Communication data processing method and apparatus, and communication device

Abstract

The present application belongs to the technical field of communications. Disclosed are a communication data processing method and apparatus, and a communication device. The communication data processing method in the embodiments of the present application comprises: a first communication device receiving a first message from a second communication device, wherein the first message is used for requesting the acquisition of a data analysis result corresponding to a data analysis task; the first communication device acquiring information of a third communication device having a capability related to vertical federated learning (VFL); and the first communication device sending a second message to the third communication device on the basis of the information of the third communication device, wherein the second message is used for requesting the performance of VFL inference on the data analysis task, or the first communication device sending a first feedback message to the second communication device, wherein the first feedback message is used for instructing the second communication device to request the third communication device to perform VFL inference.

Description

Communication data processing method, device and communication equipment

[0001] This application claims priority to the Chinese patent application filed with the State Intellectual Property Office on April 2, 2024, with application number 202410395997.9 and application name “Communication Data Processing Method, Device and Communication Equipment”, the entire contents of which are incorporated by reference into this application. Technical Field

[0002] The present application belongs to the field of communication technology, and specifically relates to a communication data processing method, apparatus and communication equipment. Background Art

[0003] In existing communication networks, models can complete a single calculation or exist independently. However, for Vertical Federated Learning (VFL) models, a single calculation requires the participation of multiple communication devices. Therefore, a communication device acting as a server or other VFL participant may not be able to send all of its own models to other communication devices for use. In this case, the method by which the Analytic Logic Function (AnLF) communication device obtains the model based on the data analysis request of the Network Function (NF) consumer and performs model inference cannot achieve VFL model inference. Therefore, how to achieve VFL model inference is an urgent problem that needs to be solved. Summary of the Invention

[0004] The embodiments of the present application provide a communication data processing method, apparatus, and communication device, which can solve the problem of how to implement reasoning of a VFL model.

[0005] In a first aspect, a communication data processing method is provided, the method comprising:

[0006] The first communication device receives a first message from the second communication device, where the first message is used to request obtaining a data analysis result corresponding to the data analysis task;

[0007] The first communication device obtains information of a third communication device having a capability related to vertical federated learning (VFL);

[0008] The first communication device sends a second message to the third communication device based on the information of the third communication device, where the second message is used to request VFL reasoning for the data analysis task, or

[0009] The first communication device sends a first feedback message to the second communication device, where the first feedback message is used to instruct the second communication device to request the third communication device to perform VFL inference.

[0010] In a second aspect, a communication data processing method is provided, the method comprising:

[0011] The second communication device sends a first message to the first communication device, where the first message is used to request obtaining a data analysis result corresponding to the data analysis task;

[0012] The second communication device receives a fifth feedback message from the first communication device, where the fifth feedback message is used to indicate a result of performing VFL reasoning on the data analysis task, or

[0013] The second communication device receives a first feedback message from the first communication device, and sends a sixth message to the third communication device based on the first feedback message, wherein the sixth message is used to request VFL reasoning for the data analysis task, and the third communication device has VFL-related capabilities.

[0014] In a third aspect, a communication data processing method is provided, the method comprising:

[0015] The third communication device receives a second message from the first communication device or a sixth message from the second communication device, where the second message or the sixth message is used to request VFL reasoning for the data analysis task;

[0016] The third communication device determines the data analysis result corresponding to the data analysis task;

[0017] The third communication device sends the data analysis result to the first communication device or the second communication device.

[0018] In a fourth aspect, a communication data processing apparatus is provided, applied to a first communication device, the apparatus comprising:

[0019] A first message receiving module is configured to receive a first message from a second communication device, wherein the first message is used to request acquisition of a data analysis result corresponding to a data analysis task;

[0020] A device information acquisition module, configured to acquire information of a third communication device having a capability related to vertical federated learning (VFL);

[0021] a second message sending module, configured to send a second message to the third communication device based on the information of the third communication device, wherein the second message is used to request VFL reasoning for the data analysis task, or

[0022] The first feedback message sending module is configured to send a first feedback message to the second communication device, where the first feedback message is used to instruct the second communication device to request the third communication device to perform VFL inference.

[0023] In a fifth aspect, a communication data processing apparatus is provided, applied to a second communication device, the apparatus comprising:

[0024] A first message sending module, configured to send a first message to a first communication device, wherein the first message is used to request acquisition of a data analysis result corresponding to a data analysis task;

[0025] a fifth feedback message receiving module, configured to receive a fifth feedback message from the first communication device, wherein the fifth feedback message is used to indicate a result of performing VFL reasoning on the data analysis task, or

[0026] A first feedback message receiving module is used to receive a first feedback message from the first communication device, and send a sixth message to a third communication device based on the first feedback message, wherein the sixth message is used to request VFL reasoning for the data analysis task, and the third communication device has VFL-related capabilities.

[0027] In a sixth aspect, a communication data processing apparatus is provided, applied to a third communication device, the apparatus comprising:

[0028] a message receiving module, configured to receive a second message from the first communication device or a sixth message from the second communication device, wherein the second message or the sixth message is used to request VFL reasoning for the data analysis task;

[0029] A data analysis module, configured to determine the data analysis results corresponding to the data analysis task;

[0030] A result feedback module is used to send the data analysis result to the first communication device or the second communication device.

[0031] In the seventh aspect, a communication device is provided, which includes a processor and a memory, wherein the memory stores a program or instruction that can be run on the processor, and when the program or instruction is executed by the processor, it implements the steps of the method described in the first aspect, or implements the steps of the method described in the second aspect, or implements the steps of the method described in the third aspect.

[0032] In an eighth aspect, a communication device is provided, comprising a processor and a communication interface, wherein the communication interface and the processor are coupled, and the processor is used to run programs or instructions to implement the steps of the method described in the first aspect, the second aspect or the third aspect.

[0033] In the ninth aspect, a readable storage medium is provided, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented, or the steps of the method described in the second aspect are implemented, or the steps of the method described in the second aspect are implemented.

[0034] In the tenth aspect, a wireless communication system is provided, comprising: a first communication device, a second communication device and a third communication device, wherein the first communication device can be used to perform the steps of the method described in the first aspect, the second communication device can be used to perform the steps of the method described in the second aspect, and the third communication device can be used to perform the steps of the method described in the third aspect.

[0035] In the eleventh aspect, a chip is provided, which includes a processor and a communication interface, the communication interface and the processor are coupled, and the processor is used to run programs or instructions to implement the method as described in the first aspect, or the method as described in the second aspect, or the method as described in the third aspect.

[0036] In the twelfth aspect, a computer program / program product is provided, which is stored in a storage medium and is executed by at least one processor to implement the steps of the method described in the first aspect, the second aspect or the third aspect.

[0037] In an embodiment of the present application, after receiving a first message from a second communication device, the first communication device obtains information of a third communication device having VFL-related capabilities. Based on the information of the third communication device, the first communication device sends a second message to the third communication device to request the third communication device to perform VFL reasoning on the data analysis task. Alternatively, the first communication device sends a first feedback message to the second communication device, and the first feedback message is used to instruct the second communication device to request the third communication device to perform VFL reasoning. In this way, the first communication device does not need to obtain the VFL model, but instead requests the third communication device to perform VFL reasoning, or the second communication device requests the third communication device to perform VFL reasoning, thereby realizing reasoning of the VFL model. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] FIG1 is a block diagram of a wireless communication system to which embodiments of the present application may be applied;

[0039] FIG2 is a schematic diagram of data and feature distribution of horizontal federated learning in an embodiment of the present application;

[0040] FIG3 is a schematic diagram of data and feature distribution of vertical federated learning in an embodiment of the present application;

[0041] FIG4 is a schematic diagram of the implementation of VFL reasoning in an embodiment of the present application;

[0042] FIG5 is a schematic diagram of a neural network in an embodiment of the present application;

[0043] FIG6 is a schematic diagram of neurons in a neural network according to an embodiment of the present application;

[0044] FIG7 is a flow chart of a communication data processing method provided in an embodiment of the present application;

[0045] FIG8 is a flowchart of another communication data processing method provided in an embodiment of the present application;

[0046] FIG9 is a flowchart of another communication data processing method provided in an embodiment of the present application;

[0047] FIG10 is an interactive flow chart of a communication data processing method provided in an embodiment of the present application;

[0048] FIG11 is an interactive flow chart of another communication data processing method provided in an embodiment of the present application;

[0049] FIG12 is an interactive flow chart of another communication data processing method provided in an embodiment of the present application;

[0050] FIG13 is an interactive flow chart of another communication data processing method provided in an embodiment of the present application;

[0051] FIG14 is an interactive flow chart of another communication data processing method provided in an embodiment of the present application;

[0052] FIG15 is a structural block diagram of a communication data processing device provided in an embodiment of the present application;

[0053] FIG16 is a structural block diagram of another communication data processing device provided in an embodiment of the present application;

[0054] FIG17 is a structural block diagram of another communication data processing device provided in an embodiment of the present application;

[0055] FIG18 is a structural block diagram of a communication device according to an embodiment of the present application;

[0056] FIG19 is a structural block diagram of a network-side device in an embodiment of the present application;

[0057] Figure 20 is a structural block diagram of another network-side device in an embodiment of the present application. DETAILED DESCRIPTION

[0058] The following will be combined with the accompanying drawings in the embodiments of this application to clearly describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by ordinary technicians in this field are within the scope of protection of this application.

[0059] The terms "first", "second", etc. in this application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the terms used in this way are interchangeable where appropriate, so that the embodiments of the present application can be implemented in an order other than those illustrated or described herein, and the objects distinguished by "first" and "second" are generally of the same type, and do not limit the number of objects, for example, the first object can be one or more. In addition, "or" in this application represents at least one of the connected objects. For example, "A or B" covers three options, namely, Option 1: including A but not including B; Option 2: including B but not including A; Option 3: including both A and B. The character " / " generally indicates that the objects associated before and after are in an "or" relationship.

[0060] The term "indication" in this application can be either a direct indication (or explicit indication) or an indirect indication (or implicit indication). A direct indication can be understood as the sender explicitly informing the receiver of specific information, the operation to be performed, or the requested result, etc. in the instruction sent; an indirect indication can be understood as the receiver determining the corresponding information based on the instruction sent by the sender, or making a judgment and determining the operation to be performed or the requested result, etc. based on the judgment result.

[0061] It is worth noting that the technology described in the embodiments of the present application is not limited to the Long Term Evolution (LTE) / LTE-Advanced (LTE-A) system, but can also be used in 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) or other systems. The terms "system" and "network" in the embodiments of the present application are often used interchangeably, and the technology described can be used for the systems and radio technologies mentioned above, as well as for other systems and radio technologies. The following description describes a New Radio (NR) system for illustrative purposes, and NR terminology is used in most of the following description, but these technologies can also be applied to systems other than NR systems, such as 6th generation (6G) systems. th Generation, 6G) communication system.

[0062] FIG1 is a block diagram of a wireless communication system applicable to an embodiment of the present application. The wireless communication system includes a terminal 11 and a network-side device 12. The terminal 11 may be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop computer (Laptop Computer), a notebook computer, a personal digital assistant (PDA), a handheld computer, a netbook, an ultra-mobile personal computer (UMPC), a mobile internet device (MID), an augmented reality (AR), a virtual reality (VR) device, a robot, a wearable device (Wearable Device), an aircraft (Flight Vehicle), a vehicle-mounted device (VUE), a ship-mounted device, a pedestrian user equipment (PUE), a smart home (home appliances with wireless communication capabilities, such as refrigerators, televisions, washing machines, or furniture), a game console, a personal computer (PC), an ATM, or a self-service machine, or other terminal-side devices. Wearable devices include: smart watches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart bracelets, smart rings, smart necklaces, smart anklets, smart anklets, etc.), smart wristbands, smart clothing, etc. Among them, the vehicle-mounted device can also be called a vehicle-mounted terminal, a vehicle-mounted controller, a vehicle-mounted module, a vehicle-mounted component, a vehicle-mounted chip or a vehicle-mounted unit, etc. It should be noted that the specific type of the terminal 11 is not limited in the embodiment of the present application. The network side device 12 may include an access network device or a core network device, wherein the access network device may also be called a radio access network (Radio Access Network, RAN) device, a radio access network function or a radio access network unit. The access network device may include a base station, a wireless local area network (WLAN) access point (AS) or a wireless fidelity (WiFi) node, etc.Among them, the base station can be referred to as Node B (NB), Evolved Node B (eNB), the next generation Node B (gNB), New Radio Node B (NR Node B), access point, Relay Base Station (RBS), Serving Base Station (SBS), Base Transceiver Station (BTS), radio base station, radio transceiver, Basic Service Set (BSS), Extended Service Set (ESS), Home Node B (HNB), Home evolved Node B (home evolved Node B), Transmission Reception Point (TRP) or other appropriate terms in the relevant field. As long as the same technical effect is achieved, the base station is not limited to specific technical vocabulary. It should be noted that in the embodiment of the present application, only the base station in the NR system is used as an example for introduction, and the specific type of the base station is not limited.

[0063] The core network equipment may include but is not limited to at least one of the following: core network node, core network function, mobility management entity (MME), access mobility management function (AMF), session management function (SMF), user plane function (UPF), policy control function (PCF), policy and charging rules function unit (PCRF), edge application service discovery function (EASDF), unified data management (UDM), unified data repository (UDR), home user server (HSS), centralized network configuration (CNC), network storage function (NRF), network exposure function (NEF), local NEF (L-NEF), binding support function (BSF), application function ( Function, AF), etc. It should be noted that in the embodiments of the present application, only the core network device in the NR system is introduced as an example, and the specific type of the core network device is not limited.But not limited to at least one of the following: 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 (L-NEF), Binding Support Function (BSF), Application Function (AF), etc. It should be noted that in the embodiments of this application, only the core network device in the NR system is introduced as an example, and the specific type of the core network device is not limited.

[0064] FIG2 is a schematic diagram of the data and feature distribution of horizontal federated learning in an embodiment of the present application, and FIG3 is a schematic diagram of the data and feature distribution of vertical federated learning in an embodiment of the present application. As shown in FIG2, for horizontal federated learning, different participants have a large overlap of features (horizontally), but the overlap of data samples is not high (vertically). As shown in FIG3, unlike horizontal federated learning, for vertical federated learning, the overlap of samples is high (vertically), but the overlap of features is less (horizontally). The essence of vertical federated learning is the union of features, which is suitable for scenarios with a lot of user overlap and little feature overlap, such as different services of the core network (CN) domain and RAN domain in a communication network serving the same user (such as UE, i.e., the same sample), such as Mobility Management (MM) service and Session Management (SM) service, where the two services have the same user but different features. By combining the different data features of the common samples of the participants, vertical federated learning increases the feature dimension of the training samples and obtains a better model. One of the characteristics of VFL is the scenario where users overlap and features differ. Therefore, the two or more parties performing VFL training / inference need to have overlapping users, i.e., the same sample data.

[0065] Figure 4 is a schematic diagram of the implementation of VFL reasoning in an embodiment of the present application. As shown in Figure 4, VFL reasoning is performed by each training entity (VFL participant) based on the input data and model output results collected by itself. Communication equipment such as the VFL server (Server) aggregates the output results of each training entity to generate the final output result and complete the model reasoning.

[0066] Artificial Intelligence (AI) is currently being widely used in various fields. AI models can be implemented using a variety of algorithms, such as neural networks, decision trees, support vector machines, and Bayesian classifiers. This application uses neural networks as an example, but does not limit the specific types of AI models.

[0067] Figure 5 is a schematic diagram of a neural network in an embodiment of the present application. As shown in Figure 5, X1, X2, ..., Xn, etc. are input values, and Y is the output result. The circles represent neurons, which are where calculations are performed, and the results are passed on to the next layer. The input layer, hidden layer, and output layer composed of these numerous neurons are a neural network. The number of hidden layers and the number of neurons in each layer are the "network structure" of the neural network.

