Communication method and apparatus

Abstract

Embodiments of the present application provide a communication method and apparatus. The method relates to the technical field of communications. In the method, each network element can be registered in a second network element, so that a certain distributed learning initiator (such as a first network element) can acquire some network elements (such as at least one third network element) meeting distributed learning requirements by means of the second network element, so that the first network element can determine a participant finally participating in the distributed learning from among the obtained at least one third network element. Online determination of distributed participants is realized, and the flexibility of determining the participants is improved.

Description

A communication method and apparatus

[0001] Cross-references to related applications

[0002] This application claims priority to Chinese Patent Application No. 202410180013.5, filed on February 16, 2024, entitled "A Communication Method and Apparatus", the entire contents of which are incorporated herein by reference. Technical Field

[0003] This application relates to the field of communication technology, and in particular to a communication method and apparatus. Background Technology

[0004] Distributed learning refers to a learning method that combines features from multiple participants to jointly train a distributed learning model. The model obtained based on distributed learning can then be used to handle various tasks. For the initiator of distributed learning, the initiator receives the task to be processed and determines the features required for the distributed learning model to handle that task. Since the initiator may not be able to provide all the features needed to train the distributed learning model, in this case, the initiator needs to collaborate with other participants to conduct distributed learning.

[0005] Currently, the initiator and other participants agree on each participant in advance, which means that the participants are determined offline, resulting in poor flexibility in the method of determining participants. Summary of the Invention

[0006] This application provides a communication method and apparatus for improving the flexibility of determining participants.

[0007] In a first aspect, embodiments of this application provide a communication method. This communication method can be executed by a first network element, which can be of any type, including existing or newly added network elements, without limitation. The first network element can be a device (such as a terminal device or network device), a software module (such as an application program) or hardware module (such as a chip) within the device, or a client, without specific limitation. The method includes: sending a first request to a second network element, the first request being used to request the discovery of a network element, and the first request indicating one or more features, wherein the features corresponding to the data used for training a distributed learning model by the network element to be discovered include one or more features; receiving first information from the second network element, the first information indicating at least one third network element; receiving the first information from the second network element, the first information indicating at least one third network element, wherein M third network elements participate in the training of the distributed learning model, the M third network elements being some or all of the at least one third network element, and M being a positive integer.

[0008] A distributed learning model, also simply called a model, refers to a model that learns (or trains) using distributed learning methods. A distributed learning model can be described by functions (such as artificial intelligence (AI) functions or machine learning (ML) functions), characteristics, or algorithms that can be described as distributed learning. The first network element can be considered the initiator of the distributed learning model training and also one of the participants. At least one third network element discovered by the first network element through the second network element can be considered a candidate participant in the distributed learning model training. M third network elements can be participants in the distributed learning model training, that is, other participants besides the first network element. Optionally, the first network element can determine the M network elements for distributed learning model training from these at least one third network element based on first information, or directly use at least one third network element as a network element participating in the distributed learning model training; there is no limitation on this.

[0009] In this embodiment, the first network element can discover at least one third network element through the second network element, and select some or all of the at least one third network element to participate in the training of the distributed learning model. This provides a way for the first network element to determine the participants without prior agreement on each participant, offering an online alignment method for distributed learning model training. Furthermore, the first network element can determine other participants independently and at any time. This means that the first network element can flexibly determine other participants, and the timing of this determination is also relatively flexible, thus improving the flexibility of participant determination. Moreover, due to the greater flexibility in participant determination, this method is applicable to more scenarios involving participant determination.

[0010] In one possible implementation, the first request further indicates one or more network element types, features corresponding to each of the one or more network element types, and features corresponding to each network element type belonging to one or more features; wherein, the network element type of at least one third network element belongs to one or more network element types.

[0011] The first request can be understood (or alternatively described) as indicating one or more network element types, and the features corresponding to each of the one or more network element types. The feature corresponding to each network element type belongs to one or more features, which can be described as: each of the one or more features is a feature corresponding to one of the one or more network element types. Wherein, the feature corresponding to each of the one or more network element types belongs to one or more features, that is, the one or more features are the union of the features corresponding to one or more network element types. The feature corresponding to one of the network element types can be understood as the feature specified by the first network element for that network element type to participate in the training of the distributed learning model.

[0012] In the above implementation, the first request may also specify one or more network element types. In this way, at least one third network element discovered by the first network element also belongs to these one or more network element types. This makes the final selection of M third network elements more in line with the needs of the first network element, ensuring the smooth execution of subsequent distributed learning model training.

[0013] In one possible implementation, the first information further indicates the features corresponding to each of the at least one third network element.

[0014] The features corresponding to each third network element can be features supported by that third network element, features supported by that third network element that belong to one or more features, or features determined by the second network element for each third network element to participate in the training of the distributed learning model. However, the fact that the features corresponding to each third network element are features determined by the second network element for each third network element to participate in the training of the distributed learning model does not mean that the features corresponding to each third network element will necessarily participate in the training of the distributed learning model.

[0015] In the above embodiments, the first network element can not only discover at least one third network element through the first information, but also discover the features corresponding to at least one third network element, which facilitates the determination of the M third network elements ultimately used to participate in the training of the distributed learning model, as well as the features of these M third network elements used to participate in the training of the distributed learning model.

[0016] In one possible implementation, the feature corresponding to each third network element in at least one third network element is one of one or more features.

[0017] Among them, the features corresponding to any two third network elements in at least one third network element may have the same features (i.e., there is overlap), or they may not have the same features (i.e., there is no overlap), and no specific limitation is made in this regard.

[0018] In the above embodiments, the first information does not need to indicate some unnecessary features of the third network element, that is, features that will not be used to participate in the training of the distributed learning model. In this way, the resource overhead of transmitting the first information can be reduced relatively.

[0019] In one possible implementation, the method further includes: determining the features of each of the M third network elements used to participate in the training of the distributed learning model; and sending second information to each of the M third network elements, wherein the second information sent to one of the at least one third network element is used to indicate the features of a third network element used to participate in the training of the distributed learning model.

[0020] In the above implementation, the first network element can not only determine the M network elements used to participate in the training of the distributed learning model, but also determine the features of the M third network elements used to participate in the training of the distributed learning model. This achieves feature alignment among the various participants in the training of the distributed learning model, providing a feature alignment mechanism. Feature alignment can then improve the accuracy of distributed learning model training.

[0021] In one possible implementation, the first information includes information about at least one network element group, wherein the information about one network element group includes features corresponding to the network element group, wherein the network element group includes some or all of the network elements in at least one third network element, and the features corresponding to the network element group include one or more features. Optionally, the information about one network element group may also include identifiers of the network elements included in the network element group, and / or the identifier of the network element group itself.

[0022] In the above embodiments, the second network element can group at least one third network element. The first information indicates at least one third network element through the information of at least one network element group, and can also indicate the features corresponding to at least one third network element. Since a network element group can serve as a candidate other participant in the training of the distributed learning model, the first network element can directly select one network element group from at least one network element group as the final other participant in the training of the distributed learning model. In this way, the first network element can intuitively determine M third network elements, thereby relatively reducing the processing load of the first network element.

[0023] In one possible implementation, the first request further indicates one or more network element types, and features corresponding to each of the one or more network element types, wherein the features corresponding to each network element type belong to one or more features; wherein, the network element type of a network element group belongs to one of the one or more network element types.

[0024] In the above embodiments, by limiting the network element type corresponding to the feature, the provided features are made more targeted and accurate. For example, one of the features is the location of user equipment (UE). Assuming that both the application function (AF) and the network data analytics function (NWDAF) can obtain UE location information, but the AF obtains the UE location with higher accuracy, this method can be used to specify the AF to provide the UE location.

[0025] In one possible implementation, the method further includes: sending third information to at least one third network element, wherein the third information sent to one of the at least one third network element indicates that a third network element feeds back a feature corresponding to a third network element; and receiving fourth information from at least one third network element, wherein the fourth information received from one of the at least one third network element indicates a feature corresponding to each third network element.

