Network element registration method, model request method, network element, and readable storage medium

Abstract

The present application discloses a network element registration method, a model request method, an apparatus, a network element, a communication system and a storage medium. The network element registration method of the embodiments of the present application includes: a first network element sends a network element registration request to a second network element, the network element registration request is used to register information of the first network element in the second network element, and the information of the first network element includes model training related capability information of the first network element, where the model training related capability information of the first network element includes at least one of model platform information supported by the first network element, model description method information supported by the first network element, manufacturer information of the first network element, network element target information with which the first network element can share a model, first indication information for indicating whether the first network element supports model sharing or model interoperation with other network elements, model accuracy information achievable by the first network element, and model training speed information of the first network element.

Description

[Technical field]

[0001] The present application relates to the field of communication technology, and specifically to a network element registration method, a model request method, an apparatus, a network element, a communication system, and a storage medium. [Background technology]

[0002] With the continuous development of communication technology, network elements can mutually transmit models and use them in common, but in the process of model transmission, network elements do not know what platform framework model the other party can receive and use, do not know what platform framework model the other party has, or whether the other party has a platform framework model that they need, which will result in unnecessary waste of resources such as time, traffic, and communication signals. For example, in the process of model transmission, if the model is of a different platform framework, the transmitted model cannot be used, so the success rate of model transmission is relatively low. Summary of the Invention [Problem to be solved by the invention]

[0003] The embodiments of the present application provide a network element registration method, a model request method, an apparatus, a network element, a communication system, and a storage medium, which can solve the problem that the success rate of model transmission in the conventional technology is relatively low. [Means for solving the problem]

[0004] According to a first aspect, a network element registration method is provided, the method including: a first network element sending a network element registration request to a second network element, the network element registration request is used to register information of the first network element to the second network element, the information of the first network element including model training related capability information of the first network element, where the model training related capability information of the first network element includes at least one of: model platform information supported by the first network element; model description method information supported by the first network element; manufacturer information of the first network element; network element target information with which the first network element can share a model; first indication information for indicating whether the first network element supports model sharing or model interoperation with other network elements; model accuracy information achievable by the first network element; and model training speed information of the first network element.

[0005] In the embodiment of the present application, the first network element can register the network element information of the first network element to the second network element through a network element registration request, thereby realizing the registration of the first network element to the second network element; since the network element registration information of the first network element includes model training related capability information, other network elements can search for a first network element that meets the model training demand through the second network element, thereby solving the problem of how the first network element can register its network element information to the second network element and be searched by other network elements.

[0006] According to a second aspect, a network element registration device for a first network element is provided, the network element registration device including a sending module, the sending module is used for sending a network element registration request to a second network element, the network element registration request is used for registering information of the first network element to the second network element, the information of the first network element includes model training related capability information of the first network element, where the model training related capability information of the first network element includes at least one of model platform information supported by the first network element, model description method information supported by the first network element, manufacturer information of the first network element, network element target information with which the first network element can share a model, first indication information for indicating whether the first network element supports model sharing or model interoperation with other network elements, model accuracy information achievable by the first network element, and model training speed information of the first network element.

[0007] According to a third aspect, a network element registration method is provided, the method including: a second network element receiving a network element registration request sent by a first network element, the network element registration request is used to register information of the first network element to the second network element, the information of the first network element including model training related capability information of the first network element, where the model training related capability information of the first network element includes at least one of: model platform information supported by the first network element; model description method information supported by the first network element; manufacturer information of the first network element; network element target information with which the first network element can share a model; first indication information for indicating whether the first network element supports model sharing or model interoperation with other network elements; model accuracy information achievable by the first network element; and model training speed information of the first network element.

[0008] In an embodiment of the present application, the second network element can receive a network element registration request sent by the first network element, thereby registering the network element information of the first network element with the second network element, and realize the registration of the first network element with the second network element; since the network element registration information of the first network element includes model training related capability information, other network elements can search for a first network element that meets the model training demand through the second network element, so as to solve the problem of how the first network element can register network element information with the second network element and be searched by other network elements.

[0009] According to a fourth aspect, a network element registration device for a second network element is provided, the network element registration device including: a receiving module, the receiving module is used for receiving a network element registration request sent by a first network element, the network element registration request is used for registering information of the first network element to the second network element, the information of the first network element includes model training related capability information of the first network element, where the model training related capability information of the first network element includes at least one of model platform information supported by the first network element, model description method information supported by the first network element, manufacturer information of the first network element, network element target information with which the first network element can share a model, first indication information for indicating whether the first network element supports model sharing or model interoperation with other network elements, model accuracy information achievable by the first network element, and model training speed information of the first network element.

[0010] According to a fifth aspect, there is provided a model request method, the method including: a third network element sending a model request message to a first network element, the model request message being used to request acquisition of a model, where the model request message includes second information, the second information including at least one of model platform information, model description method information, model accuracy information, manufacturer information of the third network element, and identifier information of the third network element, and the third network element acquiring information of a model matching the second information from the first network element.

[0011] In an embodiment of the present application, the third network element can request the first network element to obtain a model through a model request, and since the model request includes the second information, the third network element can obtain model information matching the second information from the first network element, thereby improving the success rate of model transmission.

[0012] According to a sixth aspect, there is provided a model request device for use in a third network element, the model request device including: a sending module and an acquiring module, the sending module is used for sending a model request message to a first network element, the model request message is used for requesting acquisition of a model, the model request message includes second information, the second information includes at least one of model platform information, model description method information, model accuracy information, manufacturer information of the third network element, and identifier information of the third network element, and the acquiring module is used for acquiring information of a model matching the second information from the first network element.

[0013] According to a seventh aspect, there is provided a model request method, the method including: a first network element receiving a model request message sent by a third network element, the model request message being used to request acquisition of a model, where the model request message includes second information, the second information including at least one of model platform information, model description method information, model accuracy information, manufacturer information of the third network element, and identifier information of the third network element, where the second information is used by the third network element to acquire information of a model matching the second information from the first network element.

[0014] In an embodiment of the present application, a first network element can receive a model request sent by a third network element, and the model request includes second information, so that the third network element can obtain model information matching the second information from the first network element, thereby improving the success rate of model transmission.

[0015] According to an eighth aspect, there is provided a model request device for use in a first network element, the model request device including: a receiving module, the receiving module is used for receiving a model request message sent by a third network element, the model request message is used for requesting acquisition of a model, where the model request message includes second information, the second information including at least one of model platform information, model description method information, model accuracy information, manufacturer information of the third network element, and identifier information of the third network element, where the second information is used by the third network element to acquire information of a model matching the second information from the first network element.

[0016] According to a ninth aspect, there is provided a network element comprising a processor and a memory, the memory storing a program or instructions operable to run on the processor, the program or instructions being operable when executed by the processor to realise the steps of the network element registration method described in the first aspect.

[0017] According to a tenth aspect, there is provided a network element, the network element including a processor and a communication interface, where the communication interface is used for sending a network element registration request to a second network element, the network element registration request is used for registering information of a first network element to the second network element, the information of the first network element including model training related capability information of the first network element, where the model training related capability information of the first network element includes at least one of model platform information supported by the first network element, model description method information supported by the first network element, model accuracy information achievable by the first network element, and model training speed information of the first network element.

[0018] According to an eleventh aspect, there is provided a network element comprising a processor and a memory, the memory storing a program or instructions operable to run on the processor, the program or instructions, when executed by the processor, realizing the steps of the network element registration method described in the third aspect.

[0019] According to a twelfth aspect, there is provided a network element, the network element including a processor and a communication interface, where the communication interface is used for receiving a network element registration request sent by a first network element, the network element registration request is used for registering information of the first network element to a second network element, the information of the first network element including model training related capability information of the first network element, where the model training related capability information of the first network element includes at least one of model platform information supported by the first network element, model description method information supported by the first network element, model accuracy information achievable by the first network element, and model training speed information of the first network element.

[0020] According to a thirteenth aspect, there is provided a network element comprising a processor and a memory, the memory storing a program or instructions operable to run on the processor, the program or instructions being operable when executed by the processor to implement the steps of the model request method described in the fifth aspect.

[0021] According to a fourteenth aspect, a network element is provided, the network element including a processor and a communication interface, where the communication interface is used to send a model request message to a first network element, the model request message is used to request acquisition of a model, where the model request message includes second information, the second information including at least one of model platform information, model description method information, model accuracy information, manufacturer information of the third network element, and identifier information of the third network element, and the third network element acquires information of a model matching the second information from the first network element.

[0022] According to a fifteenth aspect, there is provided a network element comprising a processor and a memory, the memory storing a program or instructions operable to run on the processor, the program or instructions, when executed by the processor, implementing the steps of the model request method described in the seventh aspect.

[0023] According to a sixteenth aspect, there is provided a network element, the network element including a processor and a communication interface, wherein the communication interface is used for receiving a model request message sent by a third network element, the model request message is used for requesting acquisition of a model, wherein the model request message includes second information, the second information including at least one of model platform information, model description method information, model accuracy information, manufacturer information of the third network element, and identifier information of the third network element, wherein the second information is used by the third network element to acquire information of a model matching the second information from the first network element.

[0024] According to a seventeenth aspect, there is provided a communication system, the communication system comprising a network element registration device according to the second aspect, e.g. a network element registration device according to the fourth aspect, e.g. a model request device according to the sixth aspect and a model request device according to the eighth aspect, or comprising a network element according to the ninth, eleventh, thirteenth and fifteenth aspects, or comprising a network element according to the tenth, twelfth, fourteenth and sixteenth aspects, wherein the network element may be used for performing steps of the method according to the first, third, fifth and seventh aspects.

[0025] According to an eighteenth aspect, there is provided a readable storage medium having a program or instructions stored thereon which, when executed by a processor, performs the steps of the method according to the first aspect, or performs the steps of the method according to the third aspect, or performs the steps of the method according to the fifth aspect, or performs the steps of the method according to the seventh aspect.

[0026] According to a nineteenth aspect, there is provided a chip, the chip including a processor and a communication interface, the communication interface coupled to the processor, the processor being adapted to run a program or instructions to implement the method according to the first aspect, or to implement the method according to the third aspect, or to implement the method according to the fifth aspect, or to implement the method according to the seventh aspect.

[0027] According to a twentieth aspect, there is provided a computer program / program product, the computer program / program product being stored on a storage medium, the computer program / program product being executed by at least one processor to implement steps of the network element registration method according to the first aspect, or to implement steps of the network element registration method according to the third aspect, or to implement steps of the model request method according to the fifth aspect, or to implement steps of the model request method according to the seventh aspect. [Brief description of the drawings]

[0028] [Figure 1] 1 is a schematic architecture diagram of a wireless communication system according to an embodiment of the present application; [Diagram 2] FIG. 2 is a schematic diagram of a neural network according to an embodiment of the present application. [Diagram 3] FIG. 2 is a schematic diagram of a neuron according to an embodiment of the present application. [Figure 4] 2 is a flowchart of a network element registration method according to an embodiment of the present application; [Diagram 5]2 is a second flowchart of a network element registration method according to an embodiment of the present application. [Figure 6] 3 is a third flowchart of a network element registration method according to an embodiment of the present application. [Figure 7] 4 is a fourth flowchart of a network element registration method according to an embodiment of the present application. [Figure 8] 4 is a flowchart of a model request method according to an embodiment of the present application; [Figure 9] 2 is a second flowchart of a model request method according to an embodiment of the present application. [Figure 10] 13 is a third flowchart of a model request method according to an embodiment of the present application. [Figure 11] 2 is a flowchart of a network element registration method and a model request method according to an embodiment of the present application; [Figure 12] FIG. 2 is a structural schematic diagram of a network element registration device according to an embodiment of the present application; [Figure 13] FIG. 2 is a second structural schematic diagram of a network element registration device according to an embodiment of the present application; [Figure 14] 1 is a structural schematic diagram of a model request device according to an embodiment of the present application; [Figure 15] FIG. 2 is a second structural schematic diagram of a model request device according to an embodiment of the present application; [Figure 16] 1 is a hardware structure schematic diagram of a communication device according to an embodiment of the present application; [Figure 17] FIG. 2 is a hardware structure schematic diagram of a network element according to an embodiment of the present application; DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0029] The following clearly describes the technical solutions in the embodiments of the present application in conjunction with the drawings in the embodiments of the present application, and it is obvious that the described embodiments are only some of the embodiments of the present application, and not all of the embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present application are all within the scope of protection of the present application.