[0068] A neural network is composed of neurons. The schematic diagram of a neuron is shown in Figure 6, where a1, a2, ... aK (i.e., X1, X2, ..., Xn shown in Figure 5) are inputs, w is the weight (multiplicative coefficient), b is the bias (also known as the additive coefficient), σ(.) is the activation function, and z is the output value. The corresponding operation process can be expressed as: z = a1w1 + ... + a k w k +…+a K w K +b. Common activation functions include Sigmoid, tanh, and Rectified Linear Unit (ReLU). The parameters of each neuron and the algorithm used together constitute the "parameter information" of the entire network and are a very important part of the AI ​​model file.

[0069] In actual use, an AI model refers to a file containing elements such as network structure and parameter information. The trained AI model can be directly reused by its framework platform without the need for repeated construction or learning, and can directly perform intelligent functions such as judgment and recognition.

[0070] The Network Data Analytics Function (NWDAF), a key component of 5G networks, aims to simplify the generation and use of core network data, generating insights and taking actions to enhance the end-user experience. The NWDAF can include the Model Training Logical Function (MTLF) and the Analytics Logical Function (AnLF). The MTLF is used to generate and train models, while the AnLF performs inference to generate analytical results.

[0071] The communication data processing method provided in the embodiments of the present application is described in detail below through some embodiments and their application scenarios in conjunction with the accompanying drawings.

[0072] FIG7 is a flowchart of a communication data processing method provided in an embodiment of the present application. The communication data processing method is executed by a first communication device, which is a network element device with reasoning capabilities in a communication network, such as an intelligent network element device such as AnLF, NWDAF, or Location Management Function (LMF). As shown in FIG7 , the communication data processing method includes the following steps:

[0073] Step 701: A first communication device receives a first message from a second communication device, where the first message is used to request a data analysis result corresponding to a data analysis task.

[0074] The first communication device and the second communication device are network-side devices, such as the core network device shown in Figure 1. The second communication device can be an NF consumer, a communication device that requests data analysis results, and can be any communication device that can interact with the NRF or NWDAF, such as the PCF or LMF.

[0075] When the second communication device needs to obtain the data analysis result corresponding to the data analysis task, the second communication device may send a first message to the first communication device to request the first communication device to obtain the data analysis result corresponding to the data analysis task.

[0076] Optionally, the first message includes at least one of the following:

[0077] Relevant information about the data analysis task;

[0078] Identification information of the inference object;

[0079] The required supported inference accuracy;

[0080] The required inference latency.

[0081] The relevant information of the data analysis task includes at least one of the following: identification information of the data analysis task; time information of the data analysis task; region information of the data analysis task; and network slice information of the data analysis task. The identification information of the data analysis task is used to indicate the specific task category. The time information, region information, and network slice information of the data analysis task are limiting information of the data analysis task and are used to define the specific data analysis task.

[0082] The identification information of the inference object is used to indicate the inference target, for example, it can be a terminal identifier, NF identifier, NF instance identifier, etc.

[0083] The required inference accuracy is used to indicate information such as the accuracy of the model required to be supported, and the required inference latency is used to indicate the time information required for the model to perform inference. Optionally, the model is a model owned by the first communication device, or further, the model is a model owned by the first communication device that meets the requirements of the data analysis task.

[0084] It should be noted that the owned model can also be understood as the available model.

[0085] It should be noted that the inference accuracy and inference delay can also be understood as information such as the accuracy that can be achieved when the first communication device performs inference or information such as the time required.

[0086] It should be noted that precision can be, for example, accuracy, recall rate, mean square error (MSE), etc.

[0087] Step 702: The first communication device obtains information of a third communication device having VFL-related capabilities.

[0088] The third communication device has VFL-related capabilities and can implement VFL training, sample alignment, inference, etc. VFL-related capabilities may, for example, be the ability to provide VFL-related services or the ability to participate in VFL training, sample alignment, inference, etc. Optionally, VFL-related capabilities may include VFL server capabilities, VFL participant capabilities, VFL active participant capabilities, VFL passive participant capabilities, VFL member capabilities, etc. The third communication device may optionally be a communication device with VFL server capabilities, such as an NWDAF, AF, or other communication device.

[0089] After receiving the first message from the second communication device, the first communication device obtains information about a third communication device having VFL-related capabilities based on the request information in the first message. The information about the third communication device includes at least one of identification information, an instance ID, and a Fully Qualified Domain Name (FQDN) of the third communication device. Optionally, the information about the third communication device may further include at least one of capability information (such as VFL-related capability information) of the third communication device, shareable operation indication information (or interoperability indication information, such as an interoperable vendor ID) of the third communication device, and the like.

[0090] Step 703: The first communication device sends a second message to the third communication device based on the information of the third communication device, where the second message is used to request VFL reasoning for the data analysis task, or

[0091] The first communication device sends a first feedback message to the second communication device, where the first feedback message is used to instruct the second communication device to request the third communication device to perform VFL inference.

[0092] When the first communication device determines that the third communication device cannot provide the model, the first communication device sends a second message to the third communication device based on the information of the third communication device, or the first communication device sends a first feedback message to the second communication device.

[0093] In an optional embodiment, the first communication device finds the third communication device based on the information of the third communication device, and sends a second message to the third communication device, where the second message is used to request the third communication device to perform VFL reasoning on the data analysis task. Optionally, the purpose of requesting VFL reasoning on the data analysis task may be to request to obtain the data analysis results corresponding to the data analysis task. Optionally, the first communication device may send the second message to the third communication device through Nnwdaf_AnalyticsSubscription or Nnwdaf_AnalyticsInfo, or services such as Naf_AnlayitcsSubscription or Naf_AnalyticsInfo, or other new services.

[0094] Optionally, the second message includes at least one of the following:

[0095] Relevant information about the data analysis task;

[0096] Identification information of the inference object, where the identification information of the inference object is used to indicate the inference target;

[0097] A VFL model related identifier, wherein the VFL model related identifier is used to indicate identification information of the data analysis task related model;

[0098] The first indication information of VFL inference is used to indicate a request for VFL inference.

[0099] The relevant information of the data analysis task includes at least one of the following: identification information of the data analysis task; time information of the data analysis task; region information of the data analysis task; and network slice information of the data analysis task. The identification information of the data analysis task indicates the specific task category. The time information, region information, and network slice information of the data analysis task are limiting information of the data analysis task.

[0100] The identification information of the inference object can be, for example, a terminal identifier, an NF identifier, or an NF instance identifier. The VFL model correlation identifier is used to indicate the identification information of the relevant model of the data analysis task. The specific VFL task (including the VFL model, VFL training, and information about the communication devices involved in VFL training) can be determined through the VFL model correlation identifier. Optionally, the VFL model correlation identifier can be a model ID, a VFL model ID, or a VFL correlation ID.

[0101] The first indication information of VFL reasoning may, for example, be instructing the third communication device or the target communication device to perform a reasoning operation, i.e., VFL-related model reasoning and other operations. Specifically, the third communication device or the target communication device may be instructed to interact with a communication device such as its training entity (e.g., the sixth communication device) to obtain an intermediate reasoning result, and determine a final reasoning result based on the intermediate reasoning result, i.e., a VFL reasoning result, that is, a data analysis result.

[0102] In another optional embodiment, the first communication device may send a first feedback message to the second communication device, where the first feedback message may include information about a third communication device. The first feedback message is used to instruct the second communication device to request the third communication device to perform VFL inference on the data analysis task, so that the second communication device can interact with the third communication device to obtain data analysis results. For example, the first communication device may send the first feedback message to the second communication device through services such as Nnwdaf_AnalyticsSubscription notify or Nnwdaf_AnalyticsInfo response, or other new services.

[0103] Optionally, the first feedback message includes at least one of the following:

[0104] The data analysis failure indication information is used to indicate the failure of the data analysis task;

[0105] Indication information of a VFL model, wherein the indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model;

[0106] Information of the third communication device.

[0107] The indication information of data analysis failure may also be used to indicate that a model corresponding to a data analysis task cannot be obtained, where the data analysis task is the task corresponding to the analysis identifier (analytics ID), or the data analysis task in the first message.

[0108] The information of the third communication device includes at least one of identification information, an instance identifier, and a network element domain name of the third communication device. Other communication devices can use the information of the third communication device to locate the third communication device and send a request message to the third communication device to request data analysis results. Optionally, the information of the third communication device may further include at least one of capability information of the third communication device (such as capability information related to a VFL), shareable operation indication information of the third communication device (or interoperability indication information, such as an interoperable vendor ID), and the like.

[0109] Optionally, the indication information of data analysis failure may also be used to indicate that the model corresponding to the data analysis task has not been obtained.

[0110] In the communication data processing method provided by the embodiment of the present application, after the first communication device receives a first message from the second communication device, it obtains information of a third communication device with VFL-related capabilities. Based on the information of the third communication device, the first communication device sends a second message to the third communication device to request the third communication device to perform VFL reasoning on the data analysis task, or the first communication device sends a first feedback message to the second communication device, and the first feedback message is used to instruct the second communication device to request the third communication device to perform VFL reasoning. In this way, the first communication device does not need to obtain the VFL model, but requests the third communication device to perform VFL reasoning, or the second communication device requests the third communication device to perform VFL reasoning, thereby realizing the reasoning of the VFL model.

[0111] Optionally, before the first communication device acquires device information of a third communication device having VFL-related capabilities, the method further includes:

[0112] The first communication device sends a third message to the fourth communication device;

[0113] The third message includes at least one of the following information:

[0114] Type information of the communication device being searched, where the type information is used to indicate the type of communication device being searched;

[0115] Target area of ​​interest information, where the target area of ​​interest information is used to indicate a communication device that searches for a service target area;

[0116] Indication information indicating that the communication device being searched has VFL server capability, wherein the indication information indicating that the communication device has VFL server capability is used to indicate that a communication device capable of providing VFL service is to be searched;

[0117] Identification information of the data analysis task that the communication device being searched needs to support;

[0118] The communication device being searched needs to support providing indication information of data analysis results or inference results;

[0119] The inference accuracy that the communication device being searched needs to support;

[0120] The inference latency that the communication device needs to support.

[0121] The fourth communication device is a network element or communication device with a network storage function, or other network elements or devices with a storage function, such as NRF, NEF, UDM, etc. The target type can be a specific network element type or communication device type, such as an NWDAF network element.

[0122] The information indicating that the communication device being searched has VFL server capability may be VFL server capability information, where the VFL server capability information is used to indicate that the communication device is requested to support the VFL server function, or the VFL server capability information is used to indicate that a communication device supporting the VFL server function is preferred, or the VFL server capability information is used to indicate that a communication device supporting the VFL server function is given priority. Optionally, the VFL server capability may refer to the ability to provide VFL services or the ability to preferably provide VFL services.

[0123] It should be noted that the indication information of the VFL server capability may also be implicitly indicated, that is, the VFL server capability information is used to indicate that the communication device to be searched must have the VFL server capability.

[0124] The identification information (Analytics ID) of the data analysis task is used to indicate the category of the data analysis task.

[0125] It should be noted that at least one of the above-mentioned VFL server capability information, indication information for providing data analysis results or reasoning results, reasoning accuracy, reasoning delay, etc. can be based on the granularity of the data analysis task. That is, different data analysis tasks correspond to different VFL server capability information, indication information for providing data analysis results or reasoning results, reasoning accuracy, reasoning delay, etc. The indication information that the communication device being searched needs to support providing data analysis results or reasoning results is used to indicate that the communication device being searched needs to have a service that supports providing data analysis results or reasoning results, such as Nnwdaf_AnalyticsSubscription service or Nnwdaf_AnalyticsInfo service, or a new service (for example, if the communication device being searched is AF, it may be services such as Naf_AnlayitcsSubscription or Naf_AnalyticsInfo, or other new services).

[0126] Inference accuracy can be information such as the accuracy of the model. Inference latency is used to indicate the time required for the model to perform inference. Optionally, the model is a model owned by the communication device being searched, or further, the model is a model owned by the communication device being searched that meets the requirements of the data analysis task.

[0127] The first communication device may send a third message to the fourth communication device to request the fourth communication device to search for a communication device that satisfies at least one item of information in the third message.

[0128] In one embodiment of the present application, after the first communication device sends the third message to the fourth communication device, the method further includes:

[0129] The first communication device receives a second feedback message from the fourth communication device;

[0130] The second feedback message includes at least one of the following:

[0131] Look for failed indications;

[0132] VFL model indication information, where the VFL model indication information is used to indicate that the model supported by the communication device being searched is a VFL model;

[0133] Indication information indicating that sharing is impossible, the indication information indicating that sharing is impossible is used to indicate that the model supported by the communication device being searched cannot be shared;

[0134] Information of the third communication device.

[0135] Based on the third message of the first communication device, the fourth communication device searches for a communication device that meets at least one item of information in the third message, and sends the search result to the first communication device in the form of a second feedback message. The first communication device receives the second feedback message from the fourth communication device.

[0136] In the second feedback message, the search failure indication information is used to indicate that the communication device search has failed, or that no device information meeting the conditions has been found.

[0137] The indication information of the VFL model and the indication information that cannot be shared are used to indicate the reason for the search failure. The indication information of the VFL model is used to indicate that the model supported by the communication device being searched is a VFL model, and the VFL model is a model generated through multi-party VFL training; it can also be understood that the model corresponding to the data analysis task being searched is a VFL model, etc. Therefore, it may not be shared due to privacy issues. The indication information that cannot be shared is used to indicate that the model supported by the communication device being searched cannot be shared, such as if the model is a VFL model, generated through VFL training, a separate model cannot be used, or may involve privacy issues, so it cannot be shared, etc.; it can also be understood that the model corresponding to the data analysis task being searched cannot be shared, etc.

[0138] The first communication device can understand the reason for the search failure based on the second feedback message from the fourth communication device, and can then decide the next step of implementing VFL inference.

[0139] In another embodiment of the present application, after the first communication device sends the third message to the fourth communication device, the method further includes:

[0140] The first communication device receives a third feedback message from the fourth communication device;

[0141] The third feedback message includes at least one of the following:

[0142] Find the successful indication information;

[0143] Information of a target communication device, where the target communication device is a communication device that satisfies the information in the third message;

[0144] The indication information of successful search is used to indicate that a communication device that meets at least one item of information in the third message has been found. The target communication device is a related device that performs VFL reasoning, for example, it can be a third communication device (i.e., a communication device that provides VFL services), or it can also be a communication device other than the third communication device. The information of the target communication device may include: at least one of the identification information of the target communication device, the instance identifier (instance ID), the network element domain name (Fully Qualified Domain Name, FQDN), etc. The third feedback message may also include capability information of the target communication device, sharable operation indication information of the target communication device (or interoperability indication information, such as an interoperable vendor ID), etc.

[0145] The first communication device receives a third feedback message from the fourth communication device, and can obtain information about the third communication device through the target communication device based on the third feedback message.

[0146] Optionally, after the first communication device receives the third feedback message from the fourth communication device, the method further includes:

[0147] The first communication device sends a fourth message to the target communication device, where the fourth message is used to request acquisition of a model corresponding to the data analysis task.

[0148] The first communication device determines to obtain a model based on the first message and may send a fourth message to the target communication device based on the target communication device information in the third feedback message to request the target communication device to obtain the model corresponding to the data analysis task. Specifically, the first communication device may send the fourth message to the target communication device via a message such as Nnwdaf_MLModelProvision_Subscribe or Nnwdaf_MLModelInfo_Request service, or a new service message, to request the target communication device to obtain the model, which may include identification information of the data analysis task.