[0026] In the above embodiments, the first network element can request its corresponding features from at least one third network element on its own. This makes it easier to obtain the features corresponding to at least one third network element more accurately, and the second network element does not need to notify the first network element of the features corresponding to at least one third network element, thus reducing the processing load of the second network element.

[0027] In one possible implementation, the method further includes: sending fifth information to the second network element, the fifth information indicating features supported by the first network element.

[0028] In the above implementation, the first network element can be registered in the second network element, so that the first network element can be discovered by the second network element during the training of other distributed learning models.

[0029] Secondly, embodiments of this application provide a communication method. This communication method can be executed by a second network element. The method can be executed by the second network element, which can be of any type, including existing network elements and newly added network elements, etc., without limitation. The second network element can be a device (such as a terminal device or network device), a software module (such as an application program) or hardware module (such as a chip) within the device, or a client, etc., without specific limitation. The method includes: receiving a first request from a first network element, the first request being used to request the discovery of a network element, the first request indicating one or more features, the features corresponding to the data used for training a distributed learning model by the network element to be discovered including one or more features; sending first information to the first network element, the first information indicating at least one third network element, wherein M third network elements participate in the training of the distributed learning model, the M third network elements being some or all of the at least one third network element, and M being a positive integer.

[0030] In one possible implementation, the first request further indicates one or more network element types, wherein each network element type in the one or more network element types corresponds to a feature, and the feature corresponding to each network element type belongs to one or more features; wherein the network element type of at least one third network element belongs to one or more network element types.

[0031] In one possible implementation, the first information further indicates the features corresponding to each of the at least one third network element.

[0032] In one possible implementation, the feature corresponding to each third network element in at least one third network element is one of one or more features.

[0033] In one possible implementation, the first information includes: information of at least one network element group, wherein the information of one network element group includes a feature corresponding to the network element group, wherein the network element group includes some or all of the network elements in at least one third network element, and the feature corresponding to the network element group includes one or more features.

[0034] In one possible implementation, the first request further indicates one or more network element types, wherein each network element type in the one or more network element types corresponds to a feature, and the feature corresponding to each network element type belongs to one or more features; wherein the network element type of a network element group belongs to one of the one or more network element types.

[0035] In one possible implementation, the method further includes: receiving fifth information from a first network element, the fifth information indicating features supported by the first network element.

[0036] In one possible implementation, the method further includes: receiving sixth information from at least one third network element, wherein receiving the sixth information from one of the at least one third network elements is used to indicate a feature supported by a third network element.

[0037] Thirdly, embodiments of this application provide a communication method. This communication method can be executed by a first network element. The method can be executed by the first network element, which can be of any type, including existing network elements and newly added network elements, etc., without limitation. The first network element can be a device (such as a terminal device or network device), a software module (such as an application program) or hardware module (such as a chip) within the device, or a client, etc., without specific limitation. The method includes: determining at least one third network element; sending first information to each of the at least one third network element, wherein the first information sent to one of the at least one third network element is used to request a feature supported by the third network element, wherein some or all of the features supported by the at least one third network element are features corresponding to the data trained by the distributed learning model; receiving second information from each of the at least one third network element, wherein receiving the second information from one of the at least one third network element indicates a feature supported by the third network element; determining M third network elements for participating in the features of the distributed learning model training, wherein the M third network elements are some or all of the at least one third network element, and M is a positive integer.

[0038] In this embodiment, the first network element can directly determine the M third network elements participating in the training of the distributed learning model from at least one third network element, improving the flexibility of determining the participants. Furthermore, at least one third network element registers with the second network element, but does not register the features supported by this at least one third network element, reducing the risk of leakage of features supported by at least one third network element and ensuring the security of the features supported by at least one third network element.

[0039] In one possible implementation, the method further includes: sending third information to each of the M third network elements, wherein the third information sent to one of the at least one third network element indicates a feature of the third network element used to participate in the training of the distributed learning model.

[0040] In the above embodiments, the first network element can also notify M third network elements of the features used to participate in the training of the distributed learning model, providing a mechanism for aligning the features of the distributed learning model training.

[0041] In one possible implementation, determining at least one third network element includes: receiving fourth information from a second network element, the fourth information indicating information of at least one third network element; and determining at least one third network element based on the fourth information.

[0042] In the above embodiments, the first network element can discover at least one third network element through the second network element, so that the first network element can intuitively and easily discover this at least one third network element.

[0043] Fourthly, embodiments of this application provide a communication method. This communication method can be executed by a third network element. The method can be executed by a third network element, which can be of any type, including existing or newly added network elements, without limitation. The third network element can be a device (such as a terminal device or network device), a software module (such as an application program) or hardware module (such as a chip) within the device, or a client, without specific limitation. The method includes: receiving first information from a first network element, the first information being used to request features supported by the third network element; and sending second information to the first network element, the second information indicating features supported by the third network element.

[0044] In one possible implementation, the method further includes: receiving third information from a first network element, the third information indicating features used by the third network element to participate in the training of a distributed learning model.

[0045] In one possible implementation, the method further includes: sending fifth information to the second network element, the fifth information indicating information of the third network element.

[0046] Fifthly, embodiments of this application provide a communication device. This communication device can be a first network element as described in the first aspect above, or a software module or hardware module (e.g., a chip) configured within the first network element. The communication device includes corresponding means or modules for performing the first aspect or any possible implementation described above. For example, the communication device includes a processing module (sometimes also called a processing unit) and a transceiver module (sometimes also called a transceiver unit).

[0047] For example, a transceiver module is used to send a first request to a second network element and receive first information from the second network element, the first information indicating at least one third network element; wherein, M third network elements among the at least one third network element are used to participate in the training of a distributed learning model, and the M third network elements are some or all of the at least one third network element.

[0048] In one possible implementation, the communication device may also perform any of the possible implementations in the first aspect above, which will not be listed one by one here.

[0049] In a sixth aspect, embodiments of this application provide a communication device. This communication device can be a second network element as described in the second aspect above, or a software module or hardware module (e.g., a chip) configured within the second network element. The communication device includes corresponding means or modules for performing the second aspect above or any possible implementation. For example, the communication device includes a processing module (sometimes also called a processing unit) and a transceiver module (sometimes also called a transceiver unit).

[0050] For example, the transceiver module is used to receive a first request from a first network element and to send first information to the first network element, wherein the first information indicates at least one third network element, and M third network elements participate in the training of a distributed learning model, and the M third network elements are some or all of the at least one third network element.

[0051] In one possible implementation, the communication device may also perform any of the possible implementations in the second aspect above, which will not be listed one by one here.

[0052] In a seventh aspect, embodiments of this application provide a communication device. This communication device can be a first network element as described in the third aspect above, or a software module or hardware module (e.g., a chip) configured within the first network element. The communication device includes corresponding means or modules for performing the third aspect above or any possible implementation. For example, the communication device includes a processing module (sometimes also called a processing unit) and a transceiver module (sometimes also called a transceiver unit).

[0053] For example, the processing module is used to determine at least one third network element; the transceiver module is used to send first information to at least one third network element and receive second information from at least one third network element respectively; the processing module is also used to determine the features of M third network elements for participating in the training of the distributed learning model, wherein the M third network elements are some or all of the at least one third network element.

[0054] In one possible implementation, the communication device may also perform any of the possible implementations in the third aspect above, which will not be listed one by one here.

[0055] Eighthly, embodiments of this application provide a communication device. This communication device can be a third network element as described in the fourth aspect above, or a software module or hardware module (e.g., a chip) configured in the third network element. The communication device includes corresponding means or modules for performing the fourth aspect above or any possible implementation. For example, the communication device includes a processing module (sometimes also called a processing unit) and a transceiver module (sometimes also called a transceiver unit).

[0056] For example, a transceiver module is used to receive first information from a first network element, the first information being used to request features supported by a third network element, and to send second information to the first network element, the second information indicating features supported by the third network element.

[0057] In one possible implementation, the communication device may also perform any of the possible implementations in the fourth aspect above, which will not be listed one by one here.

[0058] Ninthly, embodiments of this application provide a communication device. The communication device includes a processor and a memory, wherein the memory stores program instructions, and the processor executes the program instructions in the memory to perform the method as described in any one of the first to fourth aspects.