[0030] The terms "first," "second," etc. in the specification and claims of the present application are intended to distinguish between similar objects and are not intended to describe a particular order or sequence. It is to be understood that the terms so used are interchangeable where appropriate, such that the embodiments of the present application may be performed in an order other than that shown or described herein, and that the objects distinguished by "first" and "second" are generally of the same type and do not limit the number of objects, e.g., the first object may be one or more. It is to be noted that "and / or" in the specification and claims represents at least one of the objects connected, and the character " / " generally represents an "or" relationship between the related objects.

[0031] It should be noted that the techniques described in the embodiments of the present application are not limited to Long Term Evolution (LTE) / LTE-Advanced (LTE-A) systems, but can also be applied to other wireless communication systems, such as Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal Frequency Division Multiple Access (OFDMA), Single-carrier Frequency Division Multiple Access (SC-FDMA), and other systems. The terms "system" and "network" in the embodiments of the present application are always used interchangeably, and the techniques described may be used in the above-mentioned systems and radio technologies, or in other systems and radio technologies. Although the following description describes a New Radio (NR) system for illustrative purposes and uses NR terminology in most of the description below, these techniques may be applied to applications other than NR system applications, such as 6th Generation (6G) communication systems.

[0032] 1 is a block diagram of a wireless communication system to which the embodiment of the present application can be applied. The wireless communication system includes a terminal 11 and a network side device 12. Here, the terminal 11 may be a terminal device such as a mobile phone, a tablet personal computer, a laptop computer (also called a notebook computer), a personal digital assistant (PDA), a palmtop computer, a netbook, an ultra-mobile personal computer (UMPC), a mobile Internet device (MID), an augmented reality (AR) / virtual reality (VR) device, a robot, a wearable device, a vehicle-mounted equipment (VUE), a pedestrian terminal (PUE), a smart home (home equipment having a wireless communication function, such as a refrigerator, a television, a washing machine, or furniture), a game console, a personal computer (PC), a deposit machine, or a self-service machine, and the wearable device includes a smart watch, a smart wristband, a smart earphone, a smart glasses, a smart accessory (a smart bracelet, a smart hand chain, a smart ring, a smart necklace, a smart ankle bracelet, a smart anklet, etc.), a smart band, a smart clothing, etc. It should be noted that the terminal 11 in the embodiment of the present application is not limited to a specific type. The network side equipment 12 may include an access network equipment or a core network equipment, where the access network equipment 12 may be called a radio access network equipment, a radio access network (Radio Access Network, RAN), a radio access network function or a radio access network unit.The access network equipment 12 may include a base station, a WLAN access point, or a WiFi node, and the base station may be called a Node B, an evolved Node B (eNB), an access point, a base transceiver station (BTS), a radio base station, a radio transceiver, a basic service set (BSS), an extended service set (ESS), a home B node, a home evolved B node, a transmitting receiving point (TRP), or any other suitable term in the art. As long as the same technical effect is achieved, the base station is not limited to a specific technical term, and it should be explained that in the embodiments of the present application, only base stations in an NR system are introduced as examples, and the specific type of the base station is not limited.The core network devices are: Core Network Node, Core Network Function, Mobility Management Entity (MME), Access and Mobility Management Function (AMF), Session Management Function (SMF), User Plane Function (UPF), Policy Control Function (PCF), Policy and Charging Rules Function (PCRF), Edge Application Server Discovery Function (EASDF), Unified Data Management (UDM), Unified Data Repository (UDR), Home Subscriber Server (HSS), Centralized network configuration (CNC), Network Repository Function (NRF), Network Exposure Function (NEF), Local NEF (Local NEF or L-NEF), Binding Support Function (BSF), Application Function (Application Node Function (ADF)), and Application Node Function (ADN). It should be noted that the embodiments of the present application only take the core network device in the NR system as an example, and do not limit the specific type of the core network device.

[0033] In the following, some concepts and / or terms related to the network element registration method, model request method, device, network element, communication system, and storage medium according to the embodiments of the present application will be explained and described.

[0034] 1. Artificial Intelligence (AI) and AI models: Artificial intelligence is now widely applied in various fields. AI models have various algorithm implementation methods, such as neural networks, decision trees, support vector machines, Bayesian classifiers, etc. This application takes neural networks as an example for explanation, but does not limit the specific type of AI module.

[0035] As shown in Figure 2, it is a schematic diagram of a neural network. X1, X2...Xn are input values, Y is the output result, and each JPEG2025511256000002.jpg44 indicates that each neuron is a place where an operation is performed, and the result is passed on to the next layer. The input layer, hidden layer, and output layer, which are made up of a large number of neurons, are one neural network. The number of hidden layers and the number of neurons in each layer are the "network structure" of the neural network.

[0036] Here, the neural network is composed of neurons, and as shown in Figure 3, which is a schematic diagram of a neuron. Here, a1, a2, ... aK (i.e., the above X1 ... Xn) are inputs, w is a weight (multiplication coefficient), b is a bias (addition coefficient), σ(z) is an activation function, and z is an output value. Common activation functions include Sigmoid, tanh, ReLU (Rectified Linear Unit), etc. The parameter information of each neuron combined with the algorithm used is the "parameter information" of the entire network, and is also an important part of the AI ​​model file.

[0037] In actual use, an AI model is a file containing elements such as network structure and parameter information. The trained AI model can be directly reused by the framework platform to directly perform intelligent functions such as judgment and identification without repeated construction or learning.

[0038] 2. Network Framework Neural networks have various implementation frameworks, including TensorFlow, PyTorch, Keras, MXNet, Caffe2, etc., and each framework has a different focus. For example, Caffe2 and Keras are upper-layer deep learning frameworks that can quickly verify models, while TensorFlow and PyTorch are lower-layer deep learning frameworks that can realize detailed modifications to the lower layers of neural networks.

[0039] For example, PyTorch focuses on supporting dynamic graph models, TensorFlow focuses on supporting various hardware and has a fast running speed, Caffe2 is at a lightweight level, etc. Each implementation framework uses its own method to describe neural networks and complete operations such as network construction, training, and inference. Generally, the methods of describing models in different implementation frameworks cannot be understood by other frameworks, so the models between them cannot be used interchangeably.

[0040] 3. Network Data Analytics Function (NWDAF) The network element can be decomposed into two parts: NWDAF (AnLF) and NWDAF (MTLF). The former is a network element responsible for the inference function (Analytics Logical Function, AnLF), and the latter is a network element responsible for the training function (Model Training Logical Function, MTLF).

[0041] In the following, the network element registration method and model request method according to the embodiments of the present application will be described in detail through several embodiments and application scenarios in conjunction with the drawings.

[0042] At present, the frameworks of each AI platform are independent of each other, and the models that are generated can only be converted into a very small number of models, which cannot communicate with each other and have no way to be used in other frameworks, which creates a major obstacle in practical use. Because the focus of each framework and therefore the development languages ​​supported are different, two nodes using different frameworks cannot transmit information about AI models, and two nodes using different frameworks cannot analyze networks trained in other frameworks because they have different network description methods, different data compression methods, and different file storage formats. Similarly, different developers have different framework definitions and description methods, which creates a relatively high barrier between them and makes it difficult to convert between them. Even within the same framework, the file storage format is different due to information such as different versions, and when different nodes of two framework versions transmit an AI model, the analysis results may be misaligned or may not be able to be analyzed.

[0043] In the current protocol standard, only the needs for network element registration and discovery, including network element identifier information, service and some capability information (NWDAF Capabilities, Analytics aggregation capability, Analytics metadata provisioning capability), are specified, but the necessary information regarding the model transmission is not included in the registration and discovery flow of the standard.

[0044] In an embodiment of the present application, information regarding model propagation is added in the smart network element registration and discovery process, and during registration, information regarding the model platform / framework it uses is additionally registered, and when a network element is discovered, such messages can also be filtered to select an appropriate network element.

[0045] Specifically, as a solution, the MTLF network element may register the network element information of the MTLF network element with the NRF network element through a network element registration request, thereby realizing the registration of the MTLF network element with the NRF network element. Since the network element registration information of the MTLF network element includes model training related capability information, other network elements can search for MTLF network elements that meet the model training needs through the NRF network element, thereby solving the problem of how the MTLF network element registers its network element information with the NRF network element and can be searched by other network elements.

[0046] As another solution, the NRF network element can receive a network element registration request sent by the MTLF network element, and register the network element information of the MTLF network element in the NRF network element, thereby realizing the registration of the MTLF network element in the NRF network element. Since the network element registration information of the MTLF network element includes model training related capability information, other network elements can search for MTLF network elements that meet the model training needs through the NRF network element, thereby solving the problem of how the MTLF network element registers its network element information in the NRF network element and can be searched by other network elements.

[0047] As another method, the AnLF network element may request the MTLF network element to obtain a model through a model request, and the model request includes second information, so that the AnLF network element can obtain model information matching the second information from the MTLF network element, thereby improving the success rate of model transmission.

[0048] As another method, the MTLF network element may receive a model request sent by an AnLF network element, and the model request includes second information, so that the AnLF network element can obtain model information matching the second information from the MTLF network element, thereby improving the success rate of model transmission.

[0049] An embodiment of the present application provides a network element registration method, and FIG. 4 shows a flowchart of the network element registration method according to an embodiment of the present application. As shown in FIG. 4, the network element registration method according to an embodiment of the present application may include the following steps 201 and 202:

[0050] Step 201: a first network element sends a network element registration request to a second network element.

[0051] In an embodiment of the present application, the network element registration request is used to register information of a first network element to a second network element, where the information of the first network element includes model training related capability information of the first network element.

[0052] Step 202, a second network element receives a network element registration request sent by a first network element.

[0053] In an embodiment of the present application, the model training related capability information of the first network element includes: Model platform information supported by the first network element; Model description method information supported by a first network element; Manufacturer information of the first network element; Network element target information with which the first network element can share a model; First indication information for indicating whether the first network element supports model sharing or model interoperation with other network elements; model accuracy information achievable by the first network element; and model training rate information of the first network element.

[0054] Optionally, in the embodiments of the present application, the first network element is a model training network element, which may be an MTLF network element, an NWDAF network element, or an NWDAF containing MTLF network element.

[0055] Optionally, in the embodiments of the present application, the second network element is a network element responsible for capability storage, which may be an NRF network element, a UDM network element, or a Data Collection Application Function (DCAF) network element.