[0149] Optionally, after the first communication device sends the fourth message to the target communication device, the method further includes:

[0150] The first communication device receives a fourth feedback message from the target communication device;

[0151] The fourth feedback message includes at least one of the following:

[0152] Indication information that a model cannot be provided, wherein the indication information that a model cannot be provided is used to indicate that a model corresponding to the data analysis task cannot be provided;

[0153] Indication information of a VFL model, wherein the indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model;

[0154] A related identifier of a VFL model, wherein the related identifier of the VFL model is used to indicate identification information of a related model of the data analysis task;

[0155] Information of the third communication device.

[0156] Optionally, when a model cannot be provided, the message may also carry information about the reason why the model cannot be provided, such as information indicating that the model is a VFL model, etc. Optionally, the information indicating that a model cannot be provided may also be used to indicate that there is no model corresponding to the data analysis task.

[0157] Among them, the indication information of the VFL model is used to indicate that the model supported by the communication device being searched is a VFL model, and the VFL model is a model generated through multi-party VFL training; it can also be understood as the VFL model of the model corresponding to the data analysis task being searched, etc.

[0158] The VFL model correlation identifier is used to indicate the identification information of the correlation model of the data analysis task. The specific VFL task (including the VFL model, VFL training, information about the communication devices participating in VFL training, etc.) can be determined through the VFL model correlation identifier. Optionally, the VFL model correlation identifier can be a model ID, a VFL model ID, or a VFL correlation ID.

[0159] It should be noted that the target communication device may be a third communication device, or may be different from the third communication device;

[0160] Before receiving the fourth message, the target communication device may send a registration message to the fourth communication device, requesting registration of relevant information of the target communication device, such as at least one of identification information of the target communication device, an instance identifier (instance ID), and a network element domain name (Fully Qualified Domain Name, FQDN). The third feedback message may also include capability information of the target communication device, sharable operation indication information (or interoperability indication information, such as an interoperable vendor ID) of the target communication device, and the like.

[0161] After receiving the fourth message, the target communication device may send a fourth feedback message to the first communication device.

[0162] When the target communication device determines, based on the identification information of the data analysis task in the fourth message, that the model required for the data analysis task is a VFL model and is unable to provide the corresponding model, it may notify the first communication device through a fourth feedback message. Upon receiving the fourth feedback message from the target communication device, the first communication device may learn the reason why the target communication device is unable to provide the model, as well as information about a third communication device that can perform VFL reasoning.

[0163] It should be noted that when the target communication device is a third communication device, the fourth feedback message may not include information about the third communication device, because the first communication device has already queried the information about the third communication device from the fourth communication device, and does not include information about the third communication device, which can reduce the amount of data contained in the sent message.

[0164] In another embodiment of the present application, the first communication device acquires information of the third communication device, including: the first communication device determines the information of the third communication device according to internal logic or pre-configuration.

[0165] Optionally, the internal logic indicates the connection between the first communication device and other communication devices, and determines the information of the third communication device through interaction with the other communication devices.

[0166] Optionally, the pre-configuration may be to pre-configure information about a third communication device capable of performing VFL inference in the first communication device. This allows the first communication device to determine the information about the third communication device based on internal logic or pre-configuration, thus resolving the situation where the information about the third communication device cannot be retrieved from the fourth communication device.

[0167] Optionally, after the first communication device sends a second message to the third communication device based on the information of the third communication device, the method further includes:

[0168] The first communication device receives a data analysis result from the third communication device.

[0169] After the first communication device sends a second message to the third communication device, the third communication device determines to perform a VFL inference operation based on the second message and performs VFL inference with other training entities participating in the VFL. After receiving the intermediate inference results fed back by the training entities, the third communication device determines the final inference result based on the intermediate inference results or its own model, thus obtaining the data analysis result. After determining the data analysis result, the third communication device sends the data analysis result to the first communication device, which then receives the data analysis result from the third communication device.

[0170] It should be noted that when the third communication device fails in executing the data analysis task or fails to obtain the model corresponding to the data analysis task, it will also feedback corresponding failure information.

[0171] Optionally, after the first communication device receives the data analysis result from the third communication device, the method further includes:

[0172] The first communication device sends a fifth feedback message to the second communication device;

[0173] The fifth feedback message includes at least one of the following:

[0174] the data analysis results;

[0175] information of a third communication device;

[0176] The data analysis failure indication information is used to indicate the failure of the data analysis task;

[0177] The indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model.

[0178] The data analysis result is the inference result obtained by executing VFL reasoning. The data analysis result varies for different data analysis tasks. For example, for a data analysis task for predictive service analysis (e.g., analytics ID = Observed Service Experience), the data analysis result may be a mean opinion score (MoS). The data analysis failure indication information can also be used to indicate that the model corresponding to the data analysis task was not obtained.

[0179] After receiving the data analysis result from the third communication device, the first communication device feeds back the data analysis result or indication information of data analysis failure to the second communication device through the fifth feedback message, thereby obtaining the result of VFL inference.

[0180] Exemplarily, when the data analysis fails, the fifth feedback message may include indication information of the data analysis failure, information of the third communication device, and indication information of the VFL model.

[0181] Optionally, the indication information of the VFL model may also be understood as failure reason value information, etc.

[0182] Exemplarily, when the data analysis is successful, the fifth feedback message may include the data analysis result, information about the third communication device, and indication information of the VFL model.

[0183] It should be noted that, in addition to the above information, the fifth feedback message may also include at least one of the following: relevant information of the data analysis task, identification information of the inference object, etc.

[0184] Optionally, before the first communication device sends the second message to the third communication device, the method further includes:

[0185] The first communication device sends a fifth message to the fifth communication device, where the fifth message is used to request that the identification information of the inference object be converted into a data format of the third communication device;

[0186] The first communication device receives identification information of the converted inference object from the fifth communication device.

[0187] The fifth communication device may be, for example, NEF, UDM, etc.

[0188] During the interaction between the first communication device and the third communication device, the data of the two parties cannot be communicated due to the different domains of the first communication device and the third communication device or other reasons, that is, they cannot be recognized or used by the other party. At this time, the first communication device, based on its own configuration / internal logic and other reasons, such as knowing the network element type of the third communication device (such as when the network element type of the third communication device is AF), determines that the identification information of the inference object it owns is inconsistent with the identification information of the inference object owned by the third communication device. It can perform a data conversion operation before transmitting the data, converting the data into a format that can be recognized or used by the third communication device, and then sending the identification information of the converted inference object to the third communication device. When performing data conversion, the first communication device can send a fifth message to the fifth communication device, and the fifth message is used to request that the identification information of the inference object be converted into the data format of the third communication device, or the fifth message is used to request to obtain the data format of the third communication device.

[0189] The fifth message may include at least one of the following:

[0190] Identification information of the inference object, such as identification information of the inference object used for transformation;

[0191] Instruction information for requesting conversion, wherein the instruction information for requesting conversion is used to indicate that conversion of the inference object identification is required, such as converting the identification information of the inference object into identification information of another type of inference object;

[0192] The target format, or target type, is used to indicate the type of inference object identifier required, such as a Subscription Permanent Identifier (SUPI), when the inference object identifier is of the Generic Public Subscription Identifier (GPSI) type and needs to be converted to a SUPI.

[0193] The type / identification of the third communication device, such as the AF instance identifier (instance ID), so that the fifth communication device (such as NEF) determines the conversion format based on the information of the third communication device; for example, through the AF instance identifier or the AF specific identifier (specific ID), the fifth communication device can determine the target type to be converted based on internal logic or other methods, such as GPSI (External Id) or GPSI (MSISDN).

[0194] Optionally, the first communication device sending a second message to the third communication device based on the information of the third communication device includes:

[0195] When determining that the third communication device cannot provide the model, the first communication device sends the second message to the third communication device;

[0196] or,

[0197] The first communication device sending a first feedback message to the second communication device includes:

[0198] When determining that the third communication device cannot provide the model, the first communication device sends the first feedback message to the second communication device.

[0199] In an optional embodiment, when the first communication device determines that the third communication device cannot provide a model, it is impossible to directly use the model for reasoning to obtain data analysis results. The first communication device can send a second message to the third communication device to request the third communication device to perform VFL reasoning.

[0200] In another optional embodiment, when the first communication device determines that the third communication device cannot provide the model, it is impossible to directly use the model for reasoning to obtain data analysis results. The first communication device can send a first feedback message to the second communication device to instruct the second communication device to request the third communication device to perform VFL reasoning.

[0201] FIG8 is a flowchart of another communication data processing method provided in an embodiment of the present application. The communication data processing method is executed by a second communication device. The second communication device may be an NF consumer, a communication device requesting data analysis results, or any communication device capable of interacting with an NRF or NWDAF, such as a PCF or LMF. As shown in FIG8 , the communication data processing method includes the following steps:

[0202] Step 801: The second communication device sends a first message to the first communication device, where the first message is used to request to obtain a data analysis result corresponding to a data analysis task.

[0203] The first communication device and the second communication device are network side devices, such as the core network devices shown in Figure 1. The first communication device is a network element device with reasoning capability in the communication network, such as AnLF, NWDAF, LMF and other intelligent network element devices.

[0204] Before the second communication device sends the first message to the first communication device, it can search for a communication device that can perform the data analysis task from the fourth communication device. Exemplarily, the second communication device sends a network element search request to the fourth communication device, requesting to find a server network element that meets the conditions; for example, a network element search request can be sent to the third communication device through the Nnrf_NFDiscovery_Request service, etc.; the network element search request can carry the type information of the communication device being searched (used to indicate the type of network element or communication device being searched, such as NWDAF network element, etc.), the target area of ​​interest information (used to indicate the service range of the communication device being searched, etc.), and can also carry the identification information (analytics ID) of the data analysis task and the service information corresponding to the identification information of the data analysis task. For example, the service information corresponding to the identification information of the data analysis task can be an inference service, such as Nnwdaf_AnalyticsSubscription or Nnwdaf_AnalyticsInfo service, or the service information corresponding to the identification information of the data analysis task can also be a new service, etc. The fourth communication device is a network element or communication device with a network storage function, and may also be other network elements or devices with storage functions, such as NRF, NEF, UDM, etc. The identification information (Analytics ID) of the data analysis task is used to identify the category of the data analysis task.

[0205] The fourth communication device searches for information of a communication device that satisfies various information in the network element search request based on the network element search request from the second communication device, i.e., finds information of the first communication device, and feeds back the information of the first communication device to the second communication device. The information of the first communication device may include network element instance identification information (NF instance ID), address information (IP address of NF), or network element domain name, etc.

[0206] The second communication device sends a first message to the first communication device based on the information of the first communication device, so as to request the first communication device to obtain the data analysis result corresponding to the data analysis task.

[0207] Optionally, the first message includes at least one of the following:

[0208] Relevant information about the data analysis task;

[0209] Identification information of the inference object;

[0210] The required supported inference accuracy;

[0211] The required inference latency.

[0212] The relevant information of the data analysis task includes at least one of the following: identification information of the data analysis task; time information of the data analysis task; region information of the data analysis task; and network slice information of the data analysis task. The identification information of the data analysis task is used to indicate the specific task category. The time information, region information, and network slice information of the data analysis task are limiting information of the data analysis task and are used to define the specific data analysis task.

[0213] The identification information of the inference object is used to indicate the inference target, for example, it can be a terminal identifier, NF identifier, NF instance identifier, etc.

[0214] The required inference accuracy is used to indicate information such as the accuracy of the model required to be supported, and the required inference latency is used to indicate the time information required for the model to perform inference. Optionally, the model is a model owned by the first communication device, or further, the model is a model owned by the first communication device that meets the requirements of the data analysis task.

[0215] It should be noted that the owned model can also be understood as the available model.

[0216] It should be noted that the inference accuracy and inference delay can also be understood as information such as the accuracy that can be achieved when the first communication device performs inference or information such as the time required.

[0217] It should be noted that precision can be, for example, accuracy, recall rate, mean square error, etc.

[0218] Step 802: The second communication device receives a fifth feedback message from the first communication device, where the fifth feedback message indicates a result of performing VFL reasoning on the data analysis task, or

[0219] The second communication device receives a first feedback message from the first communication device, and sends a sixth message to the third communication device based on the first feedback message, wherein the sixth message is used to request VFL reasoning for the data analysis task, and the third communication device has VFL-related capabilities.

[0220] The third communication device has VFL-related capabilities and can implement VFL training, sample alignment, inference, etc. VFL-related capabilities may, for example, be the ability to provide VFL-related services or the ability to participate in VFL training, sample alignment, inference, etc. Optionally, VFL-related capabilities may include VFL server capabilities, VFL participant capabilities, VFL active participant capabilities, VFL passive participant capabilities, VFL member capabilities, etc. The third communication device may optionally be a communication device with VFL server capabilities, such as an NWDAF, AF, or other communication device.

[0221] In an optional embodiment, the first communication device may request the third communication device to perform VFL reasoning on the data analysis task based on the information of the third communication device, and obtain the result of the VFL reasoning, and feed back the result of the VFL reasoning to the second communication device through the fifth feedback message. The second communication device receives the fifth feedback message from the first communication device, thereby obtaining the result of the VFL reasoning on the data analysis task. The information of the third communication device includes: at least one of the identification information, instance identification, network element domain name, etc. of the third communication device. Optionally, the information of the third communication device may also include at least one of the following: capability information of the third communication device (such as VFL-related capability information), shareable operation indication information of the third communication device (or interoperability indication information, such as interoperable vendor ID), etc.

[0222] Optionally, the fifth feedback message includes at least one of the following:

[0223] The data analysis results corresponding to the data analysis task;

[0224] information of a third communication device;

[0225] The data analysis failure indication information is used to indicate the failure of the data analysis task;

[0226] The indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model.

[0227] The data analysis result is the inference result obtained by executing VFL inference. The data analysis result varies for different data analysis tasks. For example, for a data analysis task for predictive service analysis (e.g., analytics ID = Observed Service Experience), the data analysis result may be a mean opinion score. The data analysis failure indication information can also be used to indicate that the model corresponding to the data analysis task was not obtained.

[0228] The fifth feedback message may include data analysis results or indication information of analysis failure.

[0229] Exemplarily, when the data analysis fails, the fifth feedback message may include indication information of the data analysis failure, information of the third communication device, and indication information of the VFL model.

[0230] Optionally, the indication information of the VFL model may also be understood as failure reason value information, etc.

[0231] Exemplarily, when the data analysis is successful, the fifth feedback message may include the data analysis result, information about the third communication device, and indication information of the VFL model.

[0232] It should be noted that, in addition to the above information, the fifth feedback message may also include at least one of the following: relevant information of the data analysis task, identification information of the inference object, etc.

[0233] In another optional embodiment, the second communication device receives a first feedback message from the first communication device, which may include information of the third communication device. The second communication device sends a sixth message to the third communication device based on the information of the third communication device to request the third communication device to perform VFL inference on the data analysis task, thereby obtaining the result of VFL inference.

[0234] Optionally, the first feedback message includes at least one of the following:

[0235] The data analysis failure indication information is used to indicate the failure of the data analysis task;

[0236] Indication information of a VFL model, wherein the indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model;

[0237] Information of the third communication device.

[0238] The indication information of data analysis failure may also be used to indicate that the model corresponding to the data analysis task cannot be obtained, where the data analysis task is the task corresponding to the analysis identifier (analytics ID) or the data analysis task in the first message. Optionally, the indication information of data analysis failure may also be used to indicate that the model corresponding to the data analysis task cannot be obtained.

[0239] The information of the third communication device includes at least one of identification information, an instance identifier, and a network element domain name of the third communication device. Other communication devices can use the information of the third communication device to locate the third communication device and send a request message to the third communication device to request data analysis results. Optionally, the information of the third communication device may further include at least one of capability information of the third communication device (such as capability information related to a VFL), shareable operation indication information of the third communication device (or interoperability indication information, such as an interoperable vendor ID), and the like.