[0059] Optionally, the wireless communication device may also include a communication interface, and the processor may be coupled to the communication interface, or the processor may be set independently of the communication interface.

[0060] In one implementation, when the communication device is a wireless communication device, the communication interface can be a transceiver or an input / output interface. Optionally, the transceiver can be a transceiver circuit. Optionally, the input / output interface can be an input / output circuit.

[0061] In another implementation, when the communication device is a chip or chip system, the communication interface can be an input / output interface, interface circuit, output circuit, input circuit, pin, or related circuit on the chip or chip system. The processor can also be manifested as a processing circuit or logic circuit.

[0062] Optionally, the communication device may also include other components, such as antennas, input / output modules, or interfaces. These components may be hardware, software, or a combination of both.

[0063] Tenthly, embodiments of this application provide a communication device. The communication device includes: a processing circuit and an interface circuit; wherein: the interface circuit is configured to couple with a memory external to the communication device and provide a communication interface for the processing circuit to access the memory; the processing circuit is configured to execute program instructions in the memory to implement the method as described in any one of the first to fourth aspects.

[0064] In practical implementation, the communication device can be a chip, the input circuit can be an input pin, the output circuit can be an output pin, and the processing circuit can be a transistor, gate circuit, flip-flop, and various logic circuits. The input signal received by the input circuit can be received and input by, for example, but not limited to, a receiver, and the signal output by the output circuit can be output to, for example, but not limited to, a transmitter and transmitted by the transmitter. Furthermore, the input circuit and the output circuit can be the same circuit, which is used as the input circuit and the output circuit at different times. This application does not limit the specific implementation of the processor and various circuits.

[0065] In one implementation, the communication device can be a wireless communication device, i.e., a computer device that supports wireless communication functions. Specifically, the wireless communication device can be a terminal such as a smartphone, or a wireless access network device such as a base station. A system-on-a-chip (SoC) is also called a system-on-chip (SoC). The communication chip may include a baseband processing chip and a radio frequency (RF) processing chip. The baseband processing chip is sometimes referred to as a modem or baseband chip. The RF processing chip is sometimes referred to as an RF transceiver or RF chip. In physical implementation, some or all of the communication chip chips can be integrated within the SoC chip. For example, the baseband processing chip is integrated into the SoC chip, while the RF processing chip is not integrated with the SoC chip. The interface circuit can be the RF processing chip in the wireless communication device, and the processing circuit can be the baseband processing chip in the wireless communication device.

[0066] In another implementation, the communication device can be a component of a wireless communication device, such as an integrated circuit product like a system-on-a-chip (SoC) or communication chip. The interface circuit can be an input / output interface, interface circuit, output circuit, input circuit, pins, or related circuits on the chip or chip system. The processor can also be represented as a processing circuit or logic circuit.

[0067] Eleventhly, embodiments of this application provide a chip system. The chip system includes a processor and an interface. The processor is configured to call and execute instructions from the interface, and when the processor executes the instructions, it implements the method described in any one of the first to fourth aspects.

[0068] In a twelfth aspect, embodiments of this application provide a computer-readable storage medium. This computer-readable storage medium is used to store a computer program or instructions that, when executed, implement the methods described in any one of the first to fourth aspects.

[0069] In a thirteenth aspect, embodiments of this application provide a computer program product containing instructions. When run on a computer, it implements the method described in any one of the first to fourth aspects.

[0070] Regarding the beneficial effects of the second, fourth to thirteenth aspects, please refer to the beneficial effects discussed in the first or third aspects, which will not be listed here again. Attached Figure Description

[0071] Figure 1 is a schematic diagram of the data distribution for vertical federated learning applicable to the embodiments of this application;

[0072] Figure 2 is a schematic diagram of a distributed learning scenario applicable to an embodiment of this application;

[0073] Figure 3 is a schematic diagram of a vertical federated learning process;

[0074] Figure 4 is a schematic diagram of a distributed learning scenario provided by an embodiment of this application;

[0075] Figure 5 is a schematic diagram of an intelligent network architecture based on NWDAF provided in an embodiment of this application;

[0076] Figure 6 is a schematic diagram of another distributed learning scenario provided by an embodiment of this application;

[0077] Figure 7 is a schematic diagram of a communication method provided in an embodiment of this application;

[0078] Figure 8 is a schematic diagram of another communication method provided in an embodiment of this application;

[0079] Figure 9 is a schematic diagram of another communication method provided in an embodiment of this application;

[0080] Figure 10 is a schematic diagram of another communication method provided in an embodiment of this application;

[0081] Figure 11 is a schematic diagram of another communication method provided in an embodiment of this application;

[0082] Figure 12 is a schematic diagram of the structure of a communication device provided in an embodiment of this application;

[0083] Figure 13 is a schematic diagram of another communication device provided in an embodiment of this application;

[0084] Figure 14 is a schematic diagram of the structure of another communication device provided in an embodiment of this application. Detailed Implementation

[0085] The specific implementation of this application will be described below with reference to the accompanying drawings in the embodiments of this application.

[0086] The following explanations of some terms used in the embodiments of this application are provided to facilitate understanding by those skilled in the art.

[0087] 1. Data (or dataset)

[0088] Data can be used for model training or inference. Data can take many forms, such as structured data like tables or databases, or unstructured data like text, images, audio, or video. For example, data may include at least one of the following: a user's service experience, location, average throughput, average packet latency, radio access type (RAT), reference signal receiving power (RSRP), reference signal receiving quality (RSRQ), or radio access network throughput (RAN throughput).

[0089] Data used for model training can be divided into two dimensions (or spaces): a sample space and a feature space. The sample space represents the sample objects, i.e., which objects (such as users) are being learned from. The feature space represents the features used for model training, i.e., which features of the objects are being learned from.

[0090] Features can be understood as attributes or variables of data, or as the characteristics corresponding to the data. Features can be obtained by preprocessing the raw data so that the model can better understand and predict it. Preprocessing methods include sorting, vector representation, matrix representation, statistical averaging, feature extraction, or dimensionality reduction, and are not limited to one or more of these. In model training, labels correspond to features. Features are attributes or variables that describe the data, while labels are the target variables that the model is expected to predict or classify. Let's take training a model for image classification as an example. If we want to train a model to distinguish between images of cats and dogs, then a set of labeled images serves as training data. Each image contains two parts: features and labels. In this example, features are attributes that describe the image, possibly including at least one of the following: pixel values, color histogram, or texture features. These features provide key information about the image, helping the model understand its content. The label, on the other hand, is the result of classifying the image—whether it is a cat or a dog.

[0091] For example, please refer to Table 1 below for an example of one type of data.

[0092] Table 1

[0093] As shown in Table 1 above, the sample space corresponding to the data includes User 1 and User 2, and the feature space includes wireless access type, reference signal received power, reference signal received quality, and wireless access network throughput.

[0094] 2. Distributed learning

[0095] Distributed learning refers to a learning method that combines data from multiple participants to learn a model. In this application, the model trained using distributed learning is referred to as a distributed learning model. A distributed learning model can be an AI model, an ML model, a feature, or an algorithm, etc., without specific limitations.

[0096] A typical example of distributed learning is federated learning (FL). Federated learning is a machine learning framework that effectively helps multiple users to use data and perform machine learning modeling while meeting user privacy, data security, and government regulations. As a distributed machine learning paradigm, federated learning can effectively solve the data silo problem by jointly modeling without sharing user data, thereby technically breaking down data silos and enabling artificial intelligence (AI) collaboration. Distributed learning is suitable when two training entities cannot share their models (e.g., due to privacy issues between AF and NWDAF, or between multi-vendor NWDAFs) or the input data used to train the model (to avoid large-scale data collection or privacy problems).

