that NWDAF can contain MTLF or AnLF or both logical functions. Wherein, NWDAF containing AnLF is locally configured with the NWDAF identity document (ID) containing MTLF and the analysis ID(s) supported by each NWDAF containing MTLF to retrieve the trained ML model. If necessary, NWDAF containing AnLF uses NWDAF discovery for NWDAF containing MTLF in the configuration ID set of NWDAF containing MTLF.

[0063] It should be understood that in order to ensure the accuracy of the analysis output of the analysis ID, based on the UE abnormal behavior analysis from itself or other NWDAFs, including the abnormal UE list and the observed time window, the NWDAF is used to detect and can delete the input data from the abnormal UE, and then a new ML model and / or analysis output can be generated for the analysis ID during the observed time window without the input data related to the abnormal UE list, and then the ML model information and / or analysis output is sent / updated to the subscribed NWDAF service consumers.

[0064] To support NF to discover and select an NWDAF instance containing MTLF, AnLF or both that is able to provide the required services for the required type of analytics (e.g. analytics exposure or ML model provisioning), each NWDAF instance, when registering with NRF, should provide a list of supported analytics IDs (perhaps one for each supported service) along with other NRF registration elements of the NF profile. A NF that needs to discover an NWDAF instance that provides support for some specific services for a specific type of analytics can query the NRF for NWDAFs that support the required services and the required analytics IDs.

[0065] The NWDAF architecture allows multiple NWDAF instances to be arranged in a hierarchy / tree with a flexible number of layers / branches. The number and organization of hierarchy layers and the functionality of each NWDAF instance remains a deployment choice.

[0066] In future communication systems, the network data analysis function network element may still be an NWDAF network element, or may have other names, which is not limited in this application.

[0067] It is understandable that the above network elements or functions can be physical entities in hardware devices, software instances running on dedicated hardware, or virtualized functions instantiated on a shared platform (e.g., a cloud platform). In short, an NF can be implemented by hardware or software.

[0068] Figure 1 Where Nnwdaf, Nausf, Namf, Nsmf, N1, N2, N3, N4, and N6 are interface serial numbers. For example, the meaning of the above interface serial numbers can refer to the meaning defined in the 3GPP standard protocol, and this application does not limit the meaning of the above interface serial numbers. It should be noted that, Figure 1 The interface name between the various network functions in the system architecture is only an example. In a specific implementation, the interface name of the system architecture may also be other names, which are not limited in this application. In addition, the name of the message (or signaling) transmitted between the above-mentioned network elements is only an example and does not constitute any limitation on the function of the message itself.

[0069] It should be noted that the interfaces between the control plane network elements may also be point-to-point interfaces, which will not be described in detail here.

[0070] It should be understood that the above-mentioned network architecture applicable to the embodiments of the present application is only an exemplary description, and the network architecture applicable to the embodiments of the present application is not limited to this. Any network architecture that can realize the functions of the above-mentioned network elements is applicable to the embodiments of the present application.

[0071] It should also be understood that Figure 1 The NRFs shown can be understood as network elements for implementing different functions, for example, they can be combined into network slices as needed. These network elements can be independent devices, or integrated into the same device to implement different functions, or they can be network elements in hardware devices, or they can be software functions running on dedicated hardware, or they can be virtualized functions instantiated on a platform (for example, a cloud platform). This application does not limit the specific form of the above network elements.

[0072] It should also be understood that the above naming is only defined for the convenience of distinguishing different functions and should not constitute any limitation to this application. This application does not exclude the possibility of using other naming in 5G networks and other future networks. For example, in a 6G network, some or all of the above network elements may use the terminology in 5G, or may use other names.

[0073] It should also be understood that Figure 1The interface name between each network element is only an example. The name of the interface in the specific implementation may be other names, and this application does not specifically limit this. In addition, the name of the message (or signaling) transmitted between the above network elements is only an example and does not constitute any limitation on the function of the message itself.

[0074] The network architecture and business scenarios described in the embodiments of the present application are intended to more clearly illustrate the technical solutions of the embodiments of the present application, and do not constitute a limitation on the technical solutions provided in the embodiments of the present application. A person of ordinary skill in the art can appreciate that with the evolution of the network architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of the present application are also applicable to similar technical problems.

[0075] In the 5G service-oriented system architecture, the two parties communicating based on the service-oriented interface are respectively called service consumers and service providers. Among them, the party requesting the service is called service consumer, and the party providing the service is called service producer. Service consumers can also be called consumers, consumer network elements, users, requesters or requesters, or service consumer network elements. Service providers can also be called providing network elements, service providing network elements, providers, producers or responders, etc., which are not limited in this application.

[0076] Before explaining this application, here is an explanation of vertical federated learning in related technologies:

[0077] The 5th Generation Core Network (5GC) supports vertical federated reasoning: training data is the core of building efficient machine learning models. Generally, the richer the training data, the better the performance of the model. However, due to the need to protect data privacy and take into account the diversity of data, a lot of data in 5GC and AF is not effectively utilized. To solve this problem, 5GC needs to introduce Federated Learning (FL) technology. Among them, VFL is suitable for scenarios where data sets have the same sample space but different feature spaces between participants. It allows us to improve the capabilities of machine learning models by combining data with different characteristics on 5GC and AF for joint training without sharing local data.