[0240] Optionally, the sixth message includes at least one of the following:

[0241] Relevant information about the data analysis task;

[0242] Identification information of the inference object, where the identification information of the inference object is used to indicate the inference target;

[0243] A VFL model related identifier, wherein the VFL model related identifier is used to indicate identification information of the data analysis task related model;

[0244] The first indication information of VFL inference is used to indicate a request for VFL inference.

[0245] The relevant information of the data analysis task includes at least one of the following: identification information of the data analysis task; time information of the data analysis task; region information of the data analysis task; and network slice information of the data analysis task. The identification information of the data analysis task indicates the specific task category. The time information, region information, and network slice information of the data analysis task are limiting information of the data analysis task.

[0246] The identification information of the inference object can be, for example, a terminal identifier, an NF identifier, or an NF instance identifier. The VFL model correlation identifier is used to indicate the identification information of the relevant model of the data analysis task. The specific VFL task (including the VFL model, VFL training, and information about the communication devices involved in VFL training) can be determined through the VFL model correlation identifier. Optionally, the VFL model correlation identifier can be a model ID, a VFL model ID, or a VFL correlation ID.

[0247] The first indication information of VFL reasoning may, for example, instruct the third communication device or the target communication device to perform a reasoning operation, i.e., VFL-related model reasoning and other operations. Specifically, the third communication device or the target communication device may be instructed to interact with a communication device such as its training entity (e.g., the sixth communication device) to obtain an intermediate reasoning result, and determine a final reasoning result, i.e., a VFL reasoning result, based on the intermediate reasoning result.

[0248] In the communication data processing method provided by the embodiment of the present application, the second communication device sends a first message to the first communication device, and the first message is used to request to obtain the data analysis result corresponding to the data analysis task; the second communication device receives a fifth feedback message from the first communication device, and the fifth feedback message is used to indicate the result of VFL reasoning on the data analysis task, or the second communication device receives the first feedback message from the first communication device, and sends a sixth message to the third communication device based on the first feedback message, and the sixth message is used to request VFL reasoning on the data analysis task, so that the second communication device can directly request the third communication device to perform VFL reasoning, so that the reasoning of the VFL model can be completed without obtaining the model, and no data transmission is required through VFL reasoning, thereby ensuring the security of the data.

[0249] Optionally, before the second communication device sends a sixth message to the third communication device based on the first feedback message, the method further includes:

[0250] The second communication device sends a seventh message to the fifth communication device, where the seventh message is used to request that the identification information of the inference object be converted into a data format of the third communication device;

[0251] The second communication device receives the identification information of the converted inference object from the fifth communication device.

[0252] The fifth communication device may be, for example, NEF, UDM, etc.

[0253] During the interaction between the second communication device and the third communication device, the data of the two parties may not be able to communicate with each other due to the different domains of the second communication device and the third communication device or other reasons, that is, they cannot be recognized or used by the other party. At this time, the second communication device, based on its own configuration / internal logic and other reasons, such as knowing the network element type of the third communication device (such as when the network element type of the third communication device is AF), determines that the identification information of the inference object it owns is inconsistent with the identification information of the inference object owned by the third communication device. It can perform a data conversion operation before transmitting the data, converting the data into a format that can be recognized or used by the third communication device, and then sending the identification information of the converted inference object to the third communication device. When performing data conversion, the second communication device can send a seventh message to the fifth communication device, and the seventh message is used to request that the identification information of the inference object be converted into the data format of the third communication device, or the seventh message is used to request to obtain the data format of the third communication device.

[0254] It should be noted that the data format of the third communication device may be, for example, a format that can be recognized or used by the third communication device.

[0255] The seventh message may include at least one of the following:

[0256] Identification information of the inference object, such as identification information of the inference object used for conversion;

[0257] Instruction information for requesting conversion, wherein the instruction information for requesting conversion is used to indicate that conversion of the inference object identification is required, such as converting the identification information of the inference object into identification information of another type of inference object;

[0258] The target format, or target type, is used to indicate the type of inference object identifier required, such as a Subscription Permanent Identifier (SUPI), when the inference object identifier is of the Generic Public Subscription Identifier (GPSI) type and needs to be converted to a SUPI.

[0259] The type / identification of the third communication device, such as the AF instance identifier (instance ID), so that the fifth communication device (such as NEF) determines the conversion format based on the information of the third communication device; for example, through the AF instance identifier or the AF specific identifier (specific ID), the fifth communication device can determine the target type to be converted based on internal logic or other methods, such as GPSI (External Id) or GPSI (MSISDN).

[0260] After the fifth communication device completes the conversion of the identification information of the inference object, it sends the converted identification information of the inference object to the second communication device. The second communication device receives the converted identification information of the inference object from the fifth communication device and can send a sixth message to the third communication device. The identification information of the inference object carried in the sixth message is the converted identification information of the inference object.

[0261] Optionally, after the second communication device sends a sixth message to the third communication device based on the first feedback message, the method further includes:

[0262] The second communication device receives the data analysis result from the third communication device.

[0263] After the second communication device sends the sixth message to the third communication device, the third communication device determines to perform a VFL inference operation based on the sixth message and performs VFL inference with other training entities participating in the VFL. After receiving the intermediate inference results fed back by the training entities, the third communication device determines the final inference result based on the intermediate inference results or its own model, thus obtaining the data analysis result. After determining the data analysis result, the third communication device sends the data analysis result to the second communication device, and the second communication device receives the data analysis result from the third communication device.

[0264] It should be noted that when the third communication device fails in executing the data analysis task or fails to obtain the model corresponding to the data analysis task, it will also feedback corresponding failure information.

[0265] Figure 9 is a flowchart of a communication data processing method provided in an embodiment of the present application. The communication data processing method is executed by a third communication device. The third communication device has VFL-related capabilities and can implement VFL training, sample alignment, reasoning, etc. VFL-related capabilities may, for example, be the ability to provide VFL-related services, or the ability to participate in VFL training, sample alignment, reasoning, etc. Optionally, VFL-related capabilities may include VFL server capabilities, VFL participant capabilities, VFL active participant capabilities, VFL passive participant capabilities, VFL member capabilities, etc. The third communication device may optionally be a communication device with VFL server capabilities, for example, it may be a communication device such as NWDAF, AF, etc. As shown in Figure 9, the communication data processing method includes the following steps:

[0266] Step 901: A third communication device receives a second message from a first communication device or a sixth message from a second communication device, where the second message or the sixth message is used to request VFL reasoning for a data analysis task.

[0267] The first, second, and third communication devices are network-side devices, such as the core network devices shown in Figure 1. The first communication device is a network element device with reasoning capabilities in the communication network, such as an intelligent network element device such as AnLF or LMF. The second communication device is an NF consumer, a communication device that requests data analysis results. It can be any communication device that can interact with the NRF or NWDAF, such as the PCF or LMF.

[0268] In an optional implementation, the third communication device receives a second message from the first communication device, where the second message is used to request VFL reasoning for the data analysis task.

[0269] In another optional implementation, the third communication device receives a sixth message from the second communication device, where the sixth message is used to request VFL reasoning for the data analysis task.

[0270] Optionally, the second message or the sixth message includes at least one of the following:

[0271] Relevant information about the data analysis task;

[0272] Identification information of the inference object, where the identification information of the inference object is used to indicate the inference target;

[0273] A VFL model related identifier, wherein the VFL model related identifier is used to indicate identification information of the data analysis task related model;

[0274] The first indication information of VFL inference is used to indicate a request for VFL inference.

[0275] The relevant information of the data analysis task includes at least one of the following: identification information of the data analysis task; time information of the data analysis task; region information of the data analysis task; and network slice information of the data analysis task. The identification information of the data analysis task indicates the specific task category. The time information, region information, and network slice information of the data analysis task are limiting information of the data analysis task.

[0276] The identification information of the inference object can be, for example, a terminal identifier, an NF identifier, or an NF instance identifier. The VFL model correlation identifier is used to indicate the identification information of the relevant model of the data analysis task. The specific VFL task (including the VFL model, VFL training, and information about the communication devices involved in VFL training) can be determined through the VFL model correlation identifier. Optionally, the VFL model correlation identifier can be a model ID, a VFL model ID, or a VFL correlation ID.

[0277] The first indication information of VFL reasoning may, for example, instruct the third communication device or the target communication device to perform a reasoning operation, i.e., VFL-related model reasoning and other operations. Specifically, the third communication device or the target communication device may be instructed to interact with a communication device such as its training entity (e.g., the sixth communication device) to obtain an intermediate reasoning result, and determine a final reasoning result, i.e., a VFL reasoning result, based on the intermediate reasoning result.

[0278] Step 902: The third communication device determines a data analysis result corresponding to the data analysis task.

[0279] After receiving the second message from the first communication device or the sixth message from the second communication device, the third communication device works together with other communication devices participating in VFL training to complete the VFL reasoning of the data analysis task and obtain the data analysis results.

[0280] The data analysis result is the inference result obtained by executing VFL reasoning. The data analysis results are different for different data analysis tasks. For example, for a data analysis task for predictive service analysis (such as analytics ID = Observed Service Experience), the data analysis result may be an average opinion score, etc.

[0281] Step 903: The third communication device sends the data analysis result to the first communication device or the second communication device.

[0282] In an optional implementation, the third communication device determines the data analysis result based on the second message from the first communication device, and then sends the data analysis result to the first communication device.

[0283] In another optional implementation, the third communication device determines the data analysis result based on the sixth message from the second communication device, and then sends the data analysis result to the second communication device.

[0284] In the communication data processing method provided in the embodiment of the present application, after the third communication device receives the second message from the first communication device or the sixth message from the second communication device, it obtains the data analysis result corresponding to the data analysis task and sends the data analysis result to the first communication device or the second communication device. The third communication device does not need to provide a model, but instead performs VFL reasoning and sends the obtained data analysis result to the first communication device or the second communication device, thereby realizing VFL reasoning.

[0285] On the basis of the above technical solution, before the third communication device determines the data analysis result corresponding to the data analysis task, the method further includes:

[0286] The third communication device sends an eighth message to the sixth communication device, and the eighth message includes auxiliary information or task information. The auxiliary information is used to indicate the determination method of the input data or the feedback time of the intermediate reasoning result, and the task information is used to indicate information of the data analysis task.

[0287] Among them, the sixth communication device is a training entity device, which is a communication device that can perform or participate in VFL training or reasoning, and can also refer to a communication device with VFL-related capabilities, such as NWDAF, Trusted AF, NEF, etc.

[0288] It should be noted that there may be one or more sixth communication devices.

[0289] After receiving the second message or the sixth message, the third communication device determines to perform operations such as VFL inference, and the third communication device determines assistance information or task information to complete operations such as VFL inference or distributed inference.

[0290] Optionally, the third communication device may determine the auxiliary information or the task information based on the second message or the sixth message.

[0291] Optionally, the third communication device may determine the auxiliary information or task information based on the relevant information of the data analysis task in the second message or the sixth message, the identification information of the inference object, the relevant identification of the VFL model and other information.

[0292] Optionally, the third communication device sends an eighth message to the sixth communication device, which may carry auxiliary information or task information, requesting the sixth communication device to perform VFL reasoning, and is also used to request each sixth communication device to complete reasoning based on local data and models, and generate and feedback intermediate reasoning results.

[0293] Optionally, the auxiliary information includes at least one of the following:

[0294] Time information of input data required when performing the VFL inference;

[0295] The data type of the input data required when performing the VFL inference;

[0296] Limitation information of input data required when performing the VFL inference;

[0297] The maximum response time required to feed back the intermediate inference result when executing the VFL inference.

[0298] The time information of the input data required when performing VFL reasoning is used to indicate when the sixth communication device needs to collect data as input data, or when to collect data as input data; the data type (feature space / data type / event ID) of the input data required when performing VFL reasoning is used to indicate what data type of data the sixth communication device needs to collect as input data; the limitation information of the input data required when performing VFL reasoning includes slice limitation information or area limitation information, etc., which is used to indicate which slice or specific area the sixth communication device collects data in as input data; the maximum response time required to feedback the intermediate reasoning result when performing VFL reasoning is used to indicate the sixth communication device to feedback the intermediate reasoning result before the maximum response time. The maximum response time can be a specific time point or a time period (such as within a few seconds).

[0299] Optionally, the task information includes at least one of the following:

[0300] identification information of the data analysis task;

[0301] A VFL model related identifier, wherein the VFL model related identifier is used to indicate identification information of the data analysis task related model;

[0302] The identification information of the inference object is used to indicate the inference object identification information corresponding to the input data.

[0303] Among them, the identification information (Analytics ID) of the data analysis task is used to indicate the specific task category. The relevant identification of the VFL model (VFL correlation ID / model ID) is used to indicate the identification information of the data analysis task related model, or to indicate the identification information of the VFL task. The specific VFL task (including VFL model, VFL training, information of communication devices participating in VFL training, etc.) can be determined through the relevant identification of the VFL model. Optionally, the relevant identification of the VFL model can be model ID, VFL model ID or VFL correlation ID, etc. The sixth communication device (training entity) and the third communication device (Server), etc. can determine the model to be used through the relevant identification. The third communication device can determine the relevant identification of the VFL model based on the identification information of the data analysis task. The identification information of the inference object is used to indicate the inference object identification information to which the input data must correspond, that is, the input data should correspond to or be related to the identification information of the inference object.

[0304] Optionally, after the third communication device sends the eighth message to the sixth communication device, the method further includes:

[0305] The third communication device receives a sixth feedback message from the sixth communication device, where the sixth feedback message includes an intermediate inference result.

[0306] The sixth communication device determines the input data and / or the model to be used based on the eighth message, thereby completing the reasoning and generating an intermediate reasoning result. Specifically, the sixth communication device may determine the input data based on the auxiliary information in the eighth message, such as when and what data to collect; the sixth communication device may determine which model to use based on the task information in the eighth message, specifically, the model to be used may be determined based on the identification information (analytics ID) of the data analysis task in the task message or the correlation identifier (VFL correlation ID) of the VFL model;

[0307] After the sixth communication device determines the input data and / or model to be used, a possible implementation method is to collect data and select input data, input the input data into the model, and generate model output data. The output data serves as the model inference result, that is, the intermediate inference result, thereby enabling each training entity device to complete the inference of its own model.

[0308] The third communication device receives a sixth feedback message from the sixth communication device, where the sixth feedback message includes the intermediate reasoning result.

[0309] Optionally, the sixth feedback message further includes at least one of the following:

[0310] Identification information of the inference object, where the identification information of the inference object is used to indicate the inference target;

[0311] second indication information of VFL reasoning, wherein the second indication information of VFL reasoning is used to indicate that the intermediate reasoning result is obtained through VFL reasoning;

[0312] The VFL model related identifier is used to indicate the identification information of the model related to the data analysis task;

[0313] Identification information of the data analysis task.

[0314] The VFL model correlation identifier (VFL correlation ID / model ID) is used to indicate the identification information of the model related to the data analysis task, or to indicate the identification information of the VFL task. The specific VFL task (including the VFL model, VFL training, information about the communication devices participating in the VFL training, etc.) can be determined through the VFL model correlation identifier. Optionally, the VFL model correlation identifier can be a model ID, a VFL model ID, or a VFL correlation ID.

[0315] Optionally, the sixth feedback message may further include auxiliary information or task information.

[0316] Optionally, the third communication device determines the data analysis result corresponding to the data analysis task, including:

[0317] The third communication device determines a data analysis result based on the intermediate inference result.

[0318] After receiving the intermediate reasoning result from the sixth communication device, the third communication device calculates and determines the final reasoning result, ie, the data analysis result, based on the intermediate reasoning result and / or its own model.

[0319] For example, in one possible implementation, the third communication device inputs the intermediate inference result fed back by the sixth communication device as input data into its own model, and generates model output data, that is, the inference result of the entire VFL, to obtain a data analysis result.