[0097] Based on the different characteristics of the data sources of the participating parties, federated learning can be divided into three categories: horizontal federated learning, transfer learning, and vertical federated learning (VFL). In horizontal federated learning, the overlap of the sample spaces corresponding to the data is low (or can be described as minimal overlap), but the overlap of sample features is high (or can be described as significant overlap). For example, the participants in horizontal federated learning include participant 1 and participant 2. Participant 1 provides data including feature 1 and feature 2 of user 1, and participant 2 provides data including feature 1 and feature 2 of user 2. Transfer learning is a method of applying a model from a source domain to a target domain for a target task. For example, task A (source task) is to classify cats and tigers in pictures, and task B (target task) is to distinguish the lengths of cats and tigers in pictures. Traditional machine learning methods can only train two different models for these two completely different tasks. However, using transfer learning, the model for task A can be fully utilized, and fine-tuned to obtain a model suitable for task B.

[0098] Vertical federated learning, as a machine learning technique, can be used to address model training and inference when members are unwilling to share the original dataset. It is suitable for situations where the training sample spaces (e.g., samples) of the participants have high overlap, while the data (or features) of the participants have low overlap. Vertical federated learning combines different datasets (or features) of common samples from multiple participants for federated learning; that is, the data (or features) of each participant are vertically partitioned, hence the name vertical federated learning.

[0099] For example, please refer to Figure 1, which is a schematic diagram of data distribution in a vertical federated learning model. Figure 1 uses participant A and participant B as an example of federated learning. As shown in Figure 1, the sample spaces of participant A and participant B have a high degree of overlap, while their feature spaces have a low degree of overlap. For example, participant A's dataset contains data from user 1, user 2, user 3, and user 4, where each user's data includes features 1, 2, 3, 4, and 5. Participant B's dataset contains data from user 1, user 2, user 3, user 4, and user 5, where each user's data includes features 6, 7, 8, 9, and 10. Therefore, the feature overlap between participant A and participant B is very small, but most of their samples are the same. When performing longitudinal federated learning, participants A and B can select the same samples (i.e., user 1, user 2, user 3, and user 4) from their respective datasets. The corresponding features (i.e., feature 1, feature 2, feature 3, feature 4, feature 5, feature 6, feature 7, feature 8, feature 9, and feature 10) and the labels corresponding to these features are used to train the longitudinal federated learning model.

[0100] All parties involved in distributed learning can be considered participants (or stakeholders), and participants can be divided into primary participants and secondary participants. There can be one or more secondary participants, without limitation. Primary participants possess the labels of the data required for the distributed learning task, and optionally some of the data features required for the task. Secondary participants possess some or all of the data features required for the corresponding vertical federated learning task. Primary participants can also be called active participants, active clients, etc. Secondary participants can also be called passive participants, passive clients, etc.

[0101] Distributed learning also includes coordinators, also known as service providers. Coordinators, or trusted coordinators, are responsible for maintaining the distributed learning process, authorizing and removing members, and optionally handling tasks such as key distribution and decryption of intermediate messages. Coordinators are typically independent third parties or reputable organizations within the industry. Furthermore, the participant initiating distributed learning can be called an initiator or an active participant, while other participants can be considered passive participants. The initiator can be a primary participant, a secondary participant, or a coordinator in the distributed learning process; there are no restrictions on which role to play.

[0102] The key characteristic of distributed learning model training (such as VFL joint model training) is the presence of at least two models (e.g., ML models), each associated with a different training entity (e.g., network element) (i.e., one master participant and at least one slave participant). The master participant trains its own model together with the model owned by the slave participant, where both models share the same model objective (e.g., being trained to predict the same output, such as service experience), but are distinct from each other (e.g., having different data types of inputs). Optional participants in the distributed learning also include a distributed learning model training ID, which represents a unique identifier for the distributed learning process, used to identify the unique identifiers of at least two models (i.e., models from the active and passive participants) and associate them with the distributed learning model training process.

[0103] Please refer to Figure 2, which is a schematic diagram of a distributed learning scenario applicable to an embodiment of this application. As shown in Figure 2, the scenario includes a main participant, a slave participant, and a collaborator, and any two of the main participant, slave participant, and collaborator can communicate through a communication network, such as a fifth-generation (5G) network. thGeneration (5G) communication networks, sixth-generation communication networks, or future evolution communication networks, etc.

[0104] In distributed learning, any participant (such as a primary participant, a secondary participant, or a collaborator) can be implemented through a network element. The network element involved in the embodiments of this application can be implemented through a device, a client, software (such as an application) within the device, a hardware module (such as a chip) within the device, or a cluster of devices. A client can be, for example, an application running on a terminal device, or a terminal device running an application; a client can also be, for example, an application or function within a network element, etc., without limitation. For example, a network element can specifically be a terminal device, a network device, or a third-party server, etc., without limitation. The primary participant has its own distributed learning model (or its local distributed learning model), and the secondary participant has its own distributed learning model (or its local distributed learning model).

[0105] Figure 2 illustrates the scenario with client 1 as the primary participant, client 2 as the secondary participant, and the server as the collaborator. In practice, the implementation of the primary, secondary, and collaborators is not limited. Either the primary or secondary participant can initiate the training of the distributed learning model, while the collaborator can coordinate the training of the distributed learning model between the two parties.

[0106] The terminal device is a device with wireless transceiver capabilities, which can be a fixed device, mobile device, handheld device, wearable device, vehicle-mounted device, or a wireless device (e.g., a communication module or chip system) built into the aforementioned devices. The terminal device is used to connect people, objects, and machines, and can be widely used in various scenarios, including but not limited to: cellular communication, device-to-device (D2D) communication, vehicle-to-everything (V2X) communication, machine-to-machine / machine-type communications (M2M / MTC) communication, Internet of Things (IoT), virtual reality (VR), augmented reality (AR), industrial control, self-driving, remote medical care, smart grid, smart furniture, smart office, smart wearables, smart transportation, smart city, drones, robots, and other similar scenarios. Terminal equipment may sometimes be referred to as user equipment (UE), terminal, access station, UE station, remote station, wireless communication equipment, or user device, etc.

[0107] Network equipment includes, for example, access network equipment (or, referred to as access network devices / access network elements), and / or core network equipment (or, referred to as core network devices / core network elements). Access network equipment is a device with wireless transceiver capabilities used to communicate with the terminal equipment. Access network equipment includes, but is not limited to, base stations (BTS, Node B, eNodeB / eNB, or gNodeB / gNB) in communication systems, transmission reception points (TRPs), 3GPP subsequent evolution base stations, access nodes in wireless fidelity (WiFi) systems, wireless relay nodes, wireless backhaul nodes, satellites, or drones, etc. The base station can be: macro base station, micro base station, pico base station, small cell, relay station, etc. Multiple base stations can support networks using the same access technology mentioned above, or they can support networks using different access technologies mentioned above. A base station can contain one or more co-located or non-co-located transmission and reception points. Access network equipment can also be a wireless controller, centralized unit (CU), also known as aggregation unit, and / or distributed unit (DU) in a cloud radio access network (C(R)AN) scenario. Access network equipment can also be a server, wearable device, or vehicle-mounted equipment. For example, in vehicle-to-everything (V2X) technology, the access network equipment can be a roadside unit (RSU). The following explanation uses a base station as an example. Multiple access network devices in the communication system can be base stations of the same type or different types. Base stations can communicate with terminal devices directly or via relay stations. Terminal devices can communicate with multiple base stations using different access technologies.

[0108] In the case where the access network equipment includes a CU and / or a DU, the CU and DU can be understood as a logical functional division of the access network equipment. The CU and DU can be physically separated or deployed together; this application does not specifically limit this. One CU can connect to one DU, or multiple DUs can share one CU. The CU and DU can be divided according to the protocol stack. One possible approach is to deploy the radio resource control (RRC), service data adaptation protocol (SDAP), and packet data convergence protocol (PDCP) layers on the CU, and the remaining radio link control (RLC), media access control (MAC), and physical layers on the DU. This application does not completely limit the CU and DU to be divided according to the above protocol stack method; other division methods are also possible, such as division according to service type.