[0078] For better understanding, the following terms are defined as follows:

[0079] VFL server: A network element that performs result aggregation, integrates all local results, and calculates the gradient information or loss information required for machine learning model updates. It also coordinates the VFL process (e.g., selecting participants, exchanging intermediate results).

[0080] VFL client: A network element that holds a local dataset and performs local training. There can be multiple VFL clients in VFL.

[0081] It should be noted that TS23.288 provides NWDAF specification support for HFL, which mainly includes the process of performing federated learning through multiple NWDAF instances containing Model Training Logic Function (MTLF) in the core network, but does not provide VFL support; and as described in clause 5.2.4, the existing process defined for HFL can be enhanced to support VFL. The NWDAF federated learning process in TS 23.288 mainly includes the process of performing federated learning through multiple NWDAF instances containing MTLF in the core network. With the in-depth application of machine learning and artificial intelligence technologies in the future, data may be widely distributed on various AFs. Considering the privacy protection, resource consumption, latency and other issues brought by data transmission, it is impossible to transfer all data to NWDAF for reasoning of the federated learning model. The federated learning it supports is usually horizontal federated learning, that is, the data on different NWDAFs are distributed in different sample spaces but the same feature space; while the federated learning across AFs is vertical federated learning, that is, the data distributed on different AFs are distributed in different feature spaces and the same sample space.

[0082] In the VFL discussion scheme currently underway in TR 23.700, NWDAF and AF jointly complete the training and reasoning of the VFL model. In the core network vertical federated learning scheme currently underway in TR 23.700, NWDAF needs to act as a VFL server to coordinate the VFL processes of NWDAF and AF and aggregate the results simultaneously.

[0083] Figure 2 A flow chart of a model processing method provided in an embodiment of the present application is shown as follows: Figure 2 As shown, this method is applied to Figure 1 The network architecture 100 shown, the method includes:

[0084] Step 201: A first network function receives a first request sent by a consumer.

[0085] The first request is used to request or subscribe to the analysis result of the model; the first request includes an analysis identifier or an identifier of the first model.

[0086] In an embodiment of the present application, the first network function may be a model training logic function, one implementation form of which is MTLF; it may also be a network data analysis function including a model training logic function, one implementation form of which is NWDAF including MTLF; it may also be a network data analysis function, one implementation form of which is NWDAF; obviously, in the present application, the first network function belongs to a network function service provider (NFp).

[0087] In the embodiment of the present application, the second network function is an application function as a VFL server, and one implementation form is an AF as a VFL server.

[0088] In the embodiment of the present application, the first model may be a VFL model deployed in the second network function, that is, the first model is a VFL model. The number of first models may be at least one; that is, the consumer may send an identifier of a first model, or may send identifiers of at least two models, without limitation. The analysis identifier is the Analytics ID of the second network function.

[0089] In an embodiment of the present application, the identifier of the first model includes at least one of the following: a type identifier of the first model, a function identifier of the first model, a scenario identifier applicable to the first model, a data set identifier of the first model, a communication parameter identifier applicable to the first model, or a version identifier of the first model.

[0090] In the embodiment of the present application, the analysis result is the inference result for the first model.

[0091] In the embodiment of the present application, the first request is sent in the following ways: in-band, out-of-band, media, signaling, data, message, control plane, user plane, etc. Among them, the first request can be sent through the existing media channel to better be compatible with the existing system and reduce the cost of system transformation. In addition, when the first network function is multi-party, the established media plane communication channel is a one-to-many multicast / broadcast communication channel, so that the first request is only sent once through the established multicast / broadcast communication channel, and other first network functions can receive it, effectively reducing the amount of information sent.

[0092] In the embodiment of the present application, the analysis identifier is the analysis identifier of the second network function; of course, the first request may also include the NF instance ID (NF Instance Id(s)) of the second network function and the Vendor ID corresponding to the second network function.

[0093] Step 202: The first network function determines a second network function based on the analysis identifier or the identifier of the first model.

[0094] It should be noted that before the first network function determines the second network function based on the analysis identifier or the identifier of the first model, the second network function needs to register the VFL model information first.

[0095] Exemplarily, the second network function directly sends the relevant information of the model trained by itself to the third network function, and the first network function sends the first model identifier or analysis identifier to the third network function; the third network function obtains relevant information from the first network function based on the analysis identifier or the identifier of the first model, and sends the relevant information to the third network function, and the first network function determines the second network function based on the information returned by the third network function; that is, the first network function queries the corresponding service instance from the model registration information in the third network function based on the analysis identifier or the identifier of the first model; based on the service instance, the first network function determines the second network function that can obtain the analysis result of the first model.

[0096] In the embodiment of the present application, the third network function, namely the network function repository function, one of its implementation forms is NRF.

[0097] Exemplarily, the first network function includes a list integrating relevant information of multiple second network functions; the second network function directly sends the relevant information of the model trained by itself to the first network function, and the first network function matches the second network function from the first network function that can obtain the analysis result of the first model based on the analysis identifier or the identifier of the first model.