[0056] Optionally, in an embodiment of the present application, the first network element may send a Nnrf_NFManagement_NFRegister Register network element registration request to the second network element to request the second network element to register information of the first network element in the second network element.

[0057] In an embodiment of the present application, the model platform information supported by the first network element is used to indicate the platforms that this network element can understand and use, e.g., TensorFlow, PyTorch, etc.

[0058] Optionally, in an embodiment of the present application, the first network element may be based on the model platform training model, or the model may be obtained based on the model platform training.

[0059] Optionally, in an embodiment of the present application, the model platform information supported by the first network element includes name or address information of one or more model platforms.

[0060] In an embodiment of the present application, the above model platform information is used to instruct the first network element to train and generate a model based on the model platform corresponding to the name or address information of the model platform.

[0061] In an embodiment of the present application, the model describing method information supported by the first network element is used to instruct this network element to describe the method technology used by the model.

[0062] Optionally, in the embodiment of the present application, the model description method information supported by the first network element is one of model format information and model language information.

[0063] Optionally, in the embodiment of the present application, the model language information may be an Open Neural Network Exchange (ONNX) language.

[0064] Optionally, in an embodiment of the present application, the first network element supports expressing the trained model in model format information or model language information.

[0065] Optionally, in an embodiment of the present application, the model accuracy information achievable by the first network element is used to instruct the first network element to describe the degree of discrimination or decision-making accuracy that can be achieved after training the model.

[0066] In the embodiments of the present application, model accuracy is used to indicate the degree of accuracy of the output results of a model.

[0067] In an embodiment of the present application, the model accuracy information may be the number of times the model decision result is correct divided by the total number of decisions, that is, model accuracy=number of correct results / total number.

[0068] Optionally, in the embodiment of the present application, the model accuracy information may be identified in the form of a decimal, a percentage, or high, medium, low.

[0069] Optionally, in the embodiments of the present application, the model training speed information of the first network element may be the time required for the first network element to obtain a model with a certain accuracy through training, for example, if the first network element trains one kind of model and needs 2 hours to achieve 90% accuracy, this time represents the model training speed information of the first network element.

[0070] Optionally, in an embodiment of the present application, the first instruction information indicating that the first network element supports model sharing or model interoperation with other network elements may be understood to mean that the first network element has the capability of model sharing or model interoperation with other network elements.

[0071] Optionally, in an embodiment of the present application, the information of the first network element is: Algorithm information used for the first network element training model; Manufacturer information of the first network element; Network element target information with which the first network element can share a model; Information on constraints under which the first network element shares the model; and the first network element stores the network element information of the training model.

[0072] Optionally, in the embodiments of the present application, the algorithm information used for the first network element training model may be an algorithm type related to AI such as machine learning that the first network element supports when training the model, such as deep learning, linear regression, gradient descent, etc., and the algorithm information can obtain the model supported by the first network element.

[0073] Optionally, in the embodiment of the present application, the manufacturer information of the first network element is used to indicate the manufacturer information to which this network element belongs.

[0074] In the embodiment of the present application, the network element target information with which the first network element can share a model is used to indicate the network element target information limited by this network element itself during model sharing and model transmission.

[0075] Optionally, in an embodiment of the present application, the network element target information with which the first network element can share a model includes at least one of information of one or more model inference function network elements and information of one or more model training function network elements.

[0076] Optionally, in the embodiment of the present application, the above network element object information is description information or identifier information of the network element object.

[0077] Optionally, in an embodiment of the present application, the limiting condition information under which the first network element shares the model includes at least one of time limiting condition information and area limiting condition information.

[0078] In an embodiment of the present application, the above-mentioned time restriction condition information is used to indicate the time during which the first network element allows model sharing, for example, from 7 p.m. to 6 a.m. the next day, and the area restriction condition information is used to indicate the area during which the first network element allows model sharing, for example, the location of the network element that shares the model with the first network element should be within this area restriction information, for example, "Suzhou."

[0079] In an embodiment of the present application, the network element information of the first network element stored training model is used to indicate the network elements to be stored in the model on which the first network element training has been completed.

[0080] Optionally, in the embodiments of the present application, the network element information of the first network element stored training model may be identifier information of a network element responsible for storing model information, such as a network element responsible for storing a model (Analytics Data Repository Function, ADRF), network element type information, etc.

[0081] Optionally, in an embodiment of the present application, the information of the first network element further includes at least one of version information of the first network element, type information of the first network element, identifier information of the first network element, and data analysis identifier information supported by the first network element.

[0082] In an embodiment of the present application, the version information of the first network element is used to indicate the version information of the model used by this network element.

[0083] In an embodiment of the present application, the type information of the first network element is used to indicate what type of network element has been registered this time, for example, if the type information of the first network element is an NWDAF network element type, it indicates that the network element registered this time is an NWDAF network element.

[0084] Optionally, in the embodiment of the present application, the identifier information of the first network element may be one of: Fully Qualified Domain Name (FQDN) information and IP address information.

[0085] In an embodiment of the present application, the full qualified domain name information is used to indicate the location of the first network element and the connectivity of the first network element.

[0086] In an embodiment of the present application, the data analysis identifier information supported by the first network element is used to indicate the task type that the network element is capable of performing.

[0087] Optionally, in an embodiment of the present application, the information of the first network element further includes at least one of: type information of training supported by the first network element, time information when the first network element supports federated learning training, and metadata information possessed by the first network element.

[0088] In an embodiment of the present application, the training type information supported by the first network element is used to indicate the AI ​​model training algorithm type supported by the first network element.

[0089] Optionally, in an embodiment of the present application, the type information of training supported by the first network element includes indication information of whether the first network element supports federated learning training, and at least one of a horizontal federated learning training type, a vertical federated learning training type, deep learning, linear regression, and gradient descent.

[0090] In an embodiment of the present application, the time information in which the first network element supports associative learning training is used to indicate the time to perform the associative learning training supported by the first network element, and performing the associative learning training within this time period, for example, from 10 p.m. to 6 a.m. the following morning, can obtain relatively good performance.

[0091] In an embodiment of the present application, the metadata information possessed by the first network element is used to indicate the data information that the first network element can cover, obtain and provide.

[0092] Optionally, in an embodiment of the present application, the metadata information possessed by the first network element includes at least one of an input data type, an output data type, a data amount, and a data range.

[0093] In the embodiment of the present application, the input data type is the input data type of the associative learning training, and the output data type is the output data type of the associative learning training.

[0094] In the embodiment of the present application, the data type refers to the presence or absence of a clear characteristic in the data within the region, for example, the data is concentrated in the morning and scattered in the evening.

[0095] In the embodiment of the present application, the data scope is the service scope of the first network element.

[0096] Optionally, in an embodiment of the present application, the data range includes at least one of a service area of ​​the first network element, an area from which the first network element can collect data, and an object from which the first network element can collect data.

[0097] Optionally, in an embodiment of the present application, the collectable data of the first network element includes metadata information and training data for model training.

[0098] Optionally, in an embodiment of the present application, the area in which the first network element can collect data may be any one of the first network element service area, a sub-area range under the first network element service area, and a target of finer granularity of collectable data (e.g., a single UE list).

[0099] Optionally, the embodiment of the present application further includes the following steps 203 and 204 after the above step 202.

[0100] Step 203, the second network element sends a registration response to the first network element.

[0101] In the embodiment of the present application, the registration response is used to indicate successful registration of the first network element.

[0102] Step 204, the first network element receives the registration response sent by the second network element.

[0103] Optionally, in an embodiment of the present application, the second network element may send a registration response Nnrf_NFManagement_NFRegister response to the first network element to notify the first network element of successful registration.

[0104] An embodiment of the present application further provides a network element registration method, in which a first network element can register the network element information of the first network element to a second network element through a network element registration request, thereby realizing the registration of the first network element to the second network element; since the network element registration information of the first network element includes model training related capability information, other network elements can search for a first network element that meets the model training demand through the second network element, so as to solve the problem of how the first network element can register its network element information to the second network element and be searched by other network elements.

[0105] Optionally, in an embodiment of the present application, as shown in FIG. 5 in combination with FIG. 4, after the above step 202, the network element registration method according to the embodiment of the present application further includes the following steps 205 and 206.

[0106] Step 205: the third network element sends a search request to the second network element.

[0107] In an embodiment of the present application, the search request is used to request a search for a target network element, and the search request includes at least one of the model training related capability information and the first information required by the third network element.

[0108] Step 206, the second network element receives the search request sent by the third network element.

[0109] In an embodiment of the present application, the model training related capability information required by the third network element includes at least one of: model platform information supported by the target network element required by the third network element; model description method information supported by the target network element required by the third network element; model accuracy information achieved by the target network element required by the third network element; and model training speed information of the target network element required by the third network element.

[0110] Optionally, in the embodiments of the present application, the third network element may be an AnLF network element, an NWDAF network element, or an NWDAF containing AnLF network element.

[0111] Optionally, in the embodiment of the present application, the third network element may send a Nnrf_NFDiscovery_Request search request to the second network element to request the second network element to feed back network element information that meets the conditions.

[0112] In an embodiment of the present application, the model platform information supported by the target network element required by the third network element is used to indicate a platform that the required target network element can understand and use, e.g., TensorFlow, PyTorch, etc.

[0113] Optionally, in the embodiment of the present application, the target network element may be based on the model platform training model, or the model may be obtained based on the model platform training.

[0114] Optionally, in the embodiment of the present application, the model platform information supported by the target network element includes name or address information of one or more model platforms.

[0115] In the embodiment of the present application, the above model platform information is used to instruct the target network element to train and generate a model based on the model platform corresponding to the name or address information of the model platform.

[0116] In the embodiment of the present application, the model describing method information supported by the target network element required by the third network element is used to indicate the method technology used by the required target network element description model.

[0117] Optionally, in an embodiment of the present application, the model accuracy information achieved by the target network element requested by the third network element is used to instruct the target network element requested to describe the degree of accuracy of discrimination or decision-making that can be achieved after training the model.

[0118] In the embodiments of the present application, model accuracy is used to indicate the degree of accuracy of the output results of a model.

[0119] Optionally, in an embodiment of the present application, the model training speed information of the target network element required by the third network element may be the time required for the required target network element to undergo training and obtain a model of a certain accuracy.

[0120] Optionally, in an embodiment of the present application, the search request further includes data analysis identifier information of the target network element required by the third network element.

[0121] Optionally, in an embodiment of the present application, the data analysis identifier information of the target network element required by the third network element is used to indicate a task type that the required target network element can support.

[0122] Optionally, in an embodiment of the present application, as shown in FIG. 6 in combination with FIG. 5, after the above step 206, the network element registration method according to the embodiment of the present application further includes the following step 207:

[0123] Step 207: the second network element determines a target network element based on the search request.

[0124] Optionally, the model training related capability information of the target network element matches the model training related capability information required by a third network element.

[0125] As can be seen, the second network element determines a network element whose model training related capability information matches the model training related capability information required by the third network element as a target network element.

[0126] Alternatively, when the query request includes model training related capability information required by a third network element, the above step 207 may be specifically implemented by at least one of the following steps 207a to 207d.

[0127] In step 207a, the second network element determines a target network element based on model platform information supported by the target network element required by the third network element.

[0128] In an embodiment of the present application, the target network element supports the model platform information required by the third network element.

[0129] As can be seen, the second network element determines, as the target network element, a network element that supports the model platform information required by the third network element.