[0109] Access network equipment can refer to a Centralized Unit Control Plane (CU-CP) node or a Centralized Unit User Plane (CU-UP) node, or it may include both CU-CP and CU-UP. CU-CP is responsible for control plane functions, mainly including RRC and PDCP-C. PDCP-C is primarily responsible for control plane data encryption / decryption, integrity protection, and data transmission. CU-UP is responsible for user plane functions, mainly including SDAP and PDCP-U. SDAP is primarily responsible for processing core network data and mapping flows to bearers. PDCP-U is primarily responsible for data plane encryption / decryption, integrity protection, header compression, sequence number maintenance, and data transmission.

[0110] In different systems, CU (including CU-CP or CU-UP) or DU may have different names, but those skilled in the art will understand their meaning. For example, in an open radio access network (O-RAN) system, CU may also be called O-CU (open CU), DU may also be called O-DU, CU-CP may also be called O-CU-CP, and CU-UP may also be called O-CU-UP.

[0111] Core network equipment is used to implement at least one of the functions of mobility management, data processing, session management, policy and charging. The names of the equipment implementing core network functions may differ in systems using different access technologies, and this application does not limit this. Taking a 5G system as an example, the core network equipment includes: access and mobility management function (AMF), session management function (SMF), or user plane function (UPF), etc.

[0112] The basic process of training a distributed learning model is introduced below with reference to the schematic diagram of the vertical federated learning process shown in Figure 3. Figure 3 takes distributed learning using a homomorphic encryption algorithm as an example. The main participants involved in Figure 3 are, for example, the main participants involved in Figure 2; the slave participants involved in Figure 3 are, for example, the slave participants involved in Figure 2; and the collaborators involved in Figure 3 are, for example, the collaborators involved in Figure 2.

[0113] S301. Initialize local parameters for both the primary and secondary participants.

[0114] Initialize the local parameters Θ of the participant's local distributed learning model. A The main participant initializes the local parameters Θ of its local distributed learning model. B .

[0115] S302. The collaborating party creates an encryption key pair. The collaborating party creates a homomorphic encryption key pair based on a homomorphic encryption algorithm.

[0116] S303, The collaborating party sends the public key to the secondary and primary participants.

[0117] Specifically, S303 includes S303a and S303b, where S303a involves the collaborating party sending its public key to the primary participant, and the primary participant receiving the public key from the collaborating party. S303b involves the collaborating party sending its public key to the secondary participant, and the secondary participant receiving the public key from the collaborating party.

[0118] S304. Determine the local parameters and loss function from the participants.

[0119] Participants can train their local distributed learning model based on their data, update the local parameters and loss function of the local distributed learning model, and encrypt the local parameters and loss function using a public key to obtain the encrypted local parameters, for example: The encrypted loss function is, for example, in The symbol represents homomorphic encryption.

[0120] S305. The slave participant sends the encrypted local parameters and loss function to the master participant. Correspondingly, the master participant receives the encrypted local parameters and loss function from the slave participant.

[0121] For example, local parameters from the participants and loss function Send to the main participants.

[0122] S306. The main participant determines the local parameters, intermediate parameters, and loss function.

[0123] Similarly, the primary participant can train its local distributed learning model based on its own data, update its local parameters, and encrypt the local parameters using a public key to obtain the encrypted local parameters, for example: And the encrypted loss function is, for example, The primary participant can also determine intermediate parameters based on its local parameters and those of the secondary participants. intermediate parameters For example, it could be The primary participant can also base its decisions on local parameters from the secondary participants. and loss function and loss function Wait, determine the total loss function.

[0124] S307. The primary participant sends intermediate parameters to the secondary participant. Correspondingly, the secondary participant receives intermediate parameters from the primary participant.

[0125] For example, the main participant will use intermediate parameters Send to the participating parties.

[0126] S308. The primary participant sends the loss function to the collaborating party. Correspondingly, the collaborating party receives the loss function from the primary participant.

[0127] For example, the main participant will use the loss function Send to collaborating parties.

[0128] S309. The collaborating party determines whether the iteration should be terminated.

[0129] For example, the collaborating party uses its private key to decrypt the loss function. Obtain the decrypted loss function, and determine whether the iteration should terminate based on the decrypted loss function, or it can be described as determining whether to stop training the distributed learning model.

[0130] S310, Determine the encrypted local gradient from the participants.

[0131] Calculate the local gradient from the participants. And add random noise We obtain the encrypted local gradient, i.e.

[0132] S311. The participant sends the encrypted local gradient to the collaborator. Correspondingly, the collaborator receives the encrypted local gradient from the participant.

[0133] The collaborating party uses its private key to decrypt the encrypted local gradient, obtaining the decryption result, i.e.

[0134] S312. The collaborating party sends the decryption result from the participating party. Correspondingly, it receives the decryption result from the collaborating party from the participating party.

[0135] The gradient of the decryption is obtained by subtracting random noise from the decryption result by the participants. And update the local parameters based on the decrypted gradient.

[0136] S313. The main participant determines the encrypted local gradient.

[0137] The main participant calculates the local gradient. And add random noise We obtain the encrypted local gradient, i.e.

[0138] S314. The primary participant sends the encrypted local gradient to the collaborating party. Correspondingly, the collaborating party receives the encrypted local gradient from the primary participant.

[0139] The collaborating party uses its private key to decrypt the encrypted local gradient, obtaining the decryption result, i.e.

[0140] S315. The collaborating party sends the decryption result to the primary participant. Correspondingly, the primary participant receives the decryption result from the collaborating party.

[0141] Similarly, the main participant subtracts random noise from the decryption result to obtain the decryption gradient. And update the local parameters based on the decrypted gradient.

[0142] Because the participants in distributed learning interact with intermediate computation results during training, none of them can access the other party's raw data throughout the entire training process. Instead, they complete the model training process by exchanging intermediate results. In this way, while achieving joint training, the local data privacy of the participants is also protected.

[0143] Figure 3 uses two participants as an example. When there are three or more participants, the distributed learning process can be referred to the distributed learning process in Figure 3, and will not be listed one by one here. There can be various encryption algorithms or model training methods involved in distributed learning, such as multi-party multi-class distributed learning based on privacy-preserving tag sharing, multi-party distributed learning based on secret sharing, etc., which are not specifically limited here.

[0144] Since a single participant may not possess all the features required for distributed learning, it is necessary to collaborate with other participants to train the distributed learning model. This necessitates identifying the other participants in the vertical federation. Specifically, before model training begins, feature alignment needs to be performed on each participant, determining the features each vertical federation participant will use for distributed learning model training. However, currently, participation is negotiated offline, resulting in limited flexibility in participant selection and failing to meet the needs of certain business applications.

[0145] In view of this, embodiments of this application provide a communication method. In this method, each network element can register in a second network element. Thus, a distributed learning initiator (such as a first network element) can obtain some network elements (such as at least one third network element) that meet the requirements of distributed learning through the second network element. The first network element can then determine the final participants in the distributed learning from the obtained at least one third network element, eliminating the need for pre-determining participants and implementing an online mechanism for determining distributed participants, thereby improving the flexibility of participant determination. Furthermore, the first network element can also determine the features used by the participants to participate in the training of the distributed learning model, thus providing a way to align the features of each participant, further improving the flexibility of feature alignment.

[0146] The communication method provided in this application embodiment can be applied to the distributed learning scenario involved in Figure 2 above. In addition, it can also be applied to other similar distributed learning scenarios, which will be described below with reference to the accompanying drawings.

[0147] Please refer to Figure 4, which is a schematic diagram of a distributed learning scenario provided by an embodiment of this application. Figure 4 illustrates the initiator, other participants, and the registrant. Optionally, this scenario also includes a collaborator. Any one of the initiator, other participants, the registrant, and the collaborator can be implemented through network elements. There can be one or more other participants, which is not limited. Optionally, the registrant can also be a collaborator in distributed learning.

[0148] Figure 4 illustrates an example where the initiator is the first network element, the registrant is the second network element, other participants are some or all of at least one third network element, and the collaborator is the fourth network element. The initiator (e.g., the first network element) initiates distributed learning. The registrant (e.g., the second network element) registers the initiator and other participants. Collaborators assist in training the distributed learning model. The implementation of network elements can be found in the previous discussion of network elements; repetitions will not be listed here.