[0320] Through the interaction between the third communication device and each sixth communication device, the sixth communication device performs VFL reasoning to obtain intermediate reasoning results, and the third communication device determines the data analysis result based on each intermediate reasoning result, thereby realizing distributed reasoning of VFL.

[0321] Optionally, the third communication device sending an eighth message to the sixth communication device includes:

[0322] The third communication device sends a ninth message to the fifth communication device, where the ninth message is used to request that the eighth message be forwarded to the sixth communication device.

[0323] The fifth communication device may be, for example, NEF, UDM, etc.

[0324] The third communication device may send the eighth message to the sixth communication device via the fifth communication device.

[0325] In an optional embodiment, when the third communication device sends the eighth message to the sixth communication device via a fifth communication device such as an NEF, the fifth communication device may convert the identification information of the inference object into a data format that can be recognized / used by the sixth communication device. In this embodiment, no additional / dedicated format conversion interaction is required between the third communication device (server), the sixth communication device (training entity), and the fifth communication device. Instead, the fifth communication device may discover the need for conversion during the transmission of the eighth message. For example, the fifth communication device may determine, based on internal logic or pre-configuration, that the identification information of the inference object cannot be recognized / used by the sixth communication device and decide to perform the conversion. Specifically, the third communication device may first send a ninth message to the fifth communication device, requesting that the eighth message be forwarded to the sixth communication device. The fifth communication device may determine, based on internal logic or pre-configuration, that the identification information of the inference object in the eighth message cannot be recognized / used by the sixth communication device, convert the identification information of the inference object, and forward the converted identification information and the eighth message to the sixth communication device. In this implementation method, the sixth communication device does not need to request data format conversion.

[0326] Optionally, before the third communication device sends the eighth message to the sixth communication device, the method further includes:

[0327] The third communication device sends a tenth message to the fifth communication device, where the tenth message is used to request that the identification information of the inference object be converted into a data format of the sixth communication device;

[0328] The third communication device receives the identification information of the converted inference object from the fifth communication device.

[0329] During the interaction between the third communication device and the sixth communication device, the data of the two parties may not be able to communicate with each other due to the different domains of the third communication device and the sixth communication device or other reasons, that is, they cannot be recognized or used by the other party. At this time, the third communication device, based on its own configuration / internal logic and other reasons, such as knowing the network element type of the sixth communication device (such as when the network element type of the third communication device is AF), determines that the identification information of the inference object it owns is inconsistent with the identification information of the inference object owned by the third communication device. It can perform a data conversion operation before transmitting the data, converting the data into a format that can be recognized or used by the sixth communication device, and then sending the identification information of the converted inference object to the sixth communication device. When performing data conversion, the third communication device can send a tenth message to the fifth communication device, and the tenth message is used to request that the identification information of the inference object be converted into the data format of the sixth communication device, or the tenth message is used to request to obtain the data format of the sixth communication device.

[0330] It should be noted that the data format of the sixth communication device may be, for example, a format that can be recognized or used by the sixth communication device.

[0331] The tenth message may include at least one of the following:

[0332] Identification information of the inference object, such as identification information of the inference object used for conversion;

[0333] Instruction information for requesting conversion, wherein the instruction information for requesting conversion is used to indicate that conversion of the inference object identification is required, such as converting the identification information of the inference object into identification information of another type of inference object;

[0334] The target format, or target type, is used to indicate the type of inference object identifier required, such as a Subscription Permanent Identifier (SUPI), when the inference object identifier is of the Generic Public Subscription Identifier (GPSI) type and needs to be converted to a SUPI.

[0335] The type / identification of the sixth communication device, such as the AF instance identifier (instance ID), so that the fifth communication device (such as NEF) determines the conversion format according to the information of the sixth communication device; for example, through the AF instance identifier or the AF specific identifier (specific ID), the fifth communication device can determine the target type to be converted based on internal logic or other methods, such as GPSI (External Id) or GPSI (MSISDN).

[0336] After the fifth communication device completes the conversion of the identification information of the inference object, it sends the converted identification information of the inference object to the third communication device. The third communication device receives the converted identification information of the inference object from the fifth communication device and can send the eighth message to the sixth communication device. The identification information of the inference object carried in the eighth message is the converted identification information of the inference object.

[0337] Optionally, before the third communication device determines the data analysis result according to the intermediate inference result, the method further includes:

[0338] The third communication device sends an eleventh message to the fifth communication device, where the eleventh message is used to request that identification information of the inference object in the intermediate inference result be converted into a data format of the third communication device;

[0339] The third communication device receives the identification information of the converted inference object from the fifth communication device.

[0340] After the third communication device receives the intermediate inference result fed back by the sixth communication device, it finds that it cannot identify the identification information of the inference object. At this time, the third communication device can send an eleventh message to the fifth communication device. The eleventh message is used to request that the identification information of the inference object be converted from the data format of the sixth communication device to the data format of the third communication device.

[0341] It should be noted that the data format of the third communication device may be, for example, a format that can be recognized or used by the third communication device.

[0342] The eleventh message may include at least one of the following:

[0343] Identification information of the inference object, such as identification information of the inference object used for conversion;

[0344] Instruction information for requesting conversion, wherein the instruction information for requesting conversion is used to indicate that conversion of the inference object identification is required, such as converting the identification information of the inference object into identification information of another type of inference object;

[0345] The target format, or target type, is used to indicate the type of inference object identifier required, such as a Subscription Permanent Identifier (SUPI), when the inference object identifier is of the Generic Public Subscription Identifier (GPSI) type and needs to be converted to a SUPI.

[0346] The type / identification of the third communication device, such as the AF instance identifier (instance ID), so that the fifth communication device (such as NEF) determines the conversion format based on the information of the third communication device; for example, through the AF instance identifier or the AF specific identifier (specific ID), the fifth communication device can determine the target type to be converted based on internal logic or other methods, such as GPSI (External Id) or GPSI (MSISDN).

[0347] After the fifth communication device completes the conversion of the identification information of the inference object, it sends the converted identification information of the inference object to the third communication device. The third communication device receives the converted identification information of the inference object from the fifth communication device and can determine the data analysis result based on the intermediate inference result.

[0348] Optionally, the method further includes:

[0349] The third communication device receives a fourth message from the first communication device, where the fourth message is used to request acquisition of a model corresponding to the data analysis task.

[0350] The first communication device may determine to obtain a model based on the first message and, based on the target communication device information in the third feedback message, send a fourth message to the third communication device, if the target communication device is the third communication device, to request the third communication device to obtain the model corresponding to the data analysis task. Specifically, the first communication device may send the fourth message to the third communication device via a message such as Nnwdaf_MLModelProvision_Subscribe or Nnwdaf_MLModelInfo_Request service, requesting to obtain the model, which may include identification information of the data analysis task.

[0351] Optionally, the target communication device is a related device that performs VFL reasoning, for example, it can be a third communication device (i.e., a communication device that provides VFL services), or it can be a communication device other than the third communication device. The information of the target communication device may include: at least one of the identification information of the target communication device, an instance identifier (instance ID), a network element domain name (Fully qualified domain name, FQDN), etc. The third feedback message may also include capability information of the target communication device, shareable operation indication information of the target communication device (or interoperability indication information, such as an interoperable vendor ID), etc.

[0352] It should be noted that the target communication device may be a third communication device, or may be different from the third communication device;

[0353] Before receiving the fourth message, the target communication device may send a registration message to the fourth communication device, requesting registration of relevant information of the target communication device, such as at least one of the identification information, instance ID, and network element domain name of the target communication device. When the first communication device queries the fourth communication device for a device that meets at least one of the information in the third message, if the target communication device is found, the third feedback message sent by the fourth communication device to the first communication device may further include capability information of the target communication device, sharable operation indication information (or interoperability indication information, such as an interoperable vendor ID), and the like of the target communication device.

[0354] After receiving the fourth message, the target communication device may send a fourth feedback message to the first communication device.

[0355] Optionally, after the third communication device receives the fourth message from the first communication device, the method further includes:

[0356] The third communication device sends a fourth feedback message to the first communication device;

[0357] The fourth feedback message includes at least one of the following:

[0358] Indication information that a model cannot be provided, wherein the indication information that a model cannot be provided is used to indicate that a model corresponding to the data analysis task cannot be provided;

[0359] Indication information of a VFL model, wherein the indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model;

[0360] A related identifier of a VFL model, wherein the related identifier of the VFL model is used to indicate identification information of a related model of the data analysis task;

[0361] Information of the third communication device.

[0362] Optionally, when a model cannot be provided, the message may also carry information about the reason why the model cannot be provided, such as information indicating that the model is a VFL model, etc. Optionally, the information indicating that a model cannot be provided may also be used to indicate that there is no model corresponding to the data analysis task.

[0363] Among them, the indication information of the VFL model is used to indicate that the model supported by the communication device being searched is a VFL model, and the VFL model is a model generated through multi-party VFL training; it can also be understood as the VFL model of the model corresponding to the data analysis task being searched, etc.

[0364] The VFL model correlation identifier is used to indicate the identification information of the correlation model of the data analysis task. The specific VFL task (including the VFL model, VFL training, information about the communication devices participating in VFL training, etc.) can be determined through the VFL model correlation identifier. Optionally, the VFL model correlation identifier can be a model ID, a VFL model ID, or a VFL correlation ID.

[0365] The information of the third communication device includes: at least one of identification information of the third communication device, an instance identifier (instance ID), a network element domain name, etc. Optionally, the information of the third communication device may further include at least one of the following: capability information of the third communication device (such as capability information related to the VFL), shareable operation indication information of the third communication device (or interoperability indication information, such as an interoperable vendor ID), etc.

[0366] If the third communication device determines, based on the identification information of the data analysis task in the fourth message, that the model required for the data analysis task is a VFL model and cannot provide the corresponding model, it may notify the first communication device via a fourth feedback message. For example, the third communication device may send the fourth feedback message to the first communication device via a service message such as Nnwdaf_MLModelProvision_notify or Nnwdaf_MLModelInfo response.

[0367] Optionally, the method further includes:

[0368] The third communication device sends a twelfth message to the fourth communication device, where the twelfth message is used to request registration or update of network element capability information of the third communication device.

[0369] The third communication device sends a twelfth message to the fourth communication device, where the twelfth message includes network element capability information. The twelfth message may be a registration message, used to request registration of its data transfer capability information, such as through an Nnrf_NFManagement_NFRegister Request message. Alternatively, the twelfth message may be used to update the capability information of the intelligent network element.

[0370] When the third communication device sends the twelfth message to the fourth communication device, in addition to carrying its own identification information (network element instance identification information, FQDN or address information), NF type, network element type and other information, it can also carry network element capability information, etc.

[0371] Optionally, the network element capability information includes at least one of the following:

[0372] Identification information of supported data analysis tasks;

[0373] Identification information of the owned model;

[0374] An indication that the model owned is a VFL model;

[0375] Indication that the owned model cannot be shared;

[0376] Device information of the model owned;

[0377] Supported inference precision;

[0378] Supported inference latency.

[0379] The supported data analysis task identification information (analytics ID) is used to indicate the supported task category. The owned model identification information is used to indicate the owned model, which can correspond to the data analysis task identification information, that is, support a certain data analysis task.

[0380] The information indicating that the model is a VFL model indicates that the model is a VFL model, that is, a model generated through multi-party VFL training. It can also be understood that the model corresponding to the identification information of the supported data analysis task is a VFL model. Therefore, it may not be shared due to privacy issues. The information indicating that the model cannot be shared indicates that the model cannot be shared, such as the model is a VFL model, generated through VFL training, the model cannot be used alone, or it may involve privacy issues and cannot be shared.

[0381] The device information of the owned model is used to indicate the VFL server information of the VFL model (i.e., information of the third communication device), such as the identification information, instance identifier, FQDN, etc. of the third communication device; other communication devices can find the third communication device through the device information (VFL server information) and send a data analysis request to the third communication device, requesting the model output result (i.e., data analysis result). Optionally, the device information of the owned model may include: at least one of the identification information, instance identifier, network element domain name, etc. of the third communication device. Optionally, the device information of the owned model may also include at least one of the following: capability information of the third communication device (e.g., capability information related to VFL), shareable operation indication information of the third communication device (or interoperability indication information, such as an interoperable vendor ID), etc.

[0382] The supported inference precision is used to indicate the precision that can be achieved by the owned model, which may be the accuracy of the model, etc. The supported inference delay is used to indicate the time information required for the third communication device to use the model for inference.

[0383] It should be noted that the owned model can also be understood as the available model.

[0384] It should be noted that the inference accuracy and inference delay can also be understood as information such as the accuracy that can be achieved when the third communication device performs inference or the time required.

[0385] It should be noted that precision may be, for example, accuracy, recall rate, mean square error (MSE), etc.

[0386] It should be noted that at least one of the aforementioned model identification information, model device information, inference accuracy, and inference latency can be based on the granularity of the data analysis task. That is, different data analysis tasks correspond to different model identification information, model device information, inference accuracy, and inference latency.

[0387] The third communication device registers or updates the network element capability information with the fourth communication device, so that other communication devices can find the information and perform corresponding VFL inference.

[0388] FIG10 is an interactive flow chart of a communication data processing method provided in an embodiment of the present application. As shown in FIG10 , the communication data processing method includes the following steps:

[0389] Step 1001: The second communication device sends a sixth message to the third communication device.

[0390] The sixth message is used to request VFL reasoning for the data analysis task.

[0391] Step 1002: The third communication device determines to perform VFL inference and determines auxiliary information and task information.

[0392] Step 1003: The third communication device requests the fifth communication device to obtain identification information of the inference object in the first target format.

[0393] The first target format is a data format of the sixth communication device.

[0394] The third communication device requests the fifth communication device to obtain the identification information of the inference object in the first target format, that is, the third communication device in the above embodiment sends the tenth message to the fifth communication device, and the tenth message is used to request that the identification information of the inference object be converted into the data format of the sixth communication device, and receive the identification information of the converted inference object from the fifth communication device.

[0395] Step 1004: The third communication device sends the eighth message to the sixth communication device, or the third communication device sends the eighth message to the sixth communication device through the fifth communication device.

[0396] It should be noted that step 1003 is an optional step, that is, the third communication device can actively obtain the inference object identification information in the data format of the sixth communication device from the fifth communication device, or the third communication device can send the eighth message to the sixth communication device through the fifth communication device. When the fifth communication device forwards the eighth message, if it determines that the identification information of the inference object therein is inconsistent with the data format of the sixth communication device, it will actively convert the identification information of the inference object into the data format of the sixth communication device, and forward the converted identification information of the inference object and the eighth message to the sixth communication device.

[0397] Step 1005: The sixth communication device requests the fifth communication device to obtain identification information of the inference object in the first target format.

[0398] The first target format is a data format of the sixth communication device.

[0399] It should be noted that step 1005 is an optional step. If the identification information of the inference object in the eighth message is not in the data format of the sixth communication device, that is, the data format conversion during forwarding by the fifth communication device in steps 1003 or 1004 is not performed, the sixth communication device may request the fifth communication device to convert the identification information of the inference object into the data format of the sixth communication device.

[0400] Step 1006: The sixth communication device determines a model according to the relevant identifier of the VFL, performs model inference, and obtains an intermediate inference result.

[0401] Step 1007: The sixth communication device requests the fifth communication device to obtain identification information of the inference object in the second target format.

[0402] The second target format is a data format of the third communication device.

[0403] Optionally, after completing the reasoning and obtaining the intermediate reasoning result, the sixth communication device can request the fifth communication device to convert the identification information of the reasoning object into the identification information of the reasoning object in the second target format, or can directly request to obtain the identification information of the reasoning object in the second target format.

[0404] Step 1008: The sixth communication device sends a sixth feedback message to the third communication device.