[0130] In step 207b, the second network element determines a target network element based on the model description method information supported by the target network element required by the third network element.

[0131] In an embodiment of the present application, the target network element supports the model description method information required by the third network element.

[0132] As can be seen, the second network element determines, as the target network element, a network element that supports the model description method information required by the third network element.

[0133] In step 207c, the second network element determines a target network element based on the model accuracy information achieved by the target network element required by the third network element.

[0134] In an embodiment of the present application, the target network element supports the model accuracy information to be achieved required by a third network element.

[0135] As can be seen, the second network element determines as the target network element the network element that supports the model accuracy information desired by the third network element to be achieved.

[0136] In step 207d, the second network element determines a target network element based on the model training speed information of the target network element required by the third network element.

[0137] In an embodiment of the present application, the target network element supports model training rate information required by a third network element.

[0138] As can be seen, the second network element determines as the target network element the network element that supports the model training rate information required by the third network element.

[0139] Optionally, in an embodiment of the present application, the first information includes at least one of algorithm information used in the target network element training model demanded by the third network element, manufacturer information of the target network element demanded by the third network element, limiting condition information under which the target network element demanded by the third network element shares the model, identifier information of the third network element, and instruction information shared by the model.

[0140] Optionally, in an embodiment of the present application, the algorithm information used for the target network element training model required by the third network element may be an algorithm type related to AI such as machine learning that the target network element supports when training the model, such as deep learning, linear regression, gradient descent, etc., and this algorithm information can be used to obtain the model supported by the first network element.

[0141] Optionally, in the embodiment of the present application, the manufacturer information of the target network element required by the third network element is used to indicate the manufacturer information to which the required target network element belongs.

[0142] Optionally, in an embodiment of the present application, the limitation condition information for the target network element to share the model required by the third network element includes at least one of time information supported by the target network element and area limitation information of the target network element.

[0143] In an embodiment of the present application, the time information supported by the target network element demanded by the third network element is used to indicate the time during which the demanding target network element will allow model sharing, and the area restriction information of the target network element demanded by the third network element is used to indicate the area during which the demanding target network element will allow model sharing.

[0144] Optionally, in an embodiment of the present application, the model-shared indication information is used to indicate that the network element required by the third network element is a network element that supports model sharing or model interoperation with other network elements.

[0145] Optionally, the second network element may determine the target network element based on the first information in the search request.

[0146] Optionally, in the embodiment of the present application, when the search request includes first information, the above step 207 may be specifically realized by at least one of the following steps 207e to 207i.

[0147] In step 207e, the second network element determines a target network element based on the algorithm information used for the target network element training model required by the third network element.

[0148] In an embodiment of the present application, the target network element supports algorithm information used in the training model required by a third network element.

[0149] As can be seen, the second network element determines as the target network element the network element that supports the algorithm information used in the training model required by the third network element.

[0150] Step 207f: the second network element determines a target network element based on the manufacturer information of the target network element required by the third network element.

[0151] In an embodiment of the present application, the target network element supports manufacturer information required by a third network element.

[0152] As can be seen, the second network element determines as the target network element the network element that supports the manufacturer information required by the third network element.

[0153] In step 207g, the second network element determines a target network element based on the limiting condition information that the target network element required by the third network element shares the model.

[0154] Step 207h: the second network element determines a target network element based on the identifier information of the third network element, where the network element object information with which the target network element can share the model matches the identifier information of the third network element; understandably, the network element object information with which the target network element can share the model matching the identifier information of the third network element may refer to the network element object information with which the target network element can share the model including the identifier information of the third network element.

[0155] In step 207i, the second network element determines a target network element based on the indication information shared by the model, where the target network element supports model sharing or model interoperation with other network elements.

[0156] In an embodiment of the present application, the target network element supports condition information that allows sharing of the model required by a third network element.

[0157] As can be seen, the second network element determines as the target network element the network element that supports the constraint information for sharing the model required by the third network element.

[0158] Optionally, in an embodiment of the present application, as shown in FIG. 7 in combination with FIG. 6, after the above step 207, the network element registration method according to the embodiment of the present application further includes the following steps 208 and 209.

[0159] Step 208, the second network element sends the search response to the third network element.

[0160] In an embodiment of the present application, the search response includes targeting information.

[0161] Step 209, the third network element receives the search response sent by the second network element.

[0162] In an embodiment of the present application, the target information includes at least one of identifier information of the target network element, address information of the target network element, model platform information supported by the target network element, model description method information supported by the target network element, model accuracy information achievable by the target network element, model training speed information of the target network element, algorithm information used in the target network element training model, manufacturer information of the target network element, limiting condition information under which the target network element shares the model, and network element information of the target network element stored training model.

[0163] Optionally, in an embodiment of the present application, the target information further includes at least one of: a query validity period and version information of the target network element.

[0164] It should be noted that the information about the target network element can be seen in the above embodiment and will not be further described here.

[0165] An embodiment of the present application provides a network element registration method, in which a second network element can determine a target network element based on the search request by receiving a search request sent by a third network element, and can send a search response to the third network element to allow the third network element to determine the target network element, thereby subsequently requesting the target network element to obtain a model.

[0166] It should be mentioned that in the network element registration method according to the embodiment of the present application, the execution body may be a network element registration device, or a control module for executing the network element registration method in the network element registration device.

[0167] An embodiment of the present application provides a model request method, and Fig. 8 shows a flowchart of the model request method according to the embodiment of the present application. As shown in Fig. 8, the model request method according to the embodiment of the present application may include the following steps 301 to 303.

[0168] Step 301: a third network element sends a model request message to a first network element.

[0169] In the embodiment of the present application, the model request message is used to request acquisition of a model.

[0170] Step 302, a first network element receives a model request message sent by a third network element.

[0171] In an embodiment of the present application, the model request message includes second information, which includes at least one of model platform information, model description method information, model accuracy information, manufacturer information of a third network element, and identifier information of the third network element.

[0172] In an embodiment of the present application, the second information is used by a third network element to obtain model information matching the second information from the first network element.

[0173] Optionally, in the embodiments of the present application, the first network element is a model training network element, which may be an MTLF network element, an NWDAF network element, or an NWDAF containing MTLF network element.

[0174] Optionally, in the embodiments of the present application, the third network element may be an AnLF network element, an NWDAF network element, or an NWDAF containing AnLF network element.

[0175] Optionally, in an embodiment of the present application, the third network element may send a model request Nnwdaf_MLModelInfo_Request to the first network element to request acquisition of the model.

[0176] Optionally, in embodiments of the present application, the model platform information is used to indicate a platform that the demanding network element can understand and use, e.g., TensorFlow, PyTorch, etc.

[0177] Optionally, in an embodiment of the present application, the first network element may be based on the model platform training model, or the model may be obtained based on the model platform training.

[0178] Optionally, in an embodiment of the present application, the model platform information supported by the first network element includes name or address information of one or more model platforms.

[0179] In an embodiment of the present application, the above model platform information is used to instruct the first network element to train and generate a model based on the model platform corresponding to the name or address information of the model platform.

[0180] In the embodiment of the present application, the model description method information is used to indicate the method technology used by the desired network element description model.

[0181] Optionally, in the embodiment of the present application, the model description method information is one of model format information and model language information.

[0182] In an embodiment of the present application, the above model description method information is used to instruct the first network element to feed back the model to the third network element in the model format information or model language information.

[0183] Optionally, in the embodiment of the present application, the model language information may be ONNX (One Universal Model Description Language) language.

[0184] In an embodiment of the present application, the model accuracy information is used to instruct the network element in need to describe the degree of accuracy of discrimination or decision making that can be achieved after training the model.

[0185] In the embodiments of the present application, model accuracy is used to indicate the degree of accuracy of the output results of a model.

[0186] In an embodiment of the present application, the model accuracy information may be the number of times the model decision result is correct divided by the total number of decisions, that is, model accuracy=number of correct results / total number.

[0187] Optionally, in the embodiment of the present application, the model accuracy information may be identified in the form of a decimal, a percentage, or high, medium, low.

[0188] In an embodiment of the present application, the model request method further includes: the first network element obtaining, based on the second information, information of a model matching the second information.

[0189] Optionally, the first network element obtaining information of a model matching the second information based on the second information includes: The first network element acquires a target model based on model platform information included in the second information, and the target model is generated based on a model platform corresponding to the model platform information; The first network element obtains a target model based on model description method information included in the second information, and the target model is represented based on model language information or model format information corresponding to the model description method information.

[0190] The first network element obtains a target model according to the model accuracy information included in the second information, and the model accuracy of the target model matches the model accuracy information. Optionally, the model accuracy of the target model matches the model accuracy information may refer to the model accuracy of the target model being the same as the model accuracy indicated by the model accuracy information, or may refer to the model accuracy of the target model being within the model accuracy range indicated by the model accuracy information, or may refer to the model accuracy of the target model being greater than the model accuracy indicated by the model accuracy information, and this embodiment is not specifically limited. In an embodiment of the present application, the first network element can determine that the first network element allows model sharing or model interoperation with the third network element based on the manufacturer information of the third network element included in the second information, where the first network element allows model sharing or model interoperation with the third network element may be understood as the manufacturer information of the third network element being within the manufacturer information range that the first network element allows model sharing or model interoperation.

[0191] Or, In an embodiment of the present application, the first network element determines that the first network element allows model sharing or model interoperation with the third network element based on the identifier information of the third network element included in the second information, where the first network element allows model sharing or model interoperation with the third network element may be understood as the third network element being within the scope of the sharable network element target.

[0192] An embodiment of the present application provides a model request method, in which a third network element can request a first network element to obtain a model through a model request, and since the model request includes second information, the third network element can obtain model information matching the second information from the first network element, thereby improving the success rate of model transmission.

[0193] Step 303: the third network element obtains model information matching the second information from the first network element.

[0194] Optionally, in an embodiment of the present application, the information of the model includes at least one of: file information of the model, and download address information of the model file.

[0195] Optionally, in the embodiment of the present application, the file information of the model may include elements such as complete network structure and parameter information to generate a model corresponding to the target task.

[0196] Optionally, in an embodiment of the present application, the second information is: Information about the algorithm used to train the model; and Model version information, and feedback time information of the model.

[0197] In an embodiment of the present application, the algorithm information used in the training model is used to indicate the type of AI-related algorithm, such as machine learning, that the demanding network element supports when training the model, and the algorithm information can be used to obtain the model supported by the network element.

[0198] In the embodiment of the present application, the model version information is used to indicate the model version information used by the demanding network element.

[0199] In the embodiment of the present application, the model feedback time information is used to indicate the feedback time of the network element training model that requires, for example, the time required for model training or the deadline for model feedback.

[0200] In an embodiment of the present application, the first network element obtaining model information matching the second information based on the second information includes: the first network element obtaining a target model based on algorithm information used in the training model, and the algorithm type used by the target model is the same as the algorithm type indicated by the algorithm information used in the training model; The first network element obtains a target model according to the version information of the model, and the version information of the target model is the same as the version information of the model; The first network element includes at least one of: obtaining a target model based on the feedback time information of the model, where the feedback time of the target model matches the feedback time indicated by the feedback time information of the model, e.g., a time taken for model training of the target model is less than a time taken for model training indicated by the feedback time information of the model.

[0201] Optionally, in an embodiment of the present application, the second information further includes at least one of data analysis identifier information and model limiting condition information.

[0202] Optionally, in an embodiment of the present application, the data analysis identifier information is used to indicate the type of task the demanding network element is to perform.