[0149] For example, at least one third network element and the first network element can both register in the second network element. When the first network element initiates distributed learning, it can request the third network element to participate in the distributed learning from the second network element.

[0150] The 3GPP standard defines an intelligent network architecture based on the network data analytics function (NWDAF). The purpose of this intelligent network architecture is to collect massive amounts of information from the network and utilize big data and artificial intelligence technologies (such as longitudinal federated learning) to output valuable information that assists operators in policy formulation and network resource adjustment, thereby improving user experience and reducing network load.

[0151] Please refer to Figure 5, which is a schematic diagram of an intelligent network architecture based on NWDAF provided in an embodiment of this application. It can also be regarded as a schematic diagram of another distributed learning scenario provided in an embodiment of this application. Figure 5 illustrates NWDAF, Vertical Federated Learning Server (VFL Server), Vertical Federated Learning Support Function (VFLSF), Application Function (AF), Network Function (NF), UE, and RAN. NFs include, for example, Network Exposure Function (NEF), Network Repository Function (NRF), Network Data Analytics Function (NWDAF), Policy Control Function (PCF), Access and Mobility Management Function (AMF), Session Management Function (SMF), Operations, Administration and Management (OAM) (also known as network management), and User Plane Function (UPF), etc.

[0152] Any one of the NF, UE, or RAN involved in Figure 5 can be used as an example of the first network element involved in Figure 4. At least one of the NF, UE, or RAN involved in Figure 5 can be used as an example of at least one third network element involved in Figure 4. The NRF, NEF, or VFLSF involved in Figure 5 can be used as an example of a second network element involved in Figure 4. Alternatively, the VFL Server involved in Figure 5 can serve as a collaborator in distributed learning. For example, the VFLSF can be a newly added network element or can be deployed in an existing network element, such as an NRF, NEF, NWDAF, or AF network element.

[0153] NWDAF (Network Data Environment) has functions such as data collection, model training, data analysis, and model inference. It can collect relevant data from other network elements, third-party servers, terminal devices, or network management systems, perform data analysis or model training based on this data, and provide data analysis results to network elements, third-party service servers, terminal devices, or network management systems, or provide trained models to other data analysis function network elements. NWDAF can be functionally divided into analytics logical functions (AnLF) and model training logical functions (MTLF). AnLF is a logical function in NWDAF used to perform model inference, derive analysis results (i.e., derive statistical or predictive analysis results based on the requests of analysis consumers), and provide analysis results. MTLF is a logical function in NWDAF used to train models and provide training services (e.g., provide trained models). An NWDAF may contain only AnLF, only MTLF, or both.

[0154] Please refer to Table 2 below for the analysis results provided by NWDAF, and the data collected required to provide the corresponding analysis results by NWDAF.

[0155] Table 2

[0156] As shown in Table 2 above, NWDAF can provide service experience analysis results. To provide these results, NWDAF needs to collect service-related data such as service identifiers and service experience from AF / UE, and data such as signal reception power and signal reception quality from OAM. NWDAF can train an AI model based on the collected data, and then obtain inferred analysis results based on the AI ​​model, such as predicting the service experience for a certain period in the future. NWDAF can also provide network element load analysis results. To provide these results, NWDAF needs to collect information related to network element resource usage from NRF, as well as information such as UE speed / orientation, and data such as UE path, destination, and arrival time from third-party applications. NWDAF can train an AI model based on the collected data, and then obtain inferred analysis results based on the AI ​​model, such as predicting the network element load for a certain period in the future. NWDAF can also provide MEC service experience analysis. In order to provide this analysis result, NWDAF needs to collect UE identifier, UE location, application identifier, application location and uplink / downlink transmission performance data from AF or UE, train AI model based on this data, and obtain inference analysis results based on AI model, such as obtaining the MEC service experience version analysis for a predicted future time period.

[0157] Application Providers (AFs) are used to convey application-side requests to the network side, such as Quality of Service (QoS) requirements or user state event subscriptions. AFs can be third-party functional entities or application servers deployed by the operator.

[0158] VFLSF is responsible for the registration and discovery of participants in vertical federated learning. VFLSF can be deployed independently or within network elements such as NRF, NEF, NWDAF, or AF.

[0159] The VFL Server can act as a collaborator, responsible for maintaining vertical federation processes, authorizing and removing federated learning members, and other related functions. Optionally, the VFL Server can handle tasks such as distributing encryption keys and decrypting intermediate messages.

[0160] NEF is used to provide a framework, authentication and interface related to network capability openness, and to pass information between 5G system network functions and other network functions.

[0161] NRF is used to provide network element registration and discovery capabilities in a network.

[0162] PCF is primarily responsible for generating and updating UE access policies and QoS flow control policies.

[0163] AMF is primarily responsible for managing user access and mobility, including user registration, reachability, mobility management, N1 / N2 interface signaling transmission, access authentication, and authorization.

[0164] SMF is used to manage the creation, modification, and release of User Protocol Data Unit (PDU) sessions, as well as the allocation and management of IP addresses, and the selection and control of UPFs.

[0165] OAM is used to provide system or network fault indication, performance monitoring, security management, diagnostics, configuration, and user configuration.

[0166] As a network function element of the 5G core network, UPF undertakes core network processing functions such as data traffic processing and routing forwarding.

[0167] In Figure 5, Nnef, Nnrf, Nnwdaf, Naf, Npcf, Nvflsf, Namf, Nsmf, and Nvfl-server are the service interfaces provided by NEF, NRF, NWDAF, AF, PCF, VFLSF, AMF, SFM, and VFL-Server, respectively, used to invoke the corresponding service operations. N2 in Figure 5 is the communication interface between the RAN and the core network control plane (such as AMF), N3 is the communication interface between the RAN and UPF, used for transmitting user data, and N4 is the communication interface between SMF and UPF, used for policy configuration of the UPF, etc.

[0168] Please refer to Figure 6, which is a schematic diagram of a distributed learning scenario provided by an embodiment of this application. Figure 6 uses AF as the first network element, NRF as the second network element, and NWDAF as the third network element as an example. In practice, the specific implementation of the first, second, and third network elements is not limited. In this case, AF can act as the initiator of distributed learning model training, NRF can act as the registrant of distributed learning model training, and NWDAF can act as a participant in distributed learning model training.

[0169] The communication method provided in the embodiments of this application will be described below with reference to the accompanying drawings.

[0170] The steps illustrated by dashed lines in the various embodiments of this application are optional steps. Furthermore, the first network element involved in the various embodiments of this application may be, for example, the main participant or slave participant as shown in Figure 2, the initiator or first network element as shown in Figure 4, or the first network element or AF as shown in Figure 6; the second network element may be, for example, the registrant or second network element as shown in Figure 4, or the VFLSF, NRF, or NEF as shown in Figure 5, or the second network element or NRF as shown in Figure 6; the third network element may be, for example, the main participant or slave participant as shown in Figure 2, other participants or any third network element as shown in Figure 4, or any network element as shown in Figure 5, or the third network element or NWDAF as shown in Figure 6; and the collaborating party may be, for example, the collaborating party as shown in Figure 2, or the collaborating party or fourth network element as shown in Figure 4. These network elements may have other names, or their names may change as the standard evolves; this is not limited.

[0171] Figure 7 illustrates a communication method provided by an embodiment of this application. The steps involved in Figure 7 will be described below.

[0172] S701, the first network element sends a first request to the second network element. Correspondingly, the second network element receives the first request from the first network element.

[0173] The first network element can directly send the first request to the second network element, or the second network element can also send the first request to the second network element through other network elements (which may be one or more network elements), without limitation. For example, if the first network element is AF and the second network element is NRF, then AF can send the first request to NRF through NEF. Alternatively, if the first network element is NWDAF and the second network element is NRF, then NWDAF can directly send the first request to NRF. Or, if the first network element is NWDAF and the second network element is AF, then NWDAF can send the first request to AF through NEF.