[0098] In the embodiment of the present application, the first network function, the second network function and the third network function may be a functional network element, and the functional network element or function may be a physical entity in a hardware device, a software instance running on dedicated hardware, or a virtualized function instantiated on a platform. In short, an NF may be implemented by hardware or by software.

[0099] Step 203: The first network function sends a second request to the second network function.

[0100] The second request is used to request or subscribe to the analysis result of the first model.

[0101] In the embodiment of the present application, the first request and the second request are both used to request or subscribe to the analysis results of the model, that is, to request or subscribe to the VFL model inference results. The first network function can directly pass through the request, then the second request and the first request are the same request; or, after receiving the first request, the first network function adds relevant information to the second request, then the first request and the second request are different requests.

[0102] In the embodiment of the present application, the second request is sent in the following ways: in-band, out-of-band, media, signaling, data, message, control plane, user plane, etc. Among them, the second request can be sent through the existing media channel to better be compatible with the existing system and reduce the cost of system transformation.

[0103] In the embodiment of the present application, the second request includes model inference parameters, and the model inference parameters include one or more of the following:

[0104] Analytical identification;

[0105] A terminal identifier, i.e., an identifier of a target terminal device, is used to determine a data sample for inference by the first model;

[0106] Feature configuration information;

[0107] An identifier associated with the first model.

[0108] Step 204: The second network function receives the second request.

[0109] Step 205: The second network function obtains the analysis result of the first model.

[0110] In an embodiment of the present application, obtaining the analysis result of the first model includes the second network function instructing the network function to manage, that is, the AF as a VFL client performs analysis of the federated learning model based on local data to obtain the intermediate analysis result of the first model; the second network function summarizes the intermediate analysis results to obtain the analysis result of the first model.

[0111] In an embodiment of the present application, obtaining the analysis result of the first model includes the second network function directly performing analysis of the federated learning model based on local data to obtain the analysis result.

[0112] Step 206: The second network function sends the analysis result to the first network function.

[0113] In the embodiment of the present application, the analysis results are sent in the following ways: in-band, out-of-band, media, signaling, data, message, control plane, user plane, etc. Among them, the analysis results can be sent through the existing media channel to better be compatible with the existing system and reduce the cost of system transformation. In addition, when the first network function is multi-party, the established media plane communication channel is a one-to-many multicast / broadcast communication channel, so that the analysis result is sent only once through the established multicast / broadcast communication channel, and other first network functions can receive it, effectively reducing the amount of information sent.

[0114] Step 207: The first network function receives the analysis result sent by the second network function.

[0115] The embodiment of the present application provides a model processing method, the method comprising: a first network function receives a first request sent by a consumer; wherein the first request is used to request or subscribe to the analysis result of the model; the first request includes an analysis identifier or an identifier of the first model; the first network function determines the second network function based on the analysis identifier or the identifier of the first model; the first network function sends a second request to the second network function; wherein the second request is used to request or subscribe to the analysis result of the first model; the second network function receives the second request; the second network function obtains the analysis result of the first model; the second network function sends the analysis result to the first network function; the first network function receives the analysis result sent by the second network function. That is to say, through the above scheme, the first network function, that is, the network data analysis function, one of which is implemented as NWDAF, unloads the reasoning, aggregation and training capabilities of the vertical federation model to the second network function, that is, the application function as a VFL server, one of which is implemented as AF as a VFL server, and implements the process of performing vertical federation reasoning through AF, so that NWDAF can maximize the use of external vertical federation reasoning capabilities.

[0116] In some embodiments, step 203, the first network function sends the second request to the second network function, which can be implemented by the following steps:

[0117] Sending a second request to the second network function through the fourth network function; wherein the request sent by the first network function to the fourth network function includes one or more of the following:

[0118] Analytical identification;

[0119] Terminal identification, i.e., identification of the reference terminal device;

[0120] Feature information, i.e. feature configuration information;

[0121] The fourth network function determines the identifier of the target terminal based on the identifier of the reference terminal device.

[0122] In the embodiment of the present application, the fourth network function can transparently transmit the communication data between the first network function and the second network function, for example, request and result data.

[0123] In some embodiments, the fourth network function is a network open function, one of which is implemented in the form of NEF.

[0124] It should be noted that determining the identity of the target terminal based on the identity of the reference terminal device includes the following steps: mapping the internal user equipment identity to the external user equipment identity.

[0125] In some embodiments, the method to be included in the present application further includes the following steps A1 to A5, or includes step B1:

[0126] Step A1: The first network function sends a third request to the second network function.

[0127] The third request instructs the second network function to obtain the analysis result of the federation model.

[0128] Step A2: The second network function receives the third request.

[0129] Step A3: The second network function trains the federated model to obtain a second model.

[0130] In the embodiment of the present application, the second model may include the first model.

[0131] Step A4: The second network function sends the first information to the first network function.

[0132] The first information includes one or more of the following:

[0133] An identifier of the trained second model in the second network function, or an analysis identifier;

[0134] an instance of a second network function;

[0135] The address of the second model.