[0203] Optionally, in an embodiment of the present application, the model limiting condition information includes at least one of time limiting condition information and area limiting condition information.

[0204] In an embodiment of the present application, the above-mentioned time restriction condition information is used to indicate the time during which the demanding network element allows model sharing, for example, from 7 p.m. to 6 a.m. the next day, and the area restriction condition information is used to indicate the area during which the demanding network element allows model sharing, for example, the location of the network element that shares the model with the first network element should be within this area restriction information, for example, "Suzhou".

[0205] An embodiment of the present application provides a model request method, in which a third network element can request a first network element to obtain a model through a model request, and since the model request includes second information, the third network element can obtain model information matching the second information from the first network element, thereby improving the success rate of model transmission.

[0206] Optionally, in an embodiment of the present application, as shown in FIG. 9 in combination with FIG. 8, before the above step 301, the model request method according to the embodiment of the present application further includes the following steps 304 and 305.

[0207] Step 304: the third network element sends a search request to the second network element.

[0208] In an embodiment of the present application, the search request is used to request a search for a target network element, and the search request includes at least one of the model training related capability information and the first information required by the third network element.

[0209] Step 305, the second network element receives the search request sent by the third network element.

[0210] In an embodiment of the present application, the model training related capability information required by the third network element includes at least one of: model platform information supported by the target network element required by the third network element; model description method information supported by the target network element required by the third network element; model accuracy information achieved by the target network element required by the third network element; and model training speed information of the target network element required by the third network element.

[0211] Optionally, in the embodiment of the present application, the third network element may send a Nnrf_NFDiscovery_Request search request to the second network element to request the second network element to feed back network element information that meets the conditions.

[0212] In an embodiment of the present application, the model platform information supported by the target network element required by the third network element is used to indicate a platform that the required target network element can understand and use, e.g., TensorFlow, PyTorch, etc.

[0213] In the embodiment of the present application, the model describing method information supported by the target network element required by the third network element is used to indicate the method technology used by the required target network element description model.

[0214] Optionally, in the embodiment of the present application, the model platform information supported by the target network element includes name or address information of one or more model platforms.

[0215] In the embodiment of the present application, the above model platform information is used to instruct the target network element to train and generate a model based on the model platform corresponding to the name or address information of the model platform.

[0216] Optionally, in an embodiment of the present application, the model accuracy information achieved by the target network element requested by the third network element is used to instruct the target network element requested to describe the degree of accuracy of discrimination or decision-making that can be achieved after training the model.

[0217] Optionally, in an embodiment of the present application, the model training speed information of the target network element required by the third network element may be the time required for the required target network element to undergo training and obtain a model of a certain accuracy.

[0218] Optionally, in an embodiment of the present application, the search request further includes data analysis identifier information of the target network element required by the third network element.

[0219] Optionally, in an embodiment of the present application, the data analysis identifier information of the target network element required by the third network element is used to indicate a task type that the required target network element can support.

[0220] Optionally, in an embodiment of the present application, the first information includes at least one of algorithm information used in the target network element training model demanded by the third network element, manufacturer information of the target network element demanded by the third network element, limiting condition information under which the target network element demanded by the third network element shares the model, identifier information of the third network element, and instruction information shared by the model.

[0221] Optionally, in an embodiment of the present application, the algorithm information used for the target network element training model required by the third network element may be an algorithm type related to AI such as machine learning that the target network element supports when training the model, such as deep learning, linear regression, gradient descent, etc., and the model supported by the target network element can be obtained through this algorithm information.

[0222] Optionally, in the embodiment of the present application, the manufacturer information of the target network element required by the third network element is used to indicate the manufacturer information to which the required target network element belongs.

[0223] Optionally, in an embodiment of the present application, the limitation condition information for the target network element to share the model required by the third network element includes at least one of time information supported by the target network element and area limitation information of the target network element.

[0224] In an embodiment of the present application, the time information supported by the target network element demanded by the third network element is used to indicate the time during which the demanding target network element will allow model sharing, and the area restriction information of the target network element demanded by the third network element is used to indicate the area during which the demanding target network element will allow model sharing.

[0225] Optionally, in an embodiment of the present application, the model-shared indication information is used to indicate that the network element required by the third network element is a network element that supports model sharing or model interoperation with other network elements.

[0226] Optionally, in an embodiment of the present application, as shown in FIG. 10 in conjunction with FIG. 9, after the above step 305, the model request method according to the embodiment of the present application further includes the following steps 306 and 307.

[0227] Step 306: the second network element sends a search response to the third network element.

[0228] In an embodiment of the present application, the search response includes targeting information.

[0229] Step 307, the third network element receives the search response sent by the second network element.

[0230] In an embodiment of the present application, the target information includes at least one of identifier information of the target network element, address information of the target network element, model platform information supported by the target network element, model description method information supported by the target network element, model accuracy information achievable by the target network element, model training speed information of the target network element, algorithm information used in the target network element training model, manufacturer information of the target network element, limiting condition information under which the target network element shares the model, and network element information of the target network element stored training model.

[0231] Optionally, in the embodiment of the present application, the identifier information of the target network element may be any one of Fully Qualified Domain Name (FQDN) information and IP address information.

[0232] In an embodiment of the present application, the model platform information supported by the target network element is used to indicate the platforms that the network element can understand and use, e.g., TensorFlow, PyTorch, etc.

[0233] Optionally, in the embodiment of the present application, the target network element may be based on the model platform training model, or the model may be obtained based on the model platform training.

[0234] In the embodiment of the present application, the model describing method information supported by the target network element is used to instruct this network element to describe the method technology used by the model.

[0235] Optionally, in the embodiment of the present application, the model description method information supported by the target network element is one of model format information and model language information.

[0236] Optionally, in the embodiment of the present application, the model language information may be ONNX (One Universal Model Description Language) language.

[0237] Optionally, in an embodiment of the present application, the target network element supports expressing the trained model in model format information or model language information.

[0238] Optionally, in an embodiment of the present application, the model accuracy information achievable by the target network element is used to instruct the target network element to describe the degree of accuracy of discrimination or decision-making that can be achieved after training the model.

[0239] In an embodiment of the present application, the model accuracy information may be the number of times the model decision result is correct divided by the total number of decisions, that is, model accuracy=number of correct results / total number.

[0240] Optionally, in the embodiment of the present application, the model accuracy information may be identified in the form of a decimal, a percentage, or high, medium, low.

[0241] Optionally, in the embodiments of the present application, the model training speed information of the target network element may be the time required for the target network element to obtain a model of a certain accuracy through training. For example, if the target network element trains one type of model and requires 2 hours to achieve 90% accuracy, this time represents the model training speed information of the target network element.

[0242] Optionally, in the embodiments of the present application, the algorithm information used in the target network element training model may be an algorithm type related to AI such as machine learning that the target network element supports when performing model training, such as deep learning, linear regression, gradient descent, etc., and this algorithm information can obtain the model supported by the target network element.

[0243] Optionally, in the embodiment of the present application, the manufacturer information of the target network element is used to indicate the manufacturer information to which this network element belongs.

[0244] In the embodiment of the present application, the network element target information with which the target network element can share a model is used to indicate the network element target information defined by the network element itself during model sharing and model transmission.

[0245] Optionally, in an embodiment of the present application, the network element target information with which the target network element can share a model includes at least one of information of one or more model inference function network elements and information of one or more model training function network elements.

[0246] Optionally, in an embodiment of the present application, the limiting condition information under which the target network element shares the model includes at least one of time limiting condition information and area limiting condition information.

[0247] In an embodiment of the present application, the above-mentioned time restriction condition information is used to indicate the time during which the target network element allows model sharing, for example, from 7 p.m. to 6 a.m. the next day, and the area restriction condition information is used to indicate the area during which the target network element allows model sharing, for example, the location of the network element that shares the model with the target network element should be within this area restriction information, for example, "Suzhou".

[0248] In an embodiment of the present application, the network element information of the first network element stored training model is used to indicate the network elements to be stored in the model on which the first network element training has been completed.

[0249] Optionally, in the embodiment of the present application, the network element information of the first network element stored training model may be the identifier information of the network element responsible for storing the model information, such as the ADRF network element's identifier information, network element type information, etc.

[0250] An embodiment of the present application provides a model request method, in which a third network element can search for a target network element in a second network element through a search request, and the search request includes model training related capability information required by the third network element, so that the second network element can quickly and accurately search for the target network element required by the third network element; and can request the target network element to obtain a model through a model request, and because the model request includes second information, the third network element can obtain model information matching the second information from the first network element, thereby improving the success rate of model transmission.

[0251] It should be explained that in the model request method according to the embodiment of the present application, the execution body may be a model request device, or a control module for executing the model request method in the model request device.

[0252] The following describes in detail the interaction process of the network element registration method and the model request method according to the embodiment of the present application through a specific embodiment.

[0253] As shown in FIG. 11, the method according to the embodiment of the present application includes the following steps 21 to 28.

[0254] Step 21: The model training network element MTLF (or NWDAF, NWDAF containing MTLF) sends a network element registration message to the NRF to register its capabilities.

[0255] In an embodiment of the present application, a MTLF network element may be registered via (Nnrf_NFManagement_NFRegister Register).

[0256] Step 22, the NRF network element stores the information of this MTLF network element.

[0257] Step 23, the NRF network element sends a registration response message.

[0258] In an embodiment of the present application, the NRF network element may acknowledge successful registration of the MTLF network element via a (Nnrf_NFManagement_NFRegister response) response message.

[0259] The steps are similar to those of the conventional method, but the differences are as follows:

[0260] In step 21, when the MTLF sends a capability registration message to the NRF, in addition to information such as its own identifier information and supported analytic IDs, it also sends information such as "supported training type information," "supported federated learning time," and "metadata information."

[0261] The information required for registration is as follows:

[0262] 1. NF type: network element type, which refers to the type of network element that is registered this time. In this case, NF type = NWDAF type, etc.

[0263] 2. NF instance ID, FQDN or IP address of NF, is the network element instance identifier information, which refers to the currently registered network element identifier information, such as its FQDN information (Fully Qualified Domain Name, a fully qualified domain name for indicating the location and connection of this network element) or IP address information (another identifier information).

[0264] 3. Supported analytic ID information, indicating the task types that this NWDAF network element can perform.

[0265] In addition, the network element registration information must further include at least one of model training capability information and training capability information.

[0266] 4. Supported model platform information, which indicates the platforms that this network element understands and can use, e.g., TensorFlow, PyTorch, etc.

[0267] 5. Supported model description method information, which refers to the method technology used by the description model, may be model format information or model language information, for example, the language used in the description model is ONNX (one general-purpose model description language).

[0268] 6. Model accuracy information, where accuracy is used to describe the degree of accuracy of discrimination or decision-making that can be achieved after training the model. The model accuracy may be calculated by dividing the number of times the model decision-making result is correct by the total number of decisions, where accuracy = number of correct results / total number. This information may be identified in the form of a decimal, a percentage, or high, medium, or low.

[0269] 7. Model training speed information, which may be the time required for a first network element to obtain a model with a certain accuracy through training. For example, if the network element trains a model and needs 2 hours to achieve 90% accuracy, it represents the model training speed of the network element.

[0270] Other Information: 8. Algorithm information used for training model, including the machine learning or other AI related algorithm types supported when training the model, such as "deep learning", "linear regression", "gradient descent", etc.