[0405] The sixth communication device may send a sixth feedback message to the third communication device via the fifth communication device. Optionally, when forwarding the sixth feedback message, if the fifth communication device finds that the identification information of the inference object in the sixth feedback message is inconsistent with the data format of the third communication device, the fifth communication device may convert the identification information of the inference object into the data format of the third communication device before forwarding the message.

[0406] Step 1009: The third communication device requests the fifth communication device to obtain identification information of the inference object in the second target format.

[0407] The second target format is a data format of the third communication device.

[0408] Optionally, after the third communication device receives the intermediate inference result from the sixth communication device, if it cannot identify or use the identification information of the inference object therein, it can send an eleventh message to the fifth communication device to request that the identification information of the inference object in the intermediate inference result be converted into the data format of the third communication device.

[0409] Step 1010: The third communication device determines a data analysis result based on the intermediate inference result.

[0410] Step 1011: The third communication device sends the data analysis result to the second communication device.

[0411] The specific implementation process of each step in this embodiment can refer to the above embodiment and will not be repeated here.

[0412] The communication data processing method provided in the embodiment of the present application implements VFL reasoning based on distributed processing of the third communication device and each sixth communication device.

[0413] FIG11 is an interactive flow chart of another communication data processing method provided in an embodiment of the present application. As shown in FIG11 , the communication data processing method includes the following steps:

[0414] Step 1101: The second communication device sends a network element search request to the fourth communication device.

[0415] Step 1102: The fourth communication device feeds back information of the first communication device to the second communication device.

[0416] Step 1103: The second communication device sends a first message to the first communication device.

[0417] Step 1104: The first communication device sends a third message to the fourth communication device.

[0418] Step 1105: The first communication device receives a second feedback message from the fourth communication device.

[0419] Step 1106: The first communication device sends a second message to the third communication device.

[0420] Step 1107: The third communication device interacts with each sixth communication device to complete VFL reasoning and obtain data analysis results.

[0421] Step 1108: The third communication device feeds back the data analysis result to the first communication device.

[0422] Step 1109: The first communication device sends a fifth feedback message to the second communication device.

[0423] The specific implementation process of the above steps can be referred to the above embodiments and will not be repeated here.

[0424] It should be noted that, before step 1106, the first communication device may request the fifth communication device to convert the identification information of the inference object into the data format of the third communication device. Alternatively, the first communication device may not convert the identification information of the inference object, but instead execute step 1106. Before step 1107, the third communication device may request the fifth communication device to convert the identification information of the inference object therein into the data format of the third communication device. Before step 1108, the third communication device may request the fifth communication device to convert the identification information of the inference object in the data analysis results into the data format of the first communication device.

[0425] The data processing method provided in the embodiment of the present application is implemented by a first communication device interacting with a third communication device based on a first message from a second communication device, and the third communication device completes VFL inference, thereby realizing distributed VFL inference participated by various communication devices and obtaining data analysis results.

[0426] FIG12 is an interactive flow chart of another communication data processing method provided in an embodiment of the present application. As shown in FIG12 , the communication data processing method includes the following steps:

[0427] Step 1201: The second communication device sends a network element search request to the fourth communication device.

[0428] Step 1202: The fourth communication device feeds back information of the first communication device to the second communication device.

[0429] Step 1203: The second communication device sends a first message to the first communication device.

[0430] Step 1204: The first communication device sends a third message to the fourth communication device.

[0431] Step 1205: The first communication device receives a second feedback message from the fourth communication device.

[0432] Step 1206: The first communication device sends a first feedback message to the second communication device.

[0433] Step 1207: The second communication device sends a sixth message to the third communication device.

[0434] Step 1208: The third communication device interacts with each sixth communication device to complete VFL reasoning and obtain data analysis results.

[0435] Step 1209: The third communication device feeds back the data analysis result to the second communication device.

[0436] The specific implementation process of the above steps can be referred to the above embodiments and will not be repeated here.

[0437] It should be noted that, before step 1206, the first communication device may request the fifth communication device to convert the identification information of the inference object into the data format of the second communication device. Before step 1207, the second communication device may request the fifth communication device to convert the identification information of the inference object therein into the data format of the third communication device. Before step 1209, the third communication device may request the fifth communication device to convert the identification information of the inference object in the data analysis results into the data format of the second communication device.

[0438] The data processing method provided in the embodiment of the present application sends a first message to a first communication device through a second communication device, the first communication device feeds back information of a third communication device through a first feedback message, the second communication device interacts with the third communication device, and the third communication device completes VFL inference, thereby realizing distributed VFL inference involving various communication devices and obtaining data analysis results.

[0439] FIG13 is an interactive flow chart of another communication data processing method provided in an embodiment of the present application. As shown in FIG13 , the communication data processing method includes the following steps:

[0440] Step 1301: The second communication device sends a network element search request to the fourth communication device.

[0441] Step 1302: The fourth communication device feeds back information of the first communication device to the second communication device.

[0442] Step 1303: The second communication device sends a first message to the first communication device.

[0443] Step 1304: The first communication device sends a third message to the fourth communication device.

[0444] Step 1305: The first communication device receives a second feedback message from the fourth communication device.

[0445] Step 1306: The first communication device sends a fourth message to the third communication device.

[0446] Step 1307: The first communication device receives a fourth feedback message from the third communication device.

[0447] Step 1308: The first communication device sends a second message to the third communication device.

[0448] Step 1309 : The third communication device interacts with each sixth communication device to complete VFL reasoning and obtain data analysis results.

[0449] Step 1310: The third communication device feeds back the data analysis result to the first communication device.

[0450] Step 1311: The first communication device sends a fifth feedback message to the second communication device.

[0451] The specific implementation process of the above steps can be referred to the above embodiments and will not be repeated here.

[0452] It should be noted that, before step 1306, the first communication device may request the fifth communication device to convert the identification information of the inference object into the data format of the third communication device. Before step 1310, the third communication device may request the fifth communication device to convert the identification information of the inference object in the data analysis result into the data format of the first communication device.

[0453] The data processing method provided in the embodiment of the present application is to obtain a model from a third communication device through a first communication device based on a first message from a second communication device. When the model cannot be obtained, the first communication device interacts with the third communication device, and the third communication device completes VFL inference, thereby realizing distributed VFL inference involving various communication devices and obtaining data analysis results.

[0454] FIG14 is an interactive flow chart of another communication data processing method provided in an embodiment of the present application. As shown in FIG14 , the communication data processing method includes the following steps:

[0455] Step 1401: The second communication device sends a network element search request to the fourth communication device.

[0456] Step 1402: The fourth communication device feeds back information of the first communication device to the second communication device.

[0457] Step 1403: The second communication device sends a first message to the first communication device.

[0458] Step 1404: The first communication device sends a third message to the fourth communication device.

[0459] Step 1405: The first communication device receives a second feedback message from the fourth communication device.

[0460] Step 1406: The first communication device sends a fourth message to the third communication device.

[0461] Step 1407: The first communication device receives a fourth feedback message from the third communication device.

[0462] Step 1408: The first communication device sends a first feedback message to the second communication device.

[0463] Step 1409: The second communication device sends a sixth message to the third communication device.

[0464] Step 1410: The third communication device interacts with each sixth communication device to complete VFL reasoning and obtain data analysis results.

[0465] Step 1411: The third communication device feeds back the data analysis result to the second communication device.

[0466] The specific implementation process of the above steps can be referred to the above embodiments and will not be repeated here.

[0467] It should be noted that, before step 1406, the first communication device may request the fifth communication device to convert the identification information of the inference object into the data format of the third communication device. Before step 1109, the second communication device may request the fifth communication device to convert the identification information of the inference object therein into the data format of the third communication device. Before step 1411, the third communication device may request the fifth communication device to convert the identification information of the inference object in the data analysis results into the data format of the second communication device.

[0468] The data processing method provided in the embodiment of the present application sends a first message to a first communication device through a second communication device, and the first communication device obtains a model from a third communication device. When the model cannot be obtained, the first communication device feeds back information of the third communication device to the second communication device through a first feedback message. The second communication device interacts with the third communication device, and the third communication device completes VFL inference, thereby realizing distributed VFL inference involving various communication devices and obtaining data analysis results.

[0469] The communication data processing method provided in the embodiment of the present application can be executed by a communication data processing device. In the embodiment of the present application, the communication data processing device provided in the embodiment of the present application is described by taking the communication data processing device executing the communication data processing method as an example.

[0470] FIG15 is a block diagram of a communication data processing apparatus provided in an embodiment of the present application. The communication data processing apparatus is applied to a first communication device. As shown in FIG15 , the communication data processing apparatus 1500 includes:

[0471] A first message receiving module 1501 is configured to receive a first message from a second communication device, where the first message is used to request obtaining a data analysis result corresponding to a data analysis task;

[0472] The device information acquisition module 1502 is configured to acquire information of a third communication device having a capability related to vertical federated learning (VFL);

[0473] A second message sending module 1503 is configured to send a second message to the third communication device based on the information of the third communication device, where the second message is used to request VFL reasoning for the data analysis task, or

[0474] The first feedback message sending module 1504 is configured to send a first feedback message to the second communication device, where the first feedback message is used to instruct the second communication device to request the third communication device to perform VFL inference.

[0475] Optionally, the device further comprises:

[0476] A third message sending module, configured to send a third message to a fourth communication device;

[0477] The third message includes at least one of the following information:

[0478] Type information of the communication device being searched, where the type information is used to indicate the type of communication device being searched;

[0479] Target area of ​​interest information, where the target area of ​​interest information is used to indicate a communication device that searches for a service target area;

[0480] Indication information indicating that the communication device being searched has VFL server capability, wherein the indication information indicating that the communication device has VFL server capability is used to indicate that a communication device capable of providing VFL service is to be searched;

[0481] Identification information of the data analysis task that the communication device being searched needs to support;

[0482] The communication device being searched needs to support providing indication information of data analysis results or inference results;

[0483] The inference accuracy that the communication device being searched needs to support;

[0484] The inference latency that the communication device needs to support.

[0485] Optionally, the device further comprises:

[0486] A second feedback message receiving module, configured to receive a second feedback message from the fourth communication device;

[0487] The second feedback message includes at least one of the following:

[0488] Look for failed indications;

[0489] VFL model indication information, where the VFL model indication information is used to indicate that the model supported by the communication device being searched is a VFL model;

[0490] Indication information indicating that sharing is impossible, the indication information indicating that sharing is impossible is used to indicate that the model supported by the communication device being searched cannot be shared;

[0491] Information of the third communication device.

[0492] Optionally, the device further comprises:

[0493] A third feedback message receiving module, configured to receive a third feedback message from the fourth communication device;

[0494] The third feedback message includes at least one of the following:

[0495] Find the successful indication information;

[0496] Information of a target communication device, where the target communication device is a communication device that satisfies the information in the third message;

[0497] Optionally, the device further comprises:

[0498] The fourth message sending module is used to send a fourth message to the target communication device, and the fourth message is used to request to obtain the model corresponding to the data analysis task.

[0499] Optionally, the device further comprises:

[0500] a fourth feedback message receiving module, configured to receive a fourth feedback message from the target communication device;

[0501] The fourth feedback message includes at least one of the following:

[0502] Indication information that a model cannot be provided, wherein the indication information that a model cannot be provided is used to indicate that a model corresponding to the data analysis task cannot be provided, or that there is no model corresponding to the data analysis task;

[0503] Indication information of a VFL model, wherein the indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model;

[0504] A related identifier of a VFL model, wherein the related identifier of the VFL model is used to indicate identification information of a related model of the data analysis task;

[0505] Information of the third communication device.

[0506] Optionally, the device information acquisition module is specifically configured to:

[0507] The information of the third communication device is determined according to internal logic or pre-configuration.

[0508] Optionally, the second message includes at least one of the following:

[0509] Relevant information about the data analysis task;

[0510] Identification information of the inference object, where the identification information of the inference object is used to indicate the inference target;

[0511] A VFL model related identifier, wherein the VFL model related identifier is used to indicate identification information of the data analysis task related model;

[0512] The first indication information of VFL inference is used to indicate a request for VFL inference.

[0513] Optionally, the device further comprises:

[0514] The first result receiving module is used to receive the data analysis result from the third communication device.

[0515] Optionally, the device further comprises:

[0516] a fifth feedback message sending module, configured to send a fifth feedback message to the second communication device;

[0517] The fifth feedback message includes at least one of the following:

[0518] the data analysis results;

[0519] information of a third communication device;

[0520] Indication information of data analysis failure, wherein the indication information of data analysis failure is used to indicate that the data analysis task has failed, or that the model corresponding to the data analysis task has not been obtained;

[0521] The indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model.

[0522] Optionally, the device further comprises:

[0523] a fifth message sending module, configured to send a fifth message to a fifth communication device, wherein the fifth message is used to request that the identification information of the inference object be converted into a data format of the third communication device;

[0524] The first converted identification receiving module is configured to receive identification information of the converted inference object from the fifth communication device.

[0525] Optionally, the first feedback message includes at least one of the following:

[0526] Indication information of data analysis failure, wherein the indication information of data analysis failure is used to indicate that the data analysis task has failed, or that the model corresponding to the data analysis task has not been obtained;

[0527] Indication information of a VFL model, wherein the indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model;

[0528] Information of the third communication device.

[0529] Optionally, the first message includes at least one of the following:

[0530] Relevant information about the data analysis task;

[0531] Identification information of the inference object;

[0532] The required supported inference accuracy;

[0533] The required inference latency.

[0534] Optionally, the second message sending module is specifically configured to:

[0535] If it is determined that the third communication device cannot provide the model, sending the second message to the third communication device;

[0536] or,

[0537] The first feedback message sending module is specifically configured to:

[0538] In the communication data processing device provided by the embodiment of the present application, after the first communication device receives a first message from the second communication device, it obtains information of a third communication device with VFL-related capabilities. Based on the information of the third communication device, the first communication device sends a second message to the third communication device to request the third communication device to perform VFL reasoning on the data analysis task, or the first communication device sends a first feedback message to the second communication device, and the first feedback message is used to instruct the second communication device to request the third communication device to perform VFL reasoning. In this way, the first communication device does not need to obtain the VFL model, but requests the third communication device to perform VFL reasoning, or the second communication device requests the third communication device to perform VFL reasoning, thereby realizing the reasoning of the VFL model.

[0539] The model training device in the embodiments of the present application can be an electronic device, such as an electronic device with an operating system, or a component of an electronic device, such as an integrated circuit or chip. The electronic device can be a network-side device. For example, the network-side device can include, but is not limited to, the types of network-side devices 12 listed above.

[0540] The communication data processing device provided in the embodiment of the present application can implement the various processes implemented in the method embodiment of Figure 7 and achieve the same technical effects. To avoid repetition, they will not be described here.

[0541] FIG16 is a block diagram of another communication data processing apparatus provided in an embodiment of the present application. The communication data processing apparatus is applied to a second communication device. As shown in FIG16 , the communication data processing apparatus 1600 includes:

[0542] A first message sending module 1601 is configured to send a first message to a first communication device, where the first message is used to request acquisition of a data analysis result corresponding to a data analysis task;

[0543] A fifth feedback message receiving module 1602 is configured to receive a fifth feedback message from the first communication device, where the fifth feedback message indicates a result of performing VFL reasoning on the data analysis task, or

[0544] The first feedback message receiving module 1603 is used to receive a first feedback message from the first communication device, and send a sixth message to the third communication device based on the first feedback message, wherein the sixth message is used to request VFL reasoning for the data analysis task, and the third communication device has VFL-related capabilities.

[0545] Optionally, the fifth feedback message includes at least one of the following:

[0546] The data analysis results corresponding to the data analysis task;

[0547] information of a third communication device;

[0548] Indication information of data analysis failure, wherein the indication information of data analysis failure is used to indicate that the data analysis task has failed, or that the model corresponding to the data analysis task has not been obtained;

[0549] The indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model.