[0136] In the embodiment of the present application, the first information is sent in the following ways: in-band, out-of-band, media, signaling, data, message, control plane, user plane, etc. Among them, the first information can be sent through the existing media channel to better be compatible with the existing system and reduce the cost of system transformation. In addition, when the first network function is multi-party, the established media plane communication channel is a one-to-many multicast / broadcast communication channel, so that the first information is only sent once through the established multicast / broadcast communication channel, and other first network functions can receive it, effectively reducing the number of information sent.

[0137] Step A5: The first network function receives the first information.

[0138] Here, each model / network function has unique identification information. In some embodiments, the identification information includes color identification, graphic identification, text identification, digital identification, location identification, etc. Different models / network functions can use the same type of identification information; of course, the types of identification information of each model / network function may not be exactly the same, or the types of identification information of each model / network function may be completely different.

[0139] Step B1: The second network function sends a fourth request to the third network function.

[0140] The fourth request is used for model registration; the fourth request includes one or more of the following:

[0141] An identifier of the trained second model in the second network function, or an analysis identifier;

[0142] An instance of a second network function.

[0143] It should be noted that step A1 and step A5 are that the first network function triggers the second network function to perform model training, the second network function sends the relevant information of the trained model to the first network function, the first network function stores the relevant information of the trained model in the second network function, and when the first network function determines that the second network function performs model inference, it can quickly locate the second network function that meets the conditions based on the analysis identifier or the identifier of the first model, and the relevant information of the trained model stored in the second network function.

[0144] It should be noted that step B1 is that the second network function directly sends the relevant information of the trained model to the third network function. The third network function stores the relevant information of the trained model in the second network function. When the first network function determines that the second network function performs model inference, it can be based on the analysis identifier or the identifier of the first model and the relevant information of the trained model in the second network function stored in the third network function. Quickly locate the second network function that meets the conditions.

[0145] The following is an explanation of an exemplary application of the embodiments of the present application in a practical application scenario.

[0146] This application proposes a process for completing the vertical federated learning model reasoning by AF based on NWDAF control. The process consists of two parts: AF as a vertical federated learning server (VFL server) registers the vertical federated learning model information with NWDAF; consumer NF requests or subscribes to the vertical federated learning model reasoning results through NWDAF.

[0147] Part 1: AF as VFL server registers VFL model information. This process is used after VFL model training is completed. VFL server AF registers VFL model with the core network. When initiating model inference requirements or subscriptions later, NWDAF can obtain VFL server information. There are two solutions:

[0148] Solution 1: VFL server AF registers the model with NRF, and NWDAF obtains model information from NRF.

[0149] Step 301: VFL server AF sends a model registration request, namely, Nnrf_NFManagement_NFRegister Request, to NRF through NEF.

[0150] Step 302: NRF stores VFL model information (Store VFL model info).

[0151] It should be noted that the AF as a VFL server registers its model with the NRF through the NEF, including its own AFinstance, Analytics ID, and VFL model ID. Based on this information, the corresponding VFL server AF instance can be found by the Analytics ID or VFL model ID. The NRF stores the AF and VFL model information of the VFL server.

[0152] Step 303: NWDAF sends a model discovery request, namely, Nnrf_NFDiscovery_Request, to NRF; the request includes Analytics ID and VFL model ID.

[0153] Step 304: NRF sends a model discovery request response, namely, Nnrf_NFDiscovery_RequestResponse, to NWDAF; the response includes AF instance.

[0154] It should be noted that when NWDAF receives a model inference request initiated by a consumer NF, it can query the corresponding VFL server instance from NRF through AnalyticsID or VFL model ID, thereby finding the AF that can start VFL model inference.

[0155] Solution 2: VFL server AF directly returns model information to NWDAF.

[0156] Step 401: NWDAF sends NWDAF trigger VFL model training (NWDAFtrigger VFL model training) to VFL server AF through NEF.

[0157] Step 402: The VFL server AF performs VFL model training.

[0158] Step 403: The VFL server AF sends VFL model configuration information (VFL model infoprovisioning) to the NWDAF through the NEF; the configuration information includes the AF instance, Analytics ID, and VFL model ID.

[0159] It should be noted that NWDAF initiates the training process of the vertical federated learning model to the VFL server AF through NEF. After the model training is completed, the model information is directly returned to NWDAF.

[0160] Part 2: Consumer NF requests or subscribes to the longitudinal federated learning model inference results through NWDAF.

[0161] After completing VFL model training and registering model information, the model consumer NF can trigger the model inference process and obtain the inference results. In the process proposed in this solution, the model consumer NF first sends a model inference request to NWDAF, and NWDAF triggers the subsequent process, thereby decoupling the model consumer from the external AF as the VFL server.

[0162] Step 501: Data consumer NF initiates a model inference request or subscription to NWDAF. Data consumer NF sends an analysis request / subscription to NWDAF by initiating Nnwdaf_AnalyticsInfo_Request or Nnwdaf_AnalyticsSubscription_Subscribe calls; wherein the model inference request or subscription includes parameters such as Analytics ID and VFL model ID.