[0271] 9. Manufacturer information, which is information about the manufacturer to which this network element belongs.

[0272] 10. Version information, which is the version information of the model used by this network element.

[0273] 11. Network element target information that can share a model, which refers to the network element target information limited by this network element itself when sharing or transmitting a model.

[0274] 12. Network element information of the stored training model, including the network element responsible for storing the model information, such as the identifier information of the ADRF (Analytics Data Repository Function), network element type information, etc.

[0275] 13. Information on the limitations of conditions for sharing the model, including:

[0276] Supported time information indicates the times during which this network element can propagate the model, for example from 7 pm to 6 am the next day.

[0277] Area restriction information refers to the range information of the location of the sharing network elements that this network element can accept. For example, the locations of the network elements that share models with this network element should all be within this area restriction information, for example, "Suzhou".

[0278] 14. First indication information, the first indication information being used to indicate whether the model training network element supports module sharing or model interoperation with other network elements.

[0279] Step 24: The task consumer sends a task request to AnLF.

[0280] In step 25, the AnLF sends a request to the NRF to find suitable training network elements.

[0281] Step 26, the NRF sends a response message regarding the network element search.

[0282] In the embodiment of the present application, the response message must include: 1. query validity period; 2. identifier information of each network element; 3. address information, and so on.

[0283] 4. Model platform information of the searched network elements; 5. Model description method information; 6. Model description method information; 7. Model accuracy information; 8. Model training speed information; 9. Information about the algorithm used to train the model; 10. Manufacturer and version information, 11. Network element information of memory training model; 12. Limited information for sharing the model may further be included.

[0284] Step 27, the AnLF initiates model requests to the appropriate network elements based on the information returned from the NRF.

[0285] In step 28, AnLF uses the resulting model to generate analytical data and return it to the consumer.

[0286] In an embodiment of the present application, after using the model requested in step 27, the AnLF may generate a task report and return it to the consumer NF, and the AnLF may return the task report to the consumer via (Nnwdaf_AnalyticsInfo_Request response).

[0287] The steps are similar to those of the conventional method, but the differences are as follows:

[0288] In step 25, when the AnLF sends a request to the NRF to find suitable training network elements, the request to find network elements may include information such as analytic ID, as well as "model platform information" and "time required information". This information includes:

[0289] 1. Required information: An analytic ID that indicates the task types that the target MTLF can support.

[0290] 2. Filter info: This is task model limitation information, and is used to limit task details, such as the target that the model targets (e.g., mobility analysis for a certain UE), target task time, etc.

[0291] In this method, the AnLF may limit information about the model transfer, Training capability information, including:

[0292] 3. Supported model platform information, which indicates the platforms that this network element understands and can use, e.g., TensorFlow, PyTorch, etc.

[0293] 4. Supported model description method information, which refers to the method technology used by the description model, may be model format information or model language information.

[0294] 5. Model accuracy information, where accuracy is used to describe the degree of accuracy of discrimination or decision-making that can be achieved after training the model. The model accuracy may be calculated by dividing the number of times the model decision-making result is correct by the total number of decisions, where accuracy = number of correct results / total number. This information may be identified in the form of a decimal, a percentage, or high, medium, or low.

[0295] 6. Model training speed information, which may be the time required for a first network element to obtain a model with a certain accuracy through training, for example, if the network element trains a model and needs 2 hours to achieve 90% accuracy, it represents the model training speed of the network element.

[0296] Other information including:

[0297] 7. Algorithm information used for training model, including the machine learning or other AI related algorithm types supported for model training, such as "deep learning", "linear regression", "gradient descent", etc.

[0298] 8. Manufacturer information, which is the manufacturer information to which this network element belongs.

[0299] 9. Version information: the version information of the model used by this network element.

[0300] 10. Network element target information that can share a model, which refers to the network element target information limited by this network element itself when sharing or transmitting a model.

[0301] 11. Network element information of the stored training model, including the network element responsible for storing the model information, such as identifier information of the ADRF (Analytics Data Repository Function), network element type information, etc.

[0302] 12. Information on the limitations of conditions for sharing the model, including:

[0303] Supported time information indicates the times during which this network element can propagate the model, for example from 7 pm to 6 am the next day.

[0304] Area restriction information refers to the range information of the location of the sharing network elements that this network element can accept. For example, the locations of the network elements that share models with this network element should all be within this area restriction information, for example, "Suzhou".

[0305] Step 27, (which may be specifically) the AnLF sends a model request to the MTLF.

[0306] In an embodiment of the present application, the AnLF may send a model request via (Nnwdaf_MLModelInfo_Request), where the analytic ID, ML model Filter Information, etc. must be included.

[0307] It should be noted that, for the relevant explanations and beneficial effects in the above steps 21 to 27, please refer to the descriptions in the above embodiments, and will not be further described here.

[0308] 12 shows a possible structural schematic diagram of a network element registration device according to an embodiment of the present application. As shown in FIG. 12, a network element registration device 80 may include: a sending module 81.

[0309] Here, the sending module 81 is used for sending a network element registration request to the second network element, and the network element registration request is used for registering information of the first network element to the second network element, and the information of the first network element includes model training related capability information of the first network element, where the model training related capability information of the first network element includes at least one of model platform information supported by the first network element, model description method information supported by the first network element, manufacturer information of the first network element, network element target information with which the first network element can share models, first indication information for indicating whether the first network element supports model sharing or model interoperation with other network elements, model accuracy information achievable by the first network element, and model training speed information of the first network element.

[0310] An embodiment of the present application provides a network element registration device, which can register network element information of a first network element to a second network element through a network element registration request, thereby realizing the registration of the first network element to the second network element. Because the network element registration information of the first network element includes model training related capability information, other network elements can search for a first network element that meets the model training demand through the second network element, so as to solve the problem of how the first network element registers network element information to the second network element and can be searched by other network elements.

[0311] In one possible implementation manner, the model description method information supported by the first network element is one of model format information and model language information, where the first network element supports expressing the trained model in model format information or model language information.

[0312] In one possible implementation manner, the information of the first network element further includes at least one of algorithm information used in the first network element training model, manufacturer information of the first network element, network element target information with which the first network element can share the model, limiting condition information with which the first network element shares the model, and network element information of the first network element stored training model.

[0313] In one possible implementation manner, the network element target information with which the first network element can share a model includes at least one of information of one or more model inference function network elements and information of one or more model training function network elements.

[0314] In one possible implementation, the restriction condition information under which the first network element shares a model includes at least one of time restriction condition information and area restriction condition information, where the time restriction condition information is used to indicate a time during which the first network element allows model sharing, and the area restriction condition information is used to indicate an area during which the first network element allows model sharing.

[0315] In one possible implementation, the network element information of the first network element stored training model is used to indicate the network elements to be stored in the first network element training completed model.

[0316] The network element registration device according to the embodiment of the present application can realize each process realized by the first network element in the embodiment of the above method and achieve the same technical effects, and will not be described further here in order to avoid repetition of description.

[0317] 13 shows a possible structural schematic diagram of a network element registration device according to an embodiment of the present application. As shown in FIG. 13, a network element registration device 90 may include: a receiving module 91.

[0318] Here, the receiving module 91 is used for receiving a network element registration request sent by a first network element, which is used for registering information of the first network element to a second network element, and the information of the first network element includes model training related capability information of the first network element, where the model training related capability information of the first network element includes at least one of model platform information supported by the first network element, model description method information supported by the first network element, manufacturer information of the first network element, network element target information with which the first network element can share models, first indication information for indicating whether the first network element supports model sharing or model interoperation with other network elements, model accuracy information achievable by the first network element, and model training speed information of the first network element.

[0319] An embodiment of the present application provides a network element registration device, which can receive a network element registration request sent by a first network element, thereby registering network element information of the first network element in a second network element, and realizing the registration of the first network element in the second network element. Because the network element registration information of the first network element includes model training related capability information, other network elements can search for a first network element that meets the model training demand through the second network element, so as to solve the problem of how the first network element registers network element information in the second network element and can be searched by other network elements.

[0320] In one possible implementation, the receiving module 91 is further used for receiving a search request sent by a third network element, the search request is used for requesting the search of a target network element, and the search request includes at least one of model training related capability information required by the third network element and first information, where the model training related capability information required by the third network element includes at least one of model platform information supported by the target network element required by the third network element, model description method information supported by the target network element required by the third network element, model accuracy information achieved by the target network element required by the third network element, and model training speed information of the target network element required by the third network element. The first information includes at least one of algorithm information used in the training model of the target network element required by the third network element, manufacturer information of the target network element required by the third network element, limiting condition information of the target network element required by the third network element to share the model, identifier information of the third network element, and instruction information of the model to be shared; Here, the limiting condition information for the target network element to share the model required by the third network element includes at least one of time information supported by the target network element and area limiting information of the target network element.

[0321] In one possible implementation manner, the network element registration device 90 according to the embodiment of the present application further includes a determining module, which is used after the receiving module 91 receives the search request sent by the third network element, to determine a target network element according to the search request, and the model training related capability information of the target network element matches the model training related capability information required by the third network element.

[0322] In one possible implementation manner, the determination module specifically determines a target network element based on model platform information supported by a target network element required by a third network element, where the target network element supports the model platform information required by the third network element; or determines a target network element based on model description method information supported by a target network element required by the third network element, where the target network element supports the model description method information required by the third network element; or determines a target network element based on model accuracy information achieved by the target network element required by the third network element, where the target network element supports the model accuracy information to be achieved required by the third network element; or determines a target network element based on model training speed information of a target network element required by the third network element, where the target network element is used to support the model training speed information required by the third network element.

[0323] In one possible implementation manner, the determination module specifically determines a target network element based on algorithm information used in the training model of a target network element required by a third network element, where the target network element supports the algorithm information used in the training model required by the third network element; or determines a target network element based on manufacturer information of the target network element required by the third network element, where the target network element supports the manufacturer information required by the third network element; or determines a target network element based on limiting condition information for the target network element required by the third network element to share a model, where the target network element supports the limiting condition information for the model shared by the third network element; or determines a target network element based on identifier information of a third network element, where the network element object information with which the target network element can share a model matches the identifier information of the third network element; or determines a target network element based on instruction information shared by the model, where the target network element is used to support model sharing or model interoperation with other network elements.

[0324] In one possible implementation manner, the network element registration device 90 according to the embodiment of the present application further includes a sending module, which is used to send a search response to the third network element after the determining module determines the target network element according to the search request, and the search response includes target information, which includes at least one of the following: identifier information of the target network element, address information of the target network element, model platform information supported by the target network element, model description method information supported by the target network element, model accuracy information achievable by the target network element, model training speed information of the target network element, algorithm information used in the training model of the target network element, manufacturer information of the target network element, limited condition information of the target network element sharing the model, and network element information of the training model stored in the target network element.

[0325] The network element registration device according to the embodiment of the present application can realize each process realized by the second network element in the above method embodiment and achieve the same technical effects, and will not be described further here in order to avoid repetition of description.

[0326] 14 shows a possible structural schematic diagram of a model requesting device according to an embodiment of the present application. As shown in FIG. 14, the model requesting device 100 may include: a sending module 101 and an acquiring module 102.

[0327] Here, the sending module 101 is used for sending a model request message to a first network element, and the model request message is used for requesting acquisition of a model, where the model request message includes second information, and the second information includes at least one of model platform information, model description method information, model accuracy information, manufacturer information of a third network element, and identifier information of the third network element, and the acquiring module 102 is used by the third network element to acquire information of a model matching the second information from the first network element.