[0174] The first network element can acquire a task, which may be acquired from a user's corresponding terminal device or from other network elements; this is not limited. Based on the task's requirements, the first network element can determine the model (i.e., the distributed learning model) to be trained using distributed learning, and determine multiple features used to train this distributed learning model. For example, the first network element can determine multiple features corresponding to the task in a first correspondence relationship. The first correspondence relationship indicates features corresponding to different tasks. Alternatively, the first network element can determine features corresponding to multiple data points corresponding to the task in a second correspondence relationship. The first correspondence relationship indicates data corresponding to different tasks.

[0175] For example, a first network element acquires a task for predicting service experience. The first mapping indicates that the characteristics corresponding to the task for predicting service experience include RAT Type, RSRP, RSRQ, RAN Throughput, service experience, location, average throughput, and average packet latency. Thus, the first network element can determine multiple characteristics, including RAT Type, RSRP, RSRQ, RAN Throughput, service experience, location, average throughput, and average packet latency, based on this task.

[0176] Alternatively, the first network element can also obtain multiple features from other network elements for training the distributed learning model. These multiple features can be understood as the features corresponding to the data required for training the distributed learning model, or as features that the training of the distributed learning model depends on.

[0177] After the first network element determines multiple features, it can identify one or more features from these features other than those it can provide (or support). The features that the first network element can provide (or support) can refer to features that the first network element can directly acquire, and can also include features corresponding to data collected by the first network element from other network elements. These one or more features are used for training the distributed learning model. These one or more features can also be understood as the features the first network element expects (expected all features), or as all features that the first network element expects other participants to provide for training the distributed learning model.

[0178] Since the first network element cannot support these one or more features, it is necessary to discover network elements that can provide these one or more features, so that the first network element and the discovered network element can jointly train the distributed learning model. In this embodiment, the first network element can send a first request to the second network element to request the discovery of network elements. The number of network elements requested for discovery may be one or more, and this is not limited. The first request indicates (or includes) one or more features. The features requested by the first request for the discovered network element to participate in the distributed learning model training include these one or more features.

[0179] The first request may include K bits, where K is a positive integer, used to request the discovery of a network element. That is, the first request uses a single bit to request the discovery of a network element. Alternatively, the second network element may determine that the first request is for requesting the discovery of a network element based on the name of the first request or information about the corresponding service operation (such as its name). For example, if the second network element is an NRF, and the first request is sent to the NRF as an Nnrf_NF Discovery_Request service operation, then the NRF determines that the first network element is to be discovered based on the name of that service operation. In this case, there is no need for the first request to use a single bit to indicate that it is for discovering a network element, thus saving the number of bits occupied by the first request. Alternatively, one or more features may be used to request the discovery of a network element. In this case, the first request does not need to use a single bit to request the discovery of a network element, saving the bit overhead of the first request. In other words, if the first request indicates one or more features, it means that the first request is for requesting the discovery of a network element.

[0180] The first request may be indicated by any of the following items A1 to A5 to indicate one or more features. The contents of any of A1 to A5 are described below.

[0181] A1. The first request includes one or more features, including at least one of the following: location, average throughput, or average packet latency.

[0182] A2. The first request includes one or more feature identifiers (feature IDs) that are used to indicate the one or more features.

[0183] The identifier of a feature (such as the identifier of one of one or more features) can be an identifier assigned to that feature (also called a number, sequence number, or index, etc.). In one possible implementation, the identifier of one of these one or more features (which can be simply referred to as the feature identifier) ​​can be understood as an anonymous feature. The participants (such as the first network element and other participants) negotiate the meaning of the feature identifier in advance, so they can understand the meaning of the feature identifiers registered by each other. Optionally, the second network element cannot know which specific feature the feature identifier represents through the feature identifier, thus avoiding the disclosure of the features supported by each participant to other parties besides the participants in the distributed learning. In other words, the feature identifier can be understood by the participants in the distributed learning model, but not by the second network element, thus ensuring the security of the feature. In another possible implementation, the feature identifier can also be non-anonymous, and the participants do not need to negotiate the feature identifier in advance. This application embodiment does not impose any restrictions on this.

[0184] For example, a feature identified as 1234 indicates that the feature represents the average throughput; or a feature identified as 1235 indicates that the feature represents the average packet latency.

[0185] A3. The first request includes one or more event IDs. Each event ID is used to identify a list of local data for model training. The event IDs involved in the various embodiments of this application may be pre-configured or pre-defined, such as those pre-defined by a protocol, or they may be determined through negotiation between a first network element and at least one third network element, etc., and there is no limitation thereto.

[0186] An event identifier may correspond to (or can be identified as) at least one feature. Accordingly, one of one or more event identifiers may correspond to some or all of the features. For example, if an event identifier is Quality of Service (QoS) Monitoring, then the features corresponding to that event identifier include at least one of uplink packet delay (UL packet delay), downlink packet delay (DL packet delay), or round trip packet delay.

[0187] For example, the first request includes {event ID1, event ID2, event ID4, event ID5, feature ID3, feature ID4}, that is, one or more features include at least one feature corresponding to event 1, at least one feature corresponding to event 2, at least one feature corresponding to event 4, at least one feature corresponding to event 5, feature 3, and feature 4.

[0188] A4. The first request includes one or more network element types (NF type(s)) and / or one or more network element set IDs (NF Set ID(s)). The first request can indicate one or more characteristics through one or more network element types or one or more network element set IDs.

[0189] One or more network element types or one or more network element sets represent data sources that can be collected. Data corresponds to features, so indicating one or more network element types or one or more network element sets is equivalent to indicating one or more features.

[0190] For example, an NF type or an NF Set ID can correspond to multiple Event IDs, and an Event ID can correspond to one or more features. For instance, if NF type = SMF, the Event IDs that SMF can provide include, for example, QoS flow monitoring, uplink path change, and QoS flow identifier (QFI) allocation. Features corresponding to QoS flow monitoring include, for example, uplink data packet latency, downlink data latency, or round-trip data packet latency.

[0191] In one possible implementation, the first request may indicate one or more network element types (NF type(s)) in addition to indicating one or more features. For example, the first request may include identifiers of one or more network element types, which indicate these one or more network element types. In this implementation, each of these one or more features corresponds to a feature of one of the one or more network element types. Of course, if one or more features pass through one or more network element types in A4 above, then the first request can indicate both one or more features and one or more network element types by indicating these one or more network types.

[0192] When the first request indicates one or more features and one or more network element types, the first request can be described as indicating one or more network element types, and features corresponding to each of the one or more network element types. This can be understood as the first network element expecting these one or more features to be obtained from specific network element types, or as the first network element requesting one or more features at the granularity of network element type.

[0193] For example, if the first request indicates {NWDAF:(Event ID 1,Event ID 2,Event ID 4,Event ID 5),AF:(Feature ID3,Feature ID4)}, it means that the first network element expects to discover one or more NWDAF supporting features (Event ID 1,Event ID 2,Event ID 4,Event ID 5), and at the same time, the first network element expects to discover one or more AF supporting features {Feature ID3,Feature ID4}.

[0194] A5. The first request includes one or more capability types. Each capability type corresponds to at least one feature. Thus, one or more capability types indicate one or more features.

[0195] For example, the first request includes "active participant," which corresponds to features 1 and 2, indicating that one or more features include features 1 and 2.

[0196] In one possible implementation, the first request may indicate one or more features, as well as VFL capability types. This applies when the first request indicates one or more features using any of the methods described in A1-A4 above. Optionally, in this case, the first request may indicate one or more capability types, and the features corresponding to each capability type. Of course, if one or more features are specified through one or more network element types in A4 above, then the first request can indicate both one or more features and one or more network element types by indicating these one or more network element types.

[0197] For example, if the first request indicates that the capability type includes active participation, and indicates that the features corresponding to the active participation are features 3 and features 4, then it means that one or more features include features 3 and features 4.

[0198] The above A1 or A2 can be regarded as the first request indicating the fine-grained features to be discovered, that is, one or more features in this case can be regarded as fine-grained features. The above A3 to A5 can be regarded as the first request indicating the coarse-grained features to be discovered, that is, one or more features in this case can be regarded as coarse-grained features.