[0163] Step 502: Based on the requested Analytics ID and VFL model ID parameters, NWDAF selects the corresponding NF instance from the model registration information as the VFL server to obtain the VFL model analysis results. Figure 3 or Figure 4 The corresponding process has been explained.

[0164] Step 503: NWDAF sends a VFL inference request to NEF;

[0165] Among them, the parameters in the VFL inference request include the internal UE ID (data samples that need to be model inferenced), the feature profile (instructing the VFL client NWDAF which features to use to generate intermediate inference results), and the VFL model related ID (instructing the NWDAF VFL client which VFL model should be used to generate intermediate inference results).

[0166] Step 504: NEF maps the internal UE IDs to the external UE IDs.

[0167] Step 505: NEF sends a VFL inference request to the AF selected as the VFL server;

[0168] The parameters in the VFL inference request include parameters such as external user device ID, feature configuration file, and VFL model related ID.

[0169] Step 506: The AF as the VFL server sends a VFL inference request to the VFL client participating in the inference.

[0170] The parameters in the VFL inference request include parameters such as external user device ID, feature configuration file, and VFL model related ID.

[0171] Step 507: The AF as the VFL client generates an intermediate inference result using local data.

[0172] Among them, through parameters such as external UE ID and feature profile, AF identifies relevant samples and features used as local data sets; according to the VFL model related ID, AF selects a local model to generate intermediate inference results.

[0173] Here, AF as VFL client performs local model inference based on the collected local data.

[0174] Step 508: The AF as the VFL client sends a VFL reasoning response to the VFL server, including the local reasoning result and the VFL model related ID. The AF should send the response before the maximum response time expires.

[0175] Step 509: The AF as the VFL server gathers the intermedia inference results produced by all VFL clients and generates the final inference results (VFL Server AF gathers the intermedia inference results to generate the final inference results).

[0176] Step 510: The AF as the VFL server sends a VFL inference response to the NWDAF through the NEF; wherein the VFL inference response includes a final inference result and a VFL model ID.

[0177] Step 511 : NWDAF provides the model inference result to the data consumer NF through Nnwdaf_AnalyticsInfo_Request response or Nnwdaf_AnalyticsSubscription_Notify.

[0178] It should be noted that the type of message selected by the NWDAF to send depends on the service used in step 501 .

[0179] The embodiment of the present application provides a first network function, which can be used to implement Figure 2 A model processing method provided by the corresponding embodiment, referring to Figure 6 As shown, the first network function 600 includes:

[0180] The first receiving module 601 is used to receive a first request sent by a consumer; wherein the first request is used to request or subscribe to the analysis result of the model; the first request includes an analysis identifier or an identifier of the first model;

[0181] A first processing module 602, configured to determine a second network function based on the analysis identifier or the identifier of the first model;

[0182] The first sending module 603 is used to send a second request to the second network function; wherein the second request is used to request or subscribe to the analysis result of the first model;

[0183] The first receiving module 601 is used to receive the analysis result sent by the second network function.

[0184] In other embodiments of the present application, the first processing module 602 is configured for the first network function to send a first model identifier or an analysis identifier to a third network function;

[0185] The first processing module 602 is used to determine the second network function according to the information returned by the third network function.

[0186] In other embodiments of the present application, the first sending module 603 is used to send a third request to the second network function; wherein the third request instructs the second network function to obtain an analysis result of the federation model;

[0187] The first receiving module 601 is configured to receive first information sent by the second network function; wherein the first information includes one or more of the following:

[0188] An identifier of the trained second model in the second network function, or an analysis identifier;

[0189] an instance of a second network function;

[0190] The address of the second model.

[0191] In other embodiments of the present application, the second request includes model inference parameters, and the model inference parameters include one or more of the following: analysis identifier; terminal identifier; feature information; identifier related to the first model.

[0192] In other embodiments of the present application, the first sending module 603 is used to send a second request to the second network function through the fourth network function; wherein the request sent by the first network function to the fourth network function includes one or more of the following: analysis identifier; terminal identifier; feature information; identifier related to the first model.

[0193] The description of the above device embodiment is similar to the description of the above method embodiment, and has similar beneficial effects as the method embodiment. For technical details not disclosed in the device embodiment of the present application, please refer to the description of the method embodiment of the present application for understanding.

[0194] It should be noted that in the embodiments of the present application, if the above-mentioned model processing method is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application can be essentially or partly reflected in the form of a software product, which is stored in a storage medium and includes several instructions for a terminal device to execute all or part of the methods of each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a U disk, a mobile hard disk, a ROM, a magnetic disk or an optical disk. In this way, the embodiments of the present application are not limited to any specific combination of hardware and software.

[0195] The embodiment of the present application provides a second network function, which can be used to implement Figure 2 A model processing method provided by the corresponding embodiment, referring to Figure 7 As shown, the second network function 700 includes:

[0196] The second receiving module 701 is used to receive a second request sent by the first network function; wherein the second request is used to request or subscribe to the analysis result of the first model;

[0197] The second processing module 702 is used to obtain the analysis result of the first model;

[0198] The second sending module 703 is used to send the analysis result to the first network function.