[0328] An embodiment of the present application provides a model requesting device, which can request a first network element to obtain a model through a model request, and since the model request includes second information, a third network element can obtain model information matching the second information from the first network element, thereby improving the success rate of model transmission.

[0329] In one possible implementation manner, the model description method information is one of model format information and model language information, and the model description method information is used to instruct the first network element to feed back a model to a third network element in the model format information or the model language information.

[0330] In one possible implementation, the second information further includes at least one of: algorithm information used for the training model, model version information, and model feedback time information.

[0331] In one possible implementation, the information of the model includes at least one of: file information of the model and download address information of the model file.

[0332] In one possible implementation manner, the sending module 101 is further used for sending a search request to the second network element, which is used to request a search for a target network element, and the search request includes at least one of model training related capability information required by the third network element and first information, where the model training related capability information required by the third network element includes at least one of model platform information supported by the target network element required by the third network element, model description method information supported by the target network element required by the third network element, manufacturer information of the first network element, network element target information with which the first network element can share models, first indication information for indicating whether the first network element supports model sharing or model interoperation with other network elements, model accuracy information achieved by the target network element required by the third network element, and model training speed information of the target network element required by the third network element. This first information includes at least one of algorithm information used for the target network element training model demanded by the third network element, manufacturer information of the target network element demanded by the third network element, limiting condition information under which the target network element demanded by the third network element shares the model, identifier information of the third network element, and instruction information shared by the model, where the limiting condition information under which the target network element demanded by the third network element shares the model includes at least one of time information supported by the target network element and area limiting information of the target network element.

[0333] The model sharing indication information is used to indicate that the network element required by the third network element is a network element that supports model sharing or model interoperation with other network elements.

[0334] In one possible implementation manner, the model request device 100 according to the embodiment of the present application further includes a receiving module, which is used to receive a search response sent by the second network element after the sending module 101 sends a search request to the second network element, and the search response includes target information, which includes at least one of the following: identifier information of the target network element, address information of the target network element, model platform information supported by the target network element, model description method information supported by the target network element, model accuracy information achievable by the target network element, model training speed information of the target network element, algorithm information used in the training model of the target network element, manufacturer information of the target network element, limited condition information of the target network element sharing the model, and network element information of the storage training model of the target network element.

[0335] The model request device according to the embodiment of the present application can realize each process realized by the third network element in the embodiment of the above method and achieve the same technical effects, and will not be described further here in order to avoid repetition of description.

[0336] 15 shows a possible structural schematic diagram of a model requesting device according to an embodiment of the present application. As shown in FIG. 15, the model requesting device 110 may include: a receiving module 111.

[0337] Here, the receiving module 111 is used for receiving a model request message sent by a third network element, and the model request message is used to request acquisition of a model, where the model request message includes second information, and the second information includes at least one of model platform information, model description method information, model accuracy information, manufacturer information of the third network element, and identifier information of the third network element, where the second information is used by the third network element to obtain information of a model matching the second information from the first network element.

[0338] An embodiment of the present application provides a model request device, which can receive a model request sent by a third network element, and the model request includes second information, so that the third network element can obtain model information matching the second information from the first network element, thereby improving the success rate of model transmission.

[0339] In one possible implementation manner, the model description method information is one of model format information and model language information, and the model description method information is used to instruct the first network element to feed back a model to a third network element in the model format information or the model language information.

[0340] In one possible implementation, the second information further includes at least one of: algorithm information used for the training model, model version information, and model feedback time information.

[0341] In one possible implementation, the information of the model includes at least one of: file information of the model and download address information of the model file.

[0342] The model request device according to the embodiment of the present application can realize each process realized by the first network element in the embodiment of the above method and achieve the same technical effects, and will not be described further here in order to avoid repetition of description.

[0343] Optionally, as shown in FIG. 16, an embodiment of the present application further provides a communication device 5000, which includes a processor 5001 and a memory 5002, and the memory 5002 stores a program or instruction that can run on the processor 5001. For example, when the communication device 5000 is a network element, when the program or instruction is executed by the processor 5001, it can realize each step of the embodiment of the network element registration method on the first network element side and achieve the same technical effect, or it can realize each step of the embodiment of the network element registration method on the second network element side and achieve the same technical effect, or it can realize each step of the embodiment of the network element registration method on the third network element side and achieve the same technical effect, or it can realize each step of the embodiment of the model request method on the first network element side and achieve the same technical effect. In order to avoid repetition of description, no further description will be given here.

[0344] An embodiment of the present application further provides a network element, comprising a processor and a communication interface, the communication interface being used for sending a network element registration request to a second network element, the network element registration request being used for registering information of a first network element to the second network element, the information of the first network element including model training related capability information of the first network element, where the model training related capability information of the first network element includes at least one of: model platform information supported by the first network element; model description method information supported by the first network element; manufacturer information of the first network element; network element target information with which the first network element can share models; first indication information for indicating whether the first network element supports model sharing or model interoperation with other network elements; model accuracy information achievable by the first network element; and model training speed information of the first network element. This embodiment of the network element corresponds to the embodiment of the method on the first network element side, and each implementation process and realization manner of the embodiment of the method are all applicable to this embodiment of the network element, and can achieve the same technical effects.

[0345] An embodiment of the present application further provides a network element, comprising a processor and a communication interface, the communication interface being used for receiving a network element registration request sent by a first network element, the network element registration request being used for registering information of the first network element to a second network element, the information of the first network element including model training related capability information of the first network element, where the model training related capability information of the first network element includes at least one of: model platform information supported by the first network element; model description method information supported by the first network element; manufacturer information of the first network element; network element target information with which the first network element can share models; first indication information for indicating whether the first network element supports model sharing or model interoperation with other network elements; model accuracy information achievable by the first network element; and model training speed information of the first network element. This embodiment of the network element corresponds to the embodiment of the method on the second network element side, and each implementation process and realization manner of the embodiment of the method are all applicable to this embodiment of the network element, and can achieve the same technical effects.

[0346] An embodiment of the present application further provides a network element, including a processor and a communication interface, the communication interface is used to send a model request message to a first network element, the model request message is used to request acquisition of a model, where the model request message includes second information, the second information includes at least one of model platform information, model description method information, model accuracy information, manufacturer information of a third network element, and identifier information of the third network element, and the third network element obtains information of a model matching the second information from the first network element. This embodiment of the network element corresponds to the embodiment of the method on the third network element, and each implementation process and realization manner of the embodiment of the method are all applicable to this embodiment of the network element, and can achieve the same technical effects.

[0347] An embodiment of the present application further provides a network element, including a processor and a communication interface, the communication interface is used to receive a model request message sent by a third network element, the model request message is used to request acquisition of a model, where the model request message includes second information, the second information includes at least one of model platform information, model description method information, model accuracy information, manufacturer information of the third network element, and identifier information of the third network element, where the second information is used by the third network element to acquire information of a model matching the second information from the first network element. This embodiment of the network element corresponds to the embodiment of the method on the first network element side of the model request method, and each implementation process and realization manner of the above method embodiment can be applied to this embodiment of the network element, and the same technical effects can be achieved.

[0348] Specifically, FIG. 17 is a hardware structural schematic diagram for implementing a network element of an embodiment of the present application, where the network element is a first network element, or the network element is a second network element, or the network element is a third network element.

[0349] 17, the network element 1200 includes a processor 1201, a network interface 1202, and a memory 1203. Here, the network interface 1202 is, for example, a common public wireless interface.

[0350] Specifically, the network element 1200 of the embodiment of the present application further includes instructions or programs stored in the memory 1203 and capable of running on the processor 1201, and the processor 1201 can call the instructions or programs in the memory 1203, execute the methods performed by each of the above modules, and achieve the same technical effects, which will not be described further here in order to avoid repetition.

[0351] The network element according to the embodiments of the present application can realize each process realized by the first network element, the second network element and the third network element in the embodiments of the above method, and can achieve the same technical effects, and will not be described further here in order to avoid repetition.

[0352] An embodiment of the present application further provides a readable storage medium, in which a program or instruction is stored, which, when executed by a processor, can realize each process of the above method embodiment and achieve the same technical effect, and will not be described further here in order to avoid repetition of description.

[0353] Here, the processor is the processor in the communication device according to the above embodiment. The readable storage medium includes a computer readable storage medium, such as a computer read only memory (ROM), a random access memory (RAM), a magnetic disk, or an optical disk.

[0354] An embodiment of the present application further provides a chip, the chip including a processor and a communication interface, the communication interface is coupled to the processor, the processor is used to run a program or instruction and realize each process of the above method embodiment, and can achieve the same technical effect, and will not be described further here in order to avoid repetition of description.

[0355] It should be understood that the chips referred to in the embodiments of the present application may be referred to as system level chips, system chips, chip systems, or systems on chips.

[0356] The embodiments of the present application further provide a computer program / program product, which is stored in a storage medium, and which can be executed by at least one processor to realize each process of the above method embodiments and achieve the same technical effects, which will not be described further here in order to avoid repetition of description.

[0357] An embodiment of the present application further provides a communication system, which includes a first network element, a second network element and a third network element, wherein the first network element may be used to perform steps of the network element registration method described above, and may be used to perform steps of the model request method described above, the second network element may be used to perform steps of the network element registration method described above, and the third network element may be used to perform steps of the model request method described above.

[0358] It should be explained that in this specification, the terms "comprise", "include", or any other variants thereof are intended to cover the non-exclusive "comprise", whereby a process, method, article, or apparatus that includes a set of elements includes not only those elements, but also other elements not expressly listed or inherent to such process, method, article, or apparatus. In the absence of further limitations, an element limited by the phrase "comprises a" does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes this element. It should be pointed out that the scope of the method and apparatus in the embodiments of this application is not limited to performing functions in the order shown or discussed, but may include performing functions in an essentially simultaneous manner or in reverse order based on the functions involved, for example, the described method can be performed in a different order than described, and various steps can be added, omitted, or combined. Also, features described with reference to some examples can be combined in other examples.

[0359] From the above description of the embodiments, it is clear to those skilled in the art that the methods of the above embodiments can be realized in the form of software and a necessary general-purpose hardware platform. Of course, they can also be realized in hardware, but in many cases the former is a more preferred embodiment. In light of this understanding, the technical proposal of the present application may be substantially or the part that contributes to the prior art may be embodied in the form of a computer software product, which is stored in a storage medium (e.g., ROM / RAM, magnetic disk, optical disk) and includes some instructions for causing a terminal (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to execute the methods described in each embodiment of the present application.

[0360] The above describes the embodiments of the present application in conjunction with the drawings, but the present application is not limited to the above specific embodiments. The above specific embodiments are merely illustrative and not limiting. Those skilled in the art can take the suggestions of this application and make many forms without departing from the spirit and scope of the claims of this application, all of which belong to the protection scope of this application.

[0361] CROSS-REFERENCE TO RELATED APPLICATIONS This application claims priority to Chinese Patent Application No. 202210317248.5 filed in China on March 28, 2022, and claims priority to Chinese Patent Application No. 202210370099.9 filed in China on April 8, 2022, the entire contents of which are incorporated herein by reference.

Claims

1. A network element registration method, comprising: The method includes: a first network element sending a network element registration request to a second network element, the network element registration request being used to register information of the first network element to the second network element, the information of the first network element including model training related capability information of the first network element; Wherein, the model training related capability information of the first network element is: Model platform information supported by the first network element; and Model description method information supported by the first network element; Manufacturer information of the first network element; Network element target information with which the first network element can share a model; First indication information for indicating whether the first network element supports model sharing or model interoperation with other network elements; model accuracy information achievable by the first network element; and model training speed information of the first network element.