[0550] Optionally, the sixth message includes at least one of the following:

[0551] Relevant information about the data analysis task;

[0552] Identification information of the inference object, where the identification information of the inference object is used to indicate the inference target;

[0553] A VFL model related identifier, wherein the VFL model related identifier is used to indicate identification information of the data analysis task related model;

[0554] The first indication information of VFL inference is used to indicate a request for VFL inference.

[0555] Optionally, the device further comprises:

[0556] a seventh message sending module, configured to send a seventh message to the fifth communication device, wherein the seventh message is used to request that the identification information of the inference object be converted into a data format of the third communication device;

[0557] The second converted identification receiving module is configured to receive identification information of the converted inference object from the fifth communication device.

[0558] Optionally, the first feedback message includes at least one of the following:

[0559] Indication information of data analysis failure, wherein the indication information of data analysis failure is used to indicate that the data analysis task has failed, or that the model corresponding to the data analysis task has not been obtained;

[0560] Indication information of a VFL model, wherein the indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model;

[0561] Information of the third communication device.

[0562] Optionally, the device further comprises:

[0563] The second result receiving module is used to receive the data analysis result from the third communication device.

[0564] Optionally, the first message includes at least one of the following:

[0565] Relevant information about the data analysis task;

[0566] Identification information of the inference object;

[0567] The required supported inference accuracy;

[0568] The required inference latency.

[0569] In the communication data processing device provided by the embodiment of the present application, the second communication device sends a first message to the first communication device, and the first message is used to request to obtain the data analysis result corresponding to the data analysis task; the second communication device receives a fifth feedback message from the first communication device, and the fifth feedback message is used to indicate the result of VFL reasoning on the data analysis task, or the second communication device receives the first feedback message from the first communication device, and sends a sixth message to the third communication device based on the first feedback message, and the sixth message is used to request VFL reasoning on the data analysis task, so that the second communication device can directly request the third communication device to perform VFL reasoning, so that the reasoning of the VFL model can be completed without obtaining the model, and no data transmission is required through VFL reasoning, thereby ensuring the security of the data.

[0570] The model training device in the embodiments of the present application can be an electronic device, such as an electronic device with an operating system, or a component of an electronic device, such as an integrated circuit or chip. The electronic device can be a network-side device. For example, the network-side device can include, but is not limited to, the types of network-side devices 12 listed above.

[0571] The communication data processing device provided in the embodiment of the present application can implement the various processes implemented in the method embodiment of Figure 8 and achieve the same technical effects. To avoid repetition, it will not be repeated here.

[0572] FIG17 is a block diagram of another communication data processing apparatus provided in an embodiment of the present application. The communication data processing apparatus is applied to a third communication device. As shown in FIG17 , the communication data processing apparatus 1700 includes:

[0573] A message receiving module 1701 is configured to receive a second message from a first communication device or a sixth message from a second communication device, wherein the second message or the sixth message is used to request VFL reasoning for a data analysis task;

[0574] A data analysis module 1702 is configured to determine a data analysis result corresponding to the data analysis task;

[0575] The result feedback module 1703 is configured to send the data analysis result to the first communication device or the second communication device.

[0576] Optionally, the second message or the sixth message includes at least one of the following:

[0577] Relevant information about the data analysis task;

[0578] Identification information of the inference object, where the identification information of the inference object is used to indicate the inference target;

[0579] A VFL model related identifier, wherein the VFL model related identifier is used to indicate identification information of the data analysis task related model;

[0580] The first indication information of VFL inference is used to indicate a request for VFL inference.

[0581] Optionally, the device further comprises:

[0582] An eighth message sending module is used to send an eighth message to a sixth communication device, wherein the eighth message includes auxiliary information or task information, wherein the auxiliary information is used to indicate a method for determining input data or a feedback time for intermediate reasoning results, and the task information is used to indicate information about the data analysis task.

[0583] Optionally, the device further comprises:

[0584] The sixth feedback message receiving module is configured to receive a sixth feedback message from the sixth communication device, where the sixth feedback message includes an intermediate reasoning result.

[0585] Optionally, the data analysis module is specifically used to:

[0586] A data analysis result is determined based on the intermediate inference result.

[0587] Optionally, the auxiliary information includes at least one of the following:

[0588] Time information of input data required when performing the VFL inference;

[0589] The data type of the input data required when performing the VFL inference;

[0590] Limitation information of input data required when performing the VFL inference;

[0591] The maximum response time required to feed back the intermediate inference result when executing the VFL inference.

[0592] Optionally, the task information includes at least one of the following:

[0593] identification information of the data analysis task;

[0594] A VFL model related identifier, wherein the VFL model related identifier is used to indicate identification information of the data analysis task related model;

[0595] The identification information of the inference object is used to indicate the inference object identification information corresponding to the input data.

[0596] Optionally, the sixth feedback message further includes at least one of the following:

[0597] Identification information of the inference object, where the identification information of the inference object is used to indicate the inference target;

[0598] second indication information of VFL reasoning, wherein the second indication information of VFL reasoning is used to indicate that the intermediate reasoning result is obtained through VFL reasoning;

[0599] The VFL model related identifier is used to indicate the identification information of the model related to the data analysis task;

[0600] Identification information of the data analysis task.

[0601] Optionally, the eighth message sending module is specifically configured to:

[0602] A ninth message is sent to the fifth communication device, where the ninth message is used to request that the eighth message be forwarded to the sixth communication device.

[0603] Optionally, the device further comprises:

[0604] a tenth message sending module, configured to send a tenth message to the fifth communication device, wherein the tenth message is used to request that the identification information of the inference object be converted into a data format of the sixth communication device;

[0605] The third converted identification receiving module is configured to receive identification information of the converted inference object from the fifth communication device.

[0606] Optionally, the device further comprises:

[0607] an eleventh message sending module, configured to send an eleventh message to the fifth communication device, wherein the eleventh message is used to request that identification information of the inference object in the intermediate inference result be converted into a data format of the third communication device;

[0608] The fourth converted identification receiving module is used to receive the identification information of the converted inference object from the fifth communication device.

[0609] Optionally, the device further comprises:

[0610] The fourth message receiving module is used to receive a fourth message from the first communication device, where the fourth message is used to request to obtain the model corresponding to the data analysis task.

[0611] Optionally, the device further comprises:

[0612] a fourth feedback message sending module, configured to send a fourth feedback message to the first communication device;

[0613] The fourth feedback message includes at least one of the following:

[0614] Indication information that a model cannot be provided, wherein the indication information that a model cannot be provided is used to indicate that a model corresponding to the data analysis task cannot be provided, or that there is no model corresponding to the data analysis task;

[0615] Indication information of a VFL model, wherein the indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model;

[0616] A related identifier of a VFL model, wherein the related identifier of the VFL model is used to indicate identification information of a related model of the data analysis task;

[0617] Information of the third communication device.

[0618] Optionally, the device further comprises:

[0619] The twelfth message sending module is configured to send a twelfth message to the fourth communication device, where the twelfth message is used to request registration or update of the network element capability information of the third communication device.

[0620] Optionally, the network element capability information includes at least one of the following:

[0621] Identification information of supported data analysis tasks;

[0622] Identification information of the owned model;

[0623] The model owned is an indication of a VFL model;

[0624] Indication that the owned model cannot be shared;

[0625] Device information of the model owned;

[0626] Supported inference precision;

[0627] Supported inference latency.

[0628] In the communication data processing device provided in the embodiment of the present application, after the third communication device receives the second message from the first communication device or the sixth message from the second communication device, it obtains the data analysis result corresponding to the data analysis task and sends the data analysis result to the first communication device or the second communication device. The third communication device does not need to provide a model, but instead performs VFL reasoning and sends the obtained data analysis result to the first communication device or the second communication device, thereby realizing VFL reasoning.

[0629] The model training device in the embodiments of the present application can be an electronic device, such as an electronic device with an operating system, or a component of an electronic device, such as an integrated circuit or chip. The electronic device can be a network-side device. For example, the network-side device can include, but is not limited to, the types of network-side devices 12 listed above.

[0630] The communication data processing device provided in the embodiment of the present application can implement the various processes implemented in the method embodiment of Figure 9 and achieve the same technical effects. To avoid repetition, they will not be described here.

[0631] As shown in Figure 18, an embodiment of the present application also provides a communication device 1800, including a processor 1801 and a memory 1802, and the memory 1802 stores programs or instructions that can be run on the processor 1801. For example, when the communication device 1800 is a network side device, the program or instruction is executed by the processor 1801 to implement the various steps of the above-mentioned communication data processing method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0632] It can be understood that the implementation process of each implementation method mentioned in this embodiment can refer to the relevant description of the method embodiment and achieve the same or corresponding technical effects. To avoid repetition, it will not be described here.

[0633] The present application also provides a network-side device, including a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is configured to execute a program or instruction to implement the steps of the method embodiment shown in any one of Figures 7 to 9. This network-side device embodiment corresponds to the aforementioned network-side device method embodiment, and each implementation process and implementation method of the aforementioned method embodiment is applicable to this network-side device embodiment and can achieve the same technical effects.

[0634] Specifically, an embodiment of the present application also provides a network-side device. As shown in Figure 19, the network-side device 1900 includes: an antenna 1901, a radio frequency device 1902, a baseband device 1903, a processor 1904, and a memory 1905. Antenna 1901 is connected to radio frequency device 1902. In the uplink direction, radio frequency device 1902 receives information via antenna 1901 and sends the received information to baseband device 1903 for processing. In the downlink direction, baseband device 1903 processes the information to be transmitted and sends it to radio frequency device 1902. Radio frequency device 1902 processes the received information and sends it through antenna 1901.

[0635] The method executed by the network-side device in the above embodiment may be implemented in the baseband device 1903 , which includes a baseband processor.

[0636] The baseband device 1903 may, for example, include at least one baseband board, on which multiple chips are arranged, as shown in Figure 19, one of the chips is, for example, a baseband processor, which is connected to the memory 1905 through a bus interface to call the program in the memory 1905 and execute the network device operations shown in the above method embodiment.

[0637] The network side device may further include a network interface 1906 , which is, for example, a Common Public Radio Interface (CPRI).

[0638] Specifically, the network side device 1900 of an embodiment of the present invention also includes: instructions or programs stored in the memory 1905 and executable on the processor 1904. The processor 1904 calls the instructions or programs in the memory 1905 to execute the method executed by any module shown in any one of Figures 15 to 17 and achieve the same technical effect. To avoid repetition, they will not be elaborated here.

[0639] Specifically, the embodiment of the present application further provides a network side device. As shown in FIG20 , the network side device 2000 includes: a processor 2001, a network interface 2002, and a memory 2003. The network interface 2002 is, for example, a common public radio interface (CPRI).

[0640] Specifically, the network side device 2000 of an embodiment of the present invention also includes: instructions or programs stored in the memory 2003 and executable on the processor 2001. The processor 2001 calls the instructions or programs in the memory 2003 to execute the method executed by any module shown in Figures 15 to 17 and achieve the same technical effect. To avoid repetition, it will not be elaborated here.

[0641] An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the various processes of the above-mentioned communication data processing method embodiment are implemented and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.

[0642] The processor is the processor in the network-side device described in 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. In some examples, the readable storage medium may be a non-transitory readable storage medium.

[0643] An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run programs or instructions to implement the various processes of the above-mentioned communication data processing method embodiment, and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0644] It should be understood that the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.

[0645] An embodiment of the present application further provides a computer program / program product, which is stored in a storage medium. The computer program / program product is executed by at least one processor to implement the various processes of the above-mentioned communication data processing method embodiment and can achieve the same technical effect. To avoid repetition, it will not be repeated here.

[0646] An embodiment of the present application also provides a communication system, including: a first communication device, a second communication device, and a third communication device, wherein the first communication device can be used to execute the steps of the communication data processing method described in Figure 7 above, the second communication device can be used to execute the steps of the communication data processing method described in Figure 8 above, and the third communication device can be used to execute the steps of the communication data processing method described in Figure 9 above.

[0647] It should be noted that, in this article, the terms "comprise", "include" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the presence of other identical elements in the process, method, article or device comprising the element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, and may also include performing functions in a substantially simultaneous manner or in the opposite order according to the functions involved. For example, the described method may be performed in an order different from that described, and various steps may also be added, omitted or combined. In addition, the features described with reference to certain examples may be combined in other examples.

[0648] Through the description of the above embodiments, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of a computer software product plus a necessary general-purpose hardware platform, or of course, by hardware. The computer software product is stored in a storage medium (such as ROM, RAM, magnetic disk, optical disk, etc.) and includes a number of instructions for enabling a terminal or network-side device to execute the methods described in each embodiment of the present application.

[0649] The embodiments of the present application are described above in conjunction with the accompanying drawings, but the present application is not limited to the above-mentioned specific implementation methods. The above-mentioned specific implementation methods are merely illustrative and not restrictive. Under the guidance of this application, ordinary technicians in this field can also make many forms of implementation methods without departing from the purpose of this application and the scope of protection of the claims. These implementation methods are all within the protection of this application.

Claims

1. A communication data processing method, wherein: include: The first communication device receives a first message from the second communication device, where the first message is used to request obtaining a data analysis result corresponding to the data analysis task; The first communication device obtains information of a third communication device having a capability related to vertical federated learning (VFL); The first communication device sends a second message to the third communication device based on the information of the third communication device, where the second message is used to request VFL reasoning for the data analysis task, or The first communication device sends a first feedback message to the second communication device, where the first feedback message is used to instruct the second communication device to request the third communication device to perform VFL inference.

2. The method according to claim 1, wherein Before the first communication device acquires information of a third communication device having VFL-related capabilities, the method further includes: The first communication device sends a third message to the fourth communication device; The third message includes at least one of the following information: Type information of the communication device being searched, where the type information is used to indicate the type of communication device being searched; Target area of ​​interest information, where the target area of ​​interest information is used to indicate a communication device that searches for a service target area; Indication information indicating that the communication device being searched has VFL server capability, wherein the indication information indicating that the communication device has VFL server capability is used to indicate that a communication device capable of providing VFL service is to be searched; Identification information of the data analysis task that the communication device being searched needs to support; The communication device being searched needs to support providing indication information of data analysis results or inference results; The inference accuracy that the communication device being searched needs to support; The inference latency that the communication device needs to support.

3. The method according to claim 2, wherein: After the first communication device sends the third message to the fourth communication device, the method further includes: The first communication device receives a second feedback message from the fourth communication device; The second feedback message includes at least one of the following: Look for failed indications; VFL model indication information, where the VFL model indication information is used to indicate that the model supported by the communication device being searched is a VFL model; Indication information indicating that sharing is impossible, the indication information indicating that sharing is impossible is used to indicate that the model supported by the communication device being searched cannot be shared; Information of the third communication device.

4. [Corrected 09.04.2025 under Rule 91] A method according to claim 2, wherein After the first communication device sends the third message to the fourth communication device, the method further includes: The first communication device receives a third feedback message from the fourth communication device; The third feedback message includes at least one of the following: Find the successful indication information; Information of a target communication device, where the target communication device is a communication device that meets the information in the third message.

5. The method according to claim 4, wherein After the first communication device receives the third feedback message from the fourth communication device, the method further includes: The first communication device sends a fourth message to the target communication device, where the fourth message is used to request acquisition of a model corresponding to the data analysis task.

6. The method according to claim 5, wherein: After the first communication device sends the fourth message to the target communication device, the method further includes: The first communication device receives a fourth feedback message from the target communication device; The fourth feedback message includes at least one of the following: Indication information that a model cannot be provided, wherein the indication information that a model cannot be provided is used to indicate that a model corresponding to the data analysis task cannot be provided; Indication information of a VFL model, wherein the indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model; A related identifier of a VFL model, wherein the related identifier of the VFL model is used to indicate identification information of a related model of the data analysis task; Information of the third communication device.