[0199] In other embodiments of the present application, the second sending module 703 is used to send a fourth request to the third network function; wherein the fourth request is used for model registration; and the fourth request includes one or more of the following:

[0200] An identifier of the trained second model in the second network function, or an analysis identifier;

[0201] An instance of a second network function.

[0202] In other embodiments of the present application, the second receiving module 701 is used to receive a third request sent by the first network function; wherein the third request instructs the second network function to obtain an analysis result of the federation model;

[0203] The second processing module 702 is used to train the federated model to obtain a second model;

[0204] The second sending module 703 is configured to send first information to the first network function; wherein the first information includes one or more of the following:

[0205] an identification of the second model, or an analysis identification;

[0206] an instance of a second network function;

[0207] The address of the second model.

[0208] In other embodiments of the present application, the second request includes model inference parameters, and the model inference parameters include one or more of the following:

[0209] Analytical identification;

[0210] Terminal identification;

[0211] Feature information;

[0212] An identifier associated with the first model.

[0213] The description of the above device embodiment is similar to the description of the above method embodiment, and has similar beneficial effects as the method embodiment. For technical details not disclosed in the device embodiment of the present application, please refer to the description of the method embodiment of the present application for understanding.

[0214] It should be noted that in the embodiments of the present application, if the above-mentioned model processing method is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the embodiment of the present application can be essentially or partly reflected in the form of a software product, which is stored in a storage medium and includes several instructions for a terminal device to execute all or part of the methods of each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as a U disk, a mobile hard disk, a ROM, a magnetic disk or an optical disk. In this way, the embodiments of the present application are not limited to any specific combination of hardware and software.

[0215] Figure 8 800 is a schematic structural diagram of a communication device 800 provided in an embodiment of the present application. The communication device can be a first network function or a second network function. Figure 8 The communication device 800 shown includes a first processor 810, and the first processor 810 can call and run a computer program from a memory to implement the method in the embodiment of the present application.

[0216] Alternatively, if Figure 8 As shown, the communication device 800 may further include a first memory 820. The first processor 810 may call and run a computer program from the first memory 820 to implement the method in the embodiment of the present application.

[0217] The first memory 820 may be a separate device independent of the first processor 810 , or may be integrated into the first processor 810 .

[0218] Alternatively, if Figure 8 As shown, the communication device 800 may further include a transceiver 830, and the first processor 810 may control the transceiver 830 to communicate with other devices, specifically, may send information or data to other devices, or receive information or data sent by other devices.

[0219] The transceiver 830 may include a transmitter and a receiver. The transceiver 830 may further include an antenna, and the number of the antennas may be one or more.

[0220] Optionally, the communication device 800 may specifically be the first network function / second network function of an embodiment of the present application, and the communication device 800 may implement the corresponding processes implemented by the first network function / second network function in each method of the embodiment of the present application. For the sake of brevity, they will not be repeated here.

[0221] Fig. 9 It is a schematic structural diagram of the chip of an embodiment of the present application. Fig. 9 The chip 900 shown includes a second processor 910, which can call and run a computer program from a memory to implement the method in the embodiment of the present application.

[0222] Alternatively, if Fig. 9 As shown, the chip 900 may further include a second memory 920. The second processor 910 may call and run a computer program from the second memory 920 to implement the method in the embodiment of the present application.

[0223] The second memory 920 may be a separate device independent of the second processor 910 , or may be integrated into the second processor 910 .

[0224] Optionally, the chip 900 may further include an input interface 930. The second processor 910 may control the input interface 930 to communicate with other devices or chips, and specifically, may obtain information or data sent by other devices or chips.

[0225] Optionally, the chip 900 may further include an output interface 940. The second processor 910 may control the output interface 940 to communicate with other devices or chips, and specifically, may output information or data to other devices or chips.

[0226] Optionally, the chip can be applied to the first network function / second network function in the embodiments of the present application, and the chip can implement the corresponding processes implemented by the first network function / second network function in the various methods of the embodiments of the present application. For the sake of brevity, they will not be repeated here.

[0227] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0228] It should be understood that the processor of the embodiment of the present application may be an integrated circuit chip with signal processing capabilities. In the implementation process, each step of the above method embodiment can be completed by the hardware integrated logic circuit or software instructions in the processor. The above processor can be a general processor, a digital signal processor (Digital Signal Processor, DSP), an application-specific integrated circuit (Application Specific Integrated Circuit, ASIC), a field programmable gate array (Field Programmable Gate Array, FPGA) or other programmable logic devices, discrete gates or transistor logic devices, discrete hardware components. The methods, steps and logic block diagrams disclosed in the embodiments of the present application can be implemented or executed. The general processor can be a microprocessor or the processor can also be any conventional processor. The steps of the method disclosed in the embodiment of the present application can be directly embodied as a hardware decoding processor to perform, or the hardware and software modules in the decoding processor are combined and performed. The software module can be located in a random access memory, a flash memory, a read-only memory, a programmable read-only memory or an electrically erasable programmable memory, a register, and other mature storage media in the art. The storage medium is located in the memory, and the processor reads the information in the memory and completes the steps of the above method in combination with its hardware.