2. The model description method information supported by the first network element is one of model format information and model language information; The method of claim 1 , wherein the first network element supports expressing a trained model in the model format information or the model language information.

3. The information of the first network element is: Algorithm information used in the first network element training model; Information on limiting conditions under which the first network element shares a model; and network element information of the first network element storing training model.

4. 2. The method of claim 1, wherein the network element target information with which the first network element can share a model includes at least one of information of one or more model inference function network elements and information of one or more model training function network elements.

5. The limitation condition information under which the first network element shares a model includes at least one of time limitation condition information and area limitation condition information; 4. The method of claim 3, wherein the time restriction condition information is used to indicate a time during which the first network element will allow model sharing, and the area restriction condition information is used to indicate an area during which the first network element will allow model sharing.

6. 4. The method of claim 3, wherein the network element information of the first network element stored training model is used to indicate the network elements to be stored in the first network element training completed model.

7. A network element registration method, comprising: A second network element receives a network element registration request sent by a first network element, the network element registration request is used to register information of the first network element to the second network element, the information of the first network element includes model training related capability information of the first network element; Wherein, the model training related capability information of the first network element is: Model platform information supported by the first network element; and Model description method information supported by the first network element; Manufacturer information of the first network element; Network element target information with which the first network element can share a model; First indication information for indicating whether the first network element supports model sharing or model interoperation with other network elements; model accuracy information achievable by the first network element; and model training speed information of the first network element.

8. After the second network element receives a network element registration request sent by the first network element, the method includes: The second network element further includes receiving a search request sent by a third network element, the search request being used to request a search for a target network element, the search request including at least one of model training related capability information and first information required by the third network element; Wherein, the model training related capability information required by the third network element includes at least one of model platform information supported by the target network element required by the third network element, model description method information supported by the target network element required by the third network element, model accuracy information achieved by the target network element required by the third network element, and model training speed information of the target network element required by the third network element; The first information includes at least one of algorithm information used in the target network element training model required by the third network element, manufacturer information of the target network element required by the third network element, limiting condition information for the target network element required by the third network element to share a model, identifier information of the third network element, and instruction information shared by the model; The method of claim 7, wherein the limitation condition information for which the target network element required by the third network element shares a model includes at least one of time information supported by the target network element and area limitation information of the target network element.

9. After the second network element receives a search request sent by a third network element, the method includes:

10. The method of claim 8, further comprising: the second network element determining the target network element based on the search request, wherein model training related capability information of the target network element matches model training related capability information required by the third network element.

10. determining, by the second network element, the target network element based on the search request; The second network element determines the target network element based on model platform information supported by the target network element required by the third network element, where the target network element supports the model platform information required by the third network element; The second network element determines the target network element based on model description method information supported by the target network element required by the third network element, and the target network element supports the model description method information required by the third network element; The second network element determines the target network element according to model accuracy information that the target network element requires to achieve, and the target network element supports the model accuracy information that the third network element requires to achieve; 10. The method of claim 9, further comprising at least one of: the second network element determining the target network element based on model training rate information of the target network element required by the third network element, the target network element supporting the model training rate information required by the third network element.

11. After the second network element receives a search request sent by a third network element, the method includes: The method of claim 7 , further comprising the second network element determining the target network element based on first information in the search request.

12. determining, by the second network element, the target network element based on first information in the search request; The second network element determines the target network element based on algorithm information used in the target network element training model required by the third network element, and the target network element supports the algorithm information used in the training model required by the third network element; The second network element determines the target network element according to manufacturer information of the target network element required by the third network element, and the target network element supports the manufacturer information required by the third network element; The second network element determines the target network element based on the limiting condition information of the target network element that is required by the third network element to share a model, and the target network element supports the limiting condition information of the model that is required by the third network element; The second network element determines the target network element according to the identifier information of the third network element, and the network element target information with which the target network element can share a model matches the identifier information of the third network element; 12. The method of claim 11, further comprising: the second network element determining the target network element based on instruction information shared by the model, the target network element supporting model sharing or model interoperation with other network elements.

13. After the second network element determines the target network element based on the search request, the method includes: The method according to any one of claims 9 to 12, further comprising the second network element sending a search response to the third network element, the search response including target information, the target information including at least one of identifier information of the target network element, address information of the target network element, model platform information supported by the target network element, model description method information supported by the target network element, model accuracy information achievable by the target network element, model training speed information of the target network element, algorithm information used in the training model of the target network element, manufacturer information of the target network element, limiting condition information for the target network element to share the model, and network element information of the stored training model of the target network element.

14. 1. A model request method, comprising: The third network element sends a model request message to the first network element, the model request message is used to request acquisition of a model, and the model request message includes second information, the second information including at least one of model platform information, model description method information, model accuracy information, manufacturer information of the third network element, and identifier information of the third network element; A model request method, in which the third network element obtains information of a model matching the second information from the first network element.

15. The method of claim 14, wherein the model description method information is one of model format information and model language information, and the model description method information is used to instruct the first network element to feed back the model to the third network element in the model format information or the model language information.

16. The second information is Algorithm information used to train the model; and Version information of the model; and feedback time information of the model.

17. The method of claim 14 , wherein the information of the model includes at least one of file information of the model and download address information of the model file.

18. Before the third network element sends a model request message to the first network element, the method further comprises: The third network element further includes sending a search request to the second network element, the search request being used to request a search for a target network element, the search request including at least one of model training related capability information and first information required by the third network element; Wherein, the model training related capability information required by the third network element includes at least one of model platform information supported by the target network element required by the third network element, model description method information supported by the target network element required by the third network element, model accuracy information achieved by the target network element required by the third network element, and model training speed information of the target network element required by the third network element; The first information includes at least one of algorithm information used in the training model of the target network element required by the third network element, manufacturer information of the target network element required by the third network element, and limiting condition information for the target network element required by the third network element to share a model; The method of claim 14, wherein the limitation condition information for which the target network element requests the third network element to share the model includes at least one of time information supported by the target network element and area limitation information of the target network element.

19. After the third network element sends a search request to the second network element, the method includes: The method of claim 18, further comprising the third network element receiving a search response sent by the second network element, the search response including target information, the target information including at least one of identifier information of the target network element, address information of the target network element, model platform information supported by the target network element, model description method information supported by the target network element, model accuracy information achievable by the target network element, model training speed information of the target network element, algorithm information used in the target network element training model, manufacturer information of the target network element, limiting condition information for the target network element to share a model, and network element information of the target network element stored training model.

20. 1. A model request method, comprising: The method includes: a first network element receiving a model request message sent by a third network element, the model request message being used to request acquisition of a model, where the model request message includes second information, the second information including at least one of model platform information, model description method information, model accuracy information, manufacturer information of the third network element, and identifier information of the third network element; Here, the second information is used by the third network element to obtain information of a model matching the second information from the first network element, the model request method.

21. The method of claim 20, wherein the model description method information is one of model format information and model language information, and the model description method information is used to instruct the first network element to feed back the model to the third network element in the model format information or the model language information.

22. The second information is Algorithm information used to train the model; and Version information of the model; and feedback time information of the model.

23. The method according to any one of claims 20 to 22, wherein the information of the model includes at least one of file information of the model and download address information of the model file.

24. 24. The method according to claim 20, further comprising: the first network element obtaining, based on the second information, information of a model matching the second information.

25. The first network element obtaining information of a model matching the second information based on the second information, The first network element acquires a target model based on model platform information included in the second information, and the target model is generated based on a model platform corresponding to the model platform information; The method of claim 24, comprising at least one of: the first network element obtaining a target model based on model description method information included in the second information, the target model being represented based on model language information or model format information corresponding to the model description method information.

26. The method comprises: The first network element determines whether the first network element allows model sharing or model interoperation with the third network element based on manufacturer information of the third network element included in the second information; The method according to any one of claims 20 to 25, further comprising at least one of: determining, by the first network element, based on identifier information of the third network element included in the second information, that the first network element allows model sharing or model interoperation with the third network element.

27. A network element registration device for use in a first network element, comprising: a sending module; The sending module is used for sending a network element registration request to a second network element, the network element registration request is used for registering information of the first network element to the second network element, the information of the first network element includes model training related capability information of the first network element; Wherein, the model training related capability information of the first network element is: Model platform information supported by the first network element; and Model description method information supported by the first network element; Manufacturer information of the first network element; Network element target information with which the first network element can share a model; First indication information for indicating whether the first network element supports model sharing or model interoperation with other network elements; model accuracy information achievable by the first network element; and model training speed information of the first network element.

28. A network element registration device for use in a second network element, comprising: a receiving module; The receiving module is used for receiving a network element registration request sent by a first network element, the network element registration request is used for registering information of the first network element to the second network element, the information of the first network element includes model training related capability information of the first network element; Wherein, the model training related capability information of the first network element is: Model platform information supported by the first network element; and Model description method information supported by the first network element; Manufacturer information of the first network element; Network element target information with which the first network element can share a model; First indication information for indicating whether the first network element supports model sharing or model interoperation with other network elements; model accuracy information achievable by the first network element; and model training speed information of the first network element.

29. A model request device for use in a third network element, the model request device including a sending module and an obtaining module; The sending module is used to send a model request message to a first network element, and the model request message is used to request acquisition of a model, where the model request message includes second information, the second information including at least one of model platform information, model description method information, model accuracy information, manufacturer information of the third network element, and identifier information of the third network element; A model requesting device, wherein the obtaining module is used for obtaining model information matching the second information from the first network element.

30. A model request device for use in a first network element, the model request device including a receiving module, The receiving module is used to receive a model request message sent by a third network element, and the model request message is used to request acquisition of a model, where the model request message includes second information, the second information including at least one of model platform information, model description method information, model accuracy information, manufacturer information of the third network element, and identifier information of the third network element; Here, the second information is used by the third network element to obtain information of a model matching the second information from the first network element.

31. A network element comprising a processor and a memory, the memory storing a program or instructions operable on the processor, the program or instructions, when executed by the processor, realizing the steps of the network element registration method according to any one of claims 1 to 6.

32. A network element comprising a processor and a memory, the memory storing a program or instructions operable on the processor, the program or instructions, when executed by the processor, realizing the steps of the network element registration method according to any one of claims 7 to 13.

33. 20. A network element comprising a processor and a memory, the memory storing a program or instructions operable to run on the processor, the program or instructions, when executed by the processor, implementing the steps of the model request method of any one of claims 14 to 19.

34. 27. A network element comprising a processor and a memory, the memory storing a program or instructions operable to run on the processor, the program or instructions, when executed by the processor, implementing the steps of the model request method of any one of claims 20 to 26.

35. A communication system, the communication system including: a network element registration device according to claim 27; a network element registration device according to claim 28; a model request device according to claim 29; and a model request device according to claim 30; or The communication system includes a network element according to claims 31 to 34.

36. A readable storage medium having a program or instructions stored therein, the program or instructions being executed by a processor to implement steps of a network element registration method as set forth in any one of claims 1 to 6, or to implement steps of a network element registration method as set forth in any one of claims 7 to 13, or to implement steps of a model request method as set forth in any one of claims 14 to 19, or to implement steps of a model request method as set forth in any one of claims 20 to 26.

Images