7. The method according to claim 1, wherein The first communication device acquires information of a third communication device having VFL-related capabilities, including: The first communication device determines the information of the third communication device according to internal logic or pre-configuration.

8. The method according to any one of claims 1 to 7, wherein: The second message includes at least one of the following: Relevant information about the data analysis task; Identification information of the inference object, where the identification information of the inference object is used to indicate the inference target; A VFL model related identifier, wherein the VFL model related identifier is used to indicate identification information of the data analysis task related model; The first indication information of VFL inference is used to indicate a request for VFL inference.

9. The method according to any one of claims 1 to 8, wherein: After the first communication device sends a second message to the third communication device based on the information of the third communication device, the method further includes: The first communication device receives a data analysis result from the third communication device.

10. The method according to claim 9, wherein: After the first communication device receives the data analysis result from the third communication device, the method further includes: The first communication device sends a fifth feedback message to the second communication device; The fifth feedback message includes at least one of the following: the data analysis results; information of a third communication device; The data analysis failure indication information is used to indicate the failure of the data analysis task; The indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model.

11. The method according to any one of claims 1 to 10, wherein: Before the first communication device sends the second message to the third communication device, the method further includes: The first communication device sends a fifth message to the fifth communication device, where the fifth message is used to request that the identification information of the inference object be converted into a data format of the third communication device; The first communication device receives identification information of the converted inference object from the fifth communication device.

12. The method according to any one of claims 1 to 7, wherein: The first feedback message includes at least one of the following: The data analysis failure indication information is used to indicate the failure of the data analysis task; Indication information of a VFL model, wherein the indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model; Information of the third communication device.

13. The method according to any one of claims 1 to 12, wherein: The first message includes at least one of the following: Relevant information about the data analysis task; Identification information of the inference object; The required supported inference accuracy; The required inference latency.

14. The method according to any one of claims 1 to 13, wherein: The first communication device sending a second message to the third communication device based on the information of the third communication device includes: The first communication device sends the second message to the third communication device when determining that the third communication device cannot provide the model; or, The first communication device sending a first feedback message to the second communication device includes: When determining that the third communication device cannot provide the model, the first communication device sends the first feedback message to the second communication device.

15. A communication data processing method, wherein: include: The second communication device sends a first message to the first communication device, where the first message is used to request obtaining a data analysis result corresponding to the data analysis task; The second communication device receives a fifth feedback message from the first communication device, where the fifth feedback message is used to indicate a result of performing VFL reasoning on the data analysis task, or The second communication device receives a first feedback message from the first communication device, and sends a sixth message to the third communication device based on the first feedback message, wherein the sixth message is used to request VFL reasoning for the data analysis task, and the third communication device has VFL-related capabilities.

16. The method according to claim 15, wherein The fifth feedback message includes at least one of the following: The data analysis results corresponding to the data analysis task; information of a third communication device; The data analysis failure indication information is used to indicate the failure of the data analysis task; The indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model.

17. The method according to claim 15, wherein: The sixth message includes at least one of the following: Relevant information about the data analysis task; Identification information of the inference object, where the identification information of the inference object is used to indicate the inference target; A VFL model related identifier, wherein the VFL model related identifier is used to indicate identification information of the data analysis task related model; The first indication information of VFL inference is used to indicate a request for VFL inference.

18. The method according to claim 15 or 17, wherein Before the second communication device sends a sixth message to the third communication device based on the first feedback message, the method further includes: The second communication device sends a seventh message to the fifth communication device, where the seventh message is used to request that the identification information of the inference object be converted into a data format of the third communication device; The second communication device receives the identification information of the converted inference object from the fifth communication device.

19. The method according to any one of claims 15, 17 to 18, wherein The first feedback message includes at least one of the following: The data analysis failure indication information is used to indicate the failure of the data analysis task; Indication information of a VFL model, wherein the indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model; Information of the third communication device.

20. The method according to any one of claims 15, 17 to 19, wherein After the second communication device sends a sixth message to the third communication device based on the first feedback message, the method further includes: The second communication device receives the data analysis result from the third communication device.

21. The method according to any one of claims 15 to 20, wherein: The first message includes at least one of the following: Relevant information about the data analysis task; Identification information of the inference object; The required supported inference accuracy; The required inference latency.

22. A communication data processing method, wherein: include: The third communication device receives a second message from the first communication device or a sixth message from the second communication device, where the second message or the sixth message is used to request VFL reasoning for the data analysis task; The third communication device determines the data analysis result corresponding to the data analysis task; The third communication device sends the data analysis result to the first communication device or the second communication device.

23. The method according to claim 22, wherein The second message or the sixth message includes at least one of the following: Relevant information about the data analysis task; Identification information of the inference object, where the identification information of the inference object is used to indicate the inference target; A VFL model related identifier, wherein the VFL model related identifier is used to indicate identification information of the data analysis task related model; The first indication information of VFL inference is used to indicate a request for VFL inference.

24. The method according to claim 22 or 23, wherein Before the third communication device determines the data analysis result corresponding to the data analysis task, the method further includes: The third communication device sends an eighth message to the sixth communication device, and the eighth message includes auxiliary information or task information. The auxiliary information is used to indicate the determination method of the input data or the feedback time of the intermediate reasoning result, and the task information is used to indicate information of the data analysis task.

25. The method according to claim 24, wherein After the third communication device sends the eighth message to the sixth communication device, the method further includes: The third communication device receives a sixth feedback message from the sixth communication device, where the sixth feedback message includes an intermediate inference result.

26. The method according to claim 25, wherein The third communication device determines the data analysis result corresponding to the data analysis task, including: The third communication device determines a data analysis result based on the intermediate inference result.

27. The method according to any one of claims 24 to 26, wherein The auxiliary information includes at least one of the following: Time information of input data required when performing the VFL inference; The data type of the input data required when performing the VFL inference; Limitation information of input data required when performing the VFL inference; The maximum response time required to feed back the intermediate inference result when executing the VFL inference.

28. The method according to any one of claims 24 to 27, wherein The task information includes at least one of the following: identification information of the data analysis task; A VFL model related identifier, wherein the VFL model related identifier is used to indicate identification information of the data analysis task related model; The identification information of the inference object is used to indicate the inference object identification information corresponding to the input data.

29. The method according to any one of claims 25 to 28, wherein The sixth feedback message further includes at least one of the following: Identification information of the inference object, where the identification information of the inference object is used to indicate the inference target; second indication information of VFL reasoning, wherein the second indication information of VFL reasoning is used to indicate that the intermediate reasoning result is obtained through VFL reasoning; The VFL model related identifier is used to indicate the identification information of the model related to the data analysis task; Identification information of the data analysis task.

30. The method according to any one of claims 24 to 29, wherein The third communication device sending an eighth message to the sixth communication device includes: The third communication device sends a ninth message to the fifth communication device, where the ninth message is used to request that the eighth message be forwarded to the sixth communication device.

31. The method according to any one of claims 24 to 30, wherein Before the third communication device sends the eighth message to the sixth communication device, the method further includes: The third communication device sends a tenth message to the fifth communication device, where the tenth message is used to request that the identification information of the inference object be converted into a data format of the sixth communication device; The third communication device receives the identification information of the converted inference object from the fifth communication device.

32. The method according to any one of claims 26 to 31, wherein Before the third communication device determines the data analysis result according to the intermediate inference result, the method further includes: The third communication device sends an eleventh message to the fifth communication device, where the eleventh message is used to request that identification information of the inference object in the intermediate inference result be converted into a data format of the third communication device; The third communication device receives the identification information of the converted inference object from the fifth communication device.

33. The method according to any one of claims 22 to 32, wherein: Also includes: The third communication device receives a fourth message from the first communication device, where the fourth message is used to request acquisition of a model corresponding to the data analysis task.

34. The method according to claim 33, wherein After the third communication device receives the fourth message from the first communication device, the method further includes: The third communication device sends a fourth feedback message to the first communication device; The fourth feedback message includes at least one of the following: Indication information that a model cannot be provided, wherein the indication information that a model cannot be provided is used to indicate that a model corresponding to the data analysis task cannot be provided; Indication information of a VFL model, wherein the indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model; A related identifier of a VFL model, wherein the related identifier of the VFL model is used to indicate identification information of a related model of the data analysis task; Information of the third communication device.

35. The method according to any one of claims 22 to 34, wherein Also includes: The third communication device sends a twelfth message to the fourth communication device, where the twelfth message is used to request registration or update of network element capability information of the third communication device.

36. The method according to claim 35, wherein The network element capability information includes at least one of the following: Identification information of supported data analysis tasks; Identification information of the owned model; The model owned is an indication of a VFL model; Indication that the owned model cannot be shared; Device information of the model owned; Supported inference precision; Supported inference latency.

37. A communication data processing device, wherein: include: A first message receiving module is configured to receive a first message from a second communication device, wherein the first message is used to request acquisition of a data analysis result corresponding to a data analysis task; A device information acquisition module, configured to acquire information of a third communication device having a capability related to vertical federated learning (VFL); a second message sending module, configured to send a second message to the third communication device based on the information of the third communication device, wherein the second message is used to request VFL reasoning for the data analysis task, or The first feedback message sending module is configured to send a first feedback message to the second communication device, where the first feedback message is used to instruct the second communication device to request the third communication device to perform VFL inference.

38. The apparatus according to claim 37, wherein The device further comprises: A third message sending module, configured to send a third message to a fourth communication device; The third message includes at least one of the following information: Type information of the communication device being searched, where the type information is used to indicate the type of communication device being searched; Target area of ​​interest information, where the target area of ​​interest information is used to indicate a communication device that searches for a service target area; Indication information indicating that the communication device being searched has VFL server capability, wherein the indication information indicating that the communication device has VFL server capability is used to indicate that a communication device capable of providing VFL service is to be searched; Identification information of the data analysis task that the communication device being searched needs to support; The communication device being searched needs to support providing indication information of data analysis results or inference results; The inference accuracy that the communication device being searched needs to support; The inference latency that the communication device needs to support.

39. The apparatus according to claim 38, wherein The device further comprises: A second feedback message receiving module, configured to receive a second feedback message from the fourth communication device; The second feedback message includes at least one of the following: Look for failed indications; VFL model indication information, where the VFL model indication information is used to indicate that the model supported by the communication device being searched is a VFL model; Indication information indicating that sharing is impossible, the indication information indicating that sharing is impossible is used to indicate that the model supported by the communication device being searched cannot be shared; Information of the third communication device.

40. [Corrected 09.04.2025 under Rule 91] The apparatus according to claim 38, wherein The device further comprises: A third feedback message receiving module, configured to receive a third feedback message from the fourth communication device; The third feedback message includes at least one of the following: Find the successful indication information; Information of a target communication device, where the target communication device is a communication device that meets the information in the third message.

41. The apparatus according to claim 40, wherein The device further comprises: The fourth message sending module is used to send a fourth message to the target communication device, and the fourth message is used to request to obtain the model corresponding to the data analysis task.

42. The apparatus according to claim 41, wherein The device further comprises: a fourth feedback message receiving module, configured to receive a fourth feedback message from the target communication device; The fourth feedback message includes at least one of the following: Indication information that a model cannot be provided, wherein the indication information that a model cannot be provided is used to indicate that a model corresponding to the data analysis task cannot be provided; Indication information of a VFL model, wherein the indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model; A related identifier of a VFL model, wherein the related identifier of the VFL model is used to indicate identification information of a related model of the data analysis task; Information of the third communication device.

43. The apparatus of claim 37, wherein: The device information acquisition module is specifically used to: The information of the third communication device is determined according to internal logic or pre-configuration.

44. The apparatus according to any one of claims 37 to 43, wherein The device further comprises: The first result receiving module is used to receive the data analysis result from the third communication device.

45. The apparatus of claim 44, wherein: The device further comprises: a fifth feedback message sending module, configured to send a fifth feedback message to the second communication device; The fifth feedback message includes at least one of the following: the data analysis results; information of a third communication device; The data analysis failure indication information is used to indicate the failure of the data analysis task; The indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model.

46. ​​The apparatus according to any one of claims 37 to 45, wherein The second message sending module is specifically configured to: If it is determined that the third communication device cannot provide the model, sending the second message to the third communication device; or, The first feedback message sending module is specifically configured to: In a case where it is determined that the third communication device cannot provide the model, the first feedback message is sent to the second communication device.

47. A communication data processing device, comprising: A first message sending module, configured to send a first message to a first communication device, wherein the first message is used to request acquisition of a data analysis result corresponding to a data analysis task; a fifth feedback message receiving module, configured to receive a fifth feedback message from the first communication device, wherein the fifth feedback message is used to indicate a result of performing VFL reasoning on the data analysis task, or A first feedback message receiving module is used to receive a first feedback message from the first communication device, and send a sixth message to a third communication device based on the first feedback message, wherein the sixth message is used to request VFL reasoning for the data analysis task, and the third communication device has VFL-related capabilities.

47. The apparatus of claim 46, wherein: The device further comprises: The second result receiving module is used to receive the data analysis result from the third communication device.

48. A communication data processing device, wherein: include: a message receiving module, configured to receive a second message from the first communication device or a sixth message from the second communication device, wherein the second message or the sixth message is used to request VFL reasoning for the data analysis task; A data analysis module, configured to determine the data analysis results corresponding to the data analysis task; A result feedback module is used to send the data analysis result to the first communication device or the second communication device.

49. The apparatus according to claim 48, wherein The device further comprises: An eighth message sending module is used to send an eighth message to a sixth communication device, wherein the eighth message includes auxiliary information or task information, wherein the auxiliary information is used to indicate a method for determining input data or a feedback time for intermediate reasoning results, and the task information is used to indicate information about the data analysis task.

50. The apparatus of claim 49, wherein The device further comprises: The sixth feedback message receiving module is configured to receive a sixth feedback message from the sixth communication device, where the sixth feedback message includes an intermediate reasoning result.

51. The apparatus of claim 50, wherein: The data analysis module is specifically used for: A data analysis result is determined based on the intermediate inference result.

52. The apparatus according to any one of claims 48 to 51, wherein The device further comprises: The fourth message receiving module is used to receive a fourth message from the first communication device, where the fourth message is used to request to obtain the model corresponding to the data analysis task.

53. The apparatus of claim 52, wherein: The device further comprises: a fourth feedback message sending module, configured to send a fourth feedback message to the first communication device; The fourth feedback message includes at least one of the following: Indication information that a model cannot be provided, wherein the indication information that a model cannot be provided is used to indicate that a model corresponding to the data analysis task cannot be provided; Indication information of a VFL model, wherein the indication information of the VFL model is used to indicate that the model corresponding to the data analysis task is a VFL model; A related identifier of a VFL model, wherein the related identifier of the VFL model is used to indicate identification information of a related model of the data analysis task; Information of the third communication device.

54. The apparatus according to any one of claims 48 to 53, wherein The device further comprises: The twelfth message sending module is configured to send a twelfth message to the fourth communication device, where the twelfth message is used to request registration or update of the network element capability information of the third communication device.

55. A communication device, wherein: The method comprises a processor and a memory, wherein the memory stores a program or instruction that can be run on the processor, and when the program or instruction is executed by the processor, the method implements the steps of the communication data processing method according to any one of claims 1 to 14, or implements the steps of the communication data processing method according to any one of claims 15 to 21, or implements the steps of the communication data processing method according to any one of claims 22 to 36.

56. A readable storage medium, wherein: The readable storage medium stores a program or instruction, and when the program or instruction is executed by the processor, it implements the steps of the communication data processing method according to any one of claims 1 to 14, or implements the steps of the communication data processing method according to any one of claims 15 to 21, or implements the steps of the communication data processing method according to any one of claims 22 to 36.

Images