[0229] As an embodiment, the processor may include one or more general-purpose central processing units (CPUs). Each of these processors may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. The processor here may refer to one or more devices, circuits, and / or processing cores for processing data (e.g., computer execution instructions).

[0230] It can be understood that the memory in the embodiments of the present application can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a ROM, a programmable read-only memory (Programmable ROM, PROM), an erasable programmable read-only memory (Erasable PROM, EPROM), an electrically erasable programmable read-only memory (Electrically EPROM, EEPROM) or a flash memory. The volatile memory can be a random access memory (Random Access Memory, RAM), which is used as an external cache. By way of example and not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous link dynamic random access memory (SLDRAM), and direct memory bus random access memory (DR RAM). It should be noted that the memory of the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0231] It should be understood that the above-mentioned memory is exemplary but not restrictive. For example, the memory in the embodiments of the present application may also be static random access memory (static RAM, SRAM), dynamic random access memory (dynamic RAM, DRAM), synchronous dynamic random access memory (synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (double data rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (enhanced SDRAM, ESDRAM), synchronous link dynamic random access memory (synch link DRAM, SLDRAM) and direct memory bus random access memory (Direct Rambus RAM, DR RAM), etc. That is to say, the memory in the embodiments of the present application is intended to include but not limited to these and any other suitable types of memory.

[0232] An embodiment of the present application also provides a computer-readable storage medium for storing a computer program.

[0233] Optionally, the computer-readable storage medium can be applied to the first network function / second network function in the embodiments of the present application, and the computer program enables the computer to execute the corresponding processes implemented by the first network function / second network function in the various methods of the embodiments of the present application. For the sake of brevity, they are not repeated here.

[0234] The embodiment of the present application also provides a computer program product, including a computer program, which can be executed by a processor of the communication device 800 or the chip 900 to complete the steps of any of the aforementioned methods.

[0235] In the above embodiments, all or part of the embodiments may be implemented by software, hardware, firmware or any combination thereof. When implemented by software, all or part of the embodiments may be implemented in the form of a computer program product.

[0236] The computer program product includes one or more computer instructions. When loading and executing the computer program instructions on a computer, the process or function according to the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instruction can be stored in a computer-readable storage medium, or transmitted from a computer-readable storage medium to another computer-readable storage medium. For example, the computer instruction can be transmitted from a website site, a computer, a server, or a data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode to another website site, computer, server, or data center. The computer-readable storage medium can be any available medium that a computer can store or a data storage device such as a server, a data center, etc. that includes one or more available media integration. Available media can be magnetic media, (e.g., floppy disk, hard disk, tape), optical media (e.g., DVD), or semiconductor media (e.g., solid-state hard disk (SSD)), etc.

[0237] The model processing method, the first network function, the second network function, the computer-readable storage medium and the computer program product provided in the embodiments of the present application are introduced in detail above. Specific examples are used in this article to illustrate the principles and implementation methods of the present application. The description of the above embodiments is only used to help understand the method of the present application and its core idea; at the same time, for those skilled in the art, according to the ideas of the present application, there will be changes in the specific implementation methods and application scopes. In summary, the content of this specification should not be understood as a limitation on the present application.

[0238] It should be understood that the "one embodiment" or "an embodiment" or "an embodiment of the present application" or "the aforementioned embodiment" or "some implementations" or "some embodiments" mentioned throughout the specification means that the specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, "in one embodiment" or "in an embodiment" or "an embodiment of the present application" or "the aforementioned embodiment" or "some implementations" or "some embodiments" appearing throughout the specification do not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in various embodiments of the present application, the size of the sequence number of the above-mentioned processes does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiment of the present application. The above-mentioned sequence numbers of the embodiments of the present application are only for description and do not represent the advantages and disadvantages of the embodiments.

[0239] In the absence of special instructions, the first network function / the second network function performs any step in the embodiment of the present application, and the processor of the first network function / the second network function may perform the step. Unless otherwise specified, the embodiment of the present application does not limit the order in which the first network function / the second network function performs the following steps. In addition, the methods used to process data in different embodiments may be the same method or different methods.

[0240] In the several embodiments provided in the present application, it should be understood that the disclosed devices and methods can be implemented in other ways. The device embodiments described above are only schematic. For example, the division of units is only a logical function division. There may be other division methods in actual implementation, such as: multiple units or components can be combined, or can be integrated into another system, or some features can be ignored or not executed. In addition, the coupling, direct coupling, or communication connection between the components shown or discussed can be through some interfaces, and the indirect coupling or communication connection of devices or units can be electrical, mechanical or other forms.

[0241] The units described above as separate components may or may not be physically separated, and the components displayed as units may or may not be physical units; they may be located in one place or distributed on multiple network units; some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0242] In addition, all functional units in the embodiments of the present application may be integrated into one processing unit, or each unit may be a separate unit, or two or more units may be integrated into one unit; the above-mentioned integrated units may be implemented in the form of hardware or in the form of hardware plus software functional units.

[0243] The methods disclosed in several method embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments.

[0244] The features disclosed in several product embodiments provided in this application can be arbitrarily combined without conflict to obtain new product embodiments.

[0245] The features disclosed in several method or device embodiments provided in this application can be arbitrarily combined without conflict to obtain new method embodiments or device embodiments.

[0246] A person of ordinary skill in the art can understand that all or part of the steps of implementing the above method embodiment can be completed by hardware related to program instructions, and the aforementioned program can be stored in a computer storage medium. When the program is executed, it executes the steps of the above method embodiment; and the aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROM, magnetic disks or optical disks.

[0247] Alternatively, if the above-mentioned integrated unit of the present application is implemented in the form of a software function module and sold or used as an independent product, it can also be stored in a computer storage medium. Based on such an understanding, the technical solution of the embodiment of the present application can be essentially or partly embodied in the form of a software product that contributes to the relevant technology. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the methods described in each embodiment of the present application. The aforementioned storage medium includes: various media that can store program codes, such as mobile storage devices, ROMs, magnetic disks, or optical disks.

[0248] As used in the embodiments of the present application and the appended claims, the singular forms "a," "an," "said," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.

[0249] It should be noted that in each embodiment involved in the present application, all steps may be executed or part of the steps may be executed as long as a complete technical solution can be formed.

[0250] The above is only an implementation method of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A model processing method, characterized in that: Applied to a first network function, the method comprises: Receive a first request sent by a consumer; wherein the first request is used to request or subscribe to an analysis result of a model; the first request includes an analysis identifier or an identifier of a first model; Determining a second network function based on the analysis identifier or the identifier of the first model; Sending a second request to the second network function; wherein the second request is used to request or subscribe to the analysis result of the first model; An analysis result sent by the second network function is received.

2. The method according to claim 1, characterized in that The determining the second network function based on the analysis identifier or the identifier of the first model includes: The first network function sends a first model identifier or an analysis identifier to the third network function; The second network function is determined according to the information returned by the third network function.

3. The method according to claim 1, characterized in that The method further comprises: Sending a third request to the second network function; wherein the third request instructs the second network function to obtain an analysis result of the federation model; Receive first information sent by the second network function; wherein the first information includes one or more of the following: An identifier of the trained second model in the second network function, or an analysis identifier; an instance of the second network function; The address of the second model.

4. The method according to claim 1, characterized in that: The second request includes model inference parameters, and the model inference parameters include one or more of the following: Analytical identification; Terminal identification; Feature information; An identifier associated with the first model.

5. The method according to claim 1, characterized in that The sending the second request to the second network function includes: Send a second request to the second network function through the fourth network function; wherein the request sent by the first network function to the fourth network function includes one or more of the following: analysis identifier; terminal identifier; feature information; identifier related to the first model.

6. A model processing method, characterized in that: Applied to the second network function, the method comprises: Receiving a second request sent by the first network function; wherein the second request is used to request or subscribe to an analysis result of the first model; Obtaining analysis results of the first model; The analysis result is sent to the first network function.

7. The method according to claim 6, characterized in that The method further comprises: Send a fourth request to the third network function; wherein the fourth request is for model registration; and the fourth request includes one or more of the following: An identifier of the second model that has been trained in the second network function, or an analysis identifier; an instance of the second network function.

8. The method according to claim 6, characterized in that The method further comprises: receiving a third request sent by the first network function; wherein the third request instructs the second network function to obtain an analysis result of a federation model; Train the federated model to obtain the second model; Sending first information to the first network function; wherein the first information includes one or more of the following: an identification of the second model, or an analysis identification; an instance of the second network function; The address of the second model.

9. The method according to claim 6, characterized in that The second request includes model inference parameters, and the model inference parameters include one or more of the following: Analytical identification; Terminal identification; Feature information; An identifier associated with the first model.

10. A first network function, characterized in that: The first network function includes: A first receiving module, configured to receive a first request sent by a consumer; wherein the first request is used to request or subscribe to an analysis result of a model; and the first request includes an analysis identifier or an identifier of a first model; A first processing module, configured to determine a second network function based on the analysis identifier or the identifier of the first model; A first sending module, configured to send a second request to a second network function; wherein the second request is used to request or subscribe to an analysis result of the first model; The first receiving module is further used to receive the analysis result sent by the second network function.

11. A second network function, characterized in that: The second network function includes: A second receiving module, configured to receive a second request sent by the first network function; wherein the second request is used to request or subscribe to an analysis result of the first model; A second processing module, used to obtain the analysis result of the first model; The second sending module is used to send the analysis result to the first network function.

12. A first network function, characterized in that: The first network function includes: A first memory, used to store executable instructions; The first processor is configured to implement the model processing method according to any one of claims 1 to 5 when executing the executable instructions stored in the first memory.

13. A second network function, characterized in that: The second network function includes: A second memory, used for storing executable instructions; The second processor is used to implement the model processing method according to any one of claims 6 to 9 when executing the executable instructions stored in the second memory.

14. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the model processing method described in any one of claims 1 to 5, or to implement the model processing method described in any one of claims 6 to 9.

15. A computer program product comprising a computer program, characterized in that When the computer program is executed by a processor, the model processing method according to any one of claims 1 to 5 is implemented, or the model processing method according to any one of claims 6 to 9 is implemented.

Images