[0199] S702, the second network element sends first information to the first network element. Correspondingly, the second network element receives the first information from the first network element. The first information indicates at least one third network element, where M third network elements participate in the training of the distributed learning model. M is a positive integer, such as 1, 2, or 3, and is not limited thereto.

[0200] After receiving the first request, the second network element can determine (or sense, or discover) at least one third network element based on the first request. The second network element may have pre-configured or pre-defined information about at least one third network element, or it may obtain this information from at least one third network element; there is no specific limitation in this regard. Of course, in addition to determining at least one third network element, the second network element can also determine other network elements, such as the first network element, etc., without limitation in this regard.

[0201] The information of at least one third network element can include its identifier or address. Optionally, the information may also include its network element type and / or network element capabilities, without specific limitations. A third network element's capabilities may indicate, for example, that it possesses at least one of the following capabilities: distributed learning capability, training or inference of a distributed learning model. Distributed learning capability may be, for example, a vertical federated learning capability, which includes supported vertical federated learning capability types, such as a primary participant, a secondary participant, a collaborator, or a time window supported by vertical federated learning. A time window may be, for example, a time period for vertical federated learning, a time period for collecting data for vertical federated learning, or a time period for producing data for vertical federated learning. The identifier of a third network element may be its address, an identifier assigned to it, a result of calculations on its address, or its name. The address of the third network element is, for example, the Internet Protocol (IP) address of the third network element or the fully-qualified domain name (FQDN) of the third network element.

[0202] Since the second network element can identify at least one third network element, it can indicate this at least one third network element to the first network element. For example, the second network element can send a first message to the first network element, indicating this at least one third network element. The first message can also be regarded as a response message to the first request. At least one third network element can be understood as a third network element that satisfies the first request, or it can be understood as a network element discovered by the second network element according to the first request.

[0203] For example, the first information includes information about at least one third network element, such as the identifier of at least one third network element, which is equivalent to indicating at least one third network element.

[0204] In addition to indicating at least one third network element, the first information may optionally indicate features corresponding to at least one third network element. The features corresponding to at least one third network element include features corresponding to each of the at least one third network element. The features corresponding to at least one third network element may all be features among one or more features indicated in the first request, or the features corresponding to at least one third network element may include one or more features, as well as features other than one or more features. For example, the features corresponding to at least one third network element may include features supported by each of the at least one third network element; this is not specifically limited.

[0205] The following section uses the feature corresponding to one of the at least three third network elements as an example to introduce the content of the feature corresponding to at least one third network element.

[0206] B1. The feature corresponding to a third network element can be a feature supported by that third network element, such as all features supported by that third network element. For example, the features supported by that third network element can be all features that the third network element can provide.

[0207] For example, if all the features supported by the third network element include feature 1, feature 2 and feature 3, then the features corresponding to the third network element include feature 1, feature 2 and feature 3.

[0208] Under B1, the features corresponding to any two third network elements may or may not overlap; no specific limitation is made. Overlapping features between any two third network elements means that at least two third network elements have the same features or completely identical features. Non-overlapping features between any two third network elements means that the features corresponding to any two third network elements are different.

[0209] For example, at least one third network element includes third network element A and third network element B. The features corresponding to third network element A include feature 1, feature 2, and feature 3, while the features corresponding to third network element B include feature 1 and feature 4. Since feature 1 corresponding to third network element A is the same as feature 1 corresponding to third network element B, it can be considered that the features corresponding to third network element A and third network element B overlap. Alternatively, at least one third network element includes third network element C and third network element D. The features corresponding to third network element C include feature 1 and feature 2, while the features corresponding to third network element D include feature 4 and feature 5. Since the features corresponding to third network element C and third network element D are all different, it can be considered that the features corresponding to third network element C and third network element D do not overlap.

[0210] Under B1, the features corresponding to at least one third network element include one or more features indicated by the first request. Optionally, the features corresponding to at least one third network element may also include other features besides one or more features.

[0211] B2. If a feature corresponding to a third network element is a feature supported by the third network element and belongs to one or more features indicated by the first request, then the feature corresponding to a third network element is the intersection of the feature supported by the third network element and one or more features.

[0212] For example, if one or more features include feature 1 and feature 2, and all features supported by the third network element include feature 1, feature 2 and feature 3, then the features corresponding to the third network element are feature 1 and feature 2.

[0213] Under B2, the features corresponding to any two third network elements in at least one third network element may or may not overlap, and no specific limitation is made in this regard.

[0214] Under B2, the features corresponding to at least one third network element belong to one or more features.

[0215] B3. The features corresponding to a third network element are the features used by the third network element to participate in the training of the distributed learning model. The features used by the third network element to participate in the training of the distributed learning model can be understood as the features of the third network element determined by the second network element to participate in the training of the distributed learning model.

[0216] For example, one or more features include feature 1, feature 2, feature 3, and feature 4, and at least one third network element includes third network element A and third network element B. Third network element A supports feature 1, feature 2, and feature 3, and third network element B supports feature 3, feature 4, and feature 5. The second network element determines that third network element A uses feature 1 and feature 2 for distributed learning model training, and third network element B uses feature 3 and feature 4 for distributed learning model training. Therefore, for third network element A, the second network element only sends feature 1 and feature 2 to the first network element, or it can be described as the second network element instructing the first network element to use the corresponding features 1 and feature 2 of third network element A; for third network element B, the second network element only sends feature 3 and feature 4 to the first network element, or it can be described as the second network element instructing the first network element to use the corresponding features 3 and feature 4 of third network element B. The union of the features of third network element A and third network element B constitutes one or more features, and there is no intersection between the features of third network element A and third network element B.

[0217] In this case, the feature corresponding to one of the third network elements is part or all of the intersection of the features supported by the third network element and one or more features.

[0218] Under B3, the features corresponding to any two third network elements in at least one third network element may or may not overlap, and no specific limitation is made in this regard.

[0219] Optionally, under B3, the features corresponding to each of the at least one third network element can be features from one or more features, and the features corresponding to any two third network elements do not overlap. In this optional approach, it is equivalent to the second network element being at least one third network element determined by the first network element, which is a third network element that can be directly used to participate in the training of the distributed learning model, without the first network element needing to further select from at least one third network element.

[0220] The following provides an example illustrating the specific content of the first information when it also indicates the characteristics corresponding to at least one third network element.

[0221] C1. The first information includes the identifier of at least one third network element, and the features corresponding to each of the at least one third network element. The content of the features corresponding to a third network element can be referred to the content of the features corresponding to a third network element discussed earlier. Thus, the first network element can intuitively identify the features corresponding to at least one third network element.

[0222] For example, if the first information includes {AF1: feature 1, feature 2 and feature 3, UE2: feature 4 and feature 5, AMF1: feature 1, feature 5 and feature 6}, then the at least one third network element indicated by the first information includes AF1, UE2 and AMF1. The features corresponding to AF1 include feature 1, feature 2 and feature 3, the features corresponding to UE2 include feature 4 and feature 5, and the features corresponding to AMF2 include feature 1, feature 5 and feature 6.

[0223] C2. The first information includes information about at least one network element group. The information of each network element group includes the identifiers of the network elements included in each network element group, and the features corresponding to each network element group include one or more features. Optionally, the features corresponding to any two network elements in each network element group do not overlap, although the features corresponding to any two network elements may overlap.

[0224] At least one network element group includes at least one third network element, and correspondingly, one network element group within the at least one network element group includes some or all of the network elements in the at least one third network element. This is equivalent to the second network element grouping at least one third network element, facilitating the first network element's subsequent selection of network elements to participate in the training of the distributed learning model.

[0225] Optionally, the information for each network element group may also include the characteristics corresponding to each network element group, and / or the information for each network element group may also include the identifier of each network element group. The characteristics corresponding to at least one network element group are the same as the characteristics corresponding to at least one third network element. Furthermore, the characteristics corresponding to a network element group may include one or more features.

[0226] For example, the first information includes {Group 1: VFL client 1: Event ID 1, Event ID 4; VFL client 2: Event ID 2, Event ID 5; VFL client 3: Feature ID 3, Feature ID 4} and {Group 2: VFL client 4: Event ID 1, Event ID 2, Event ID 4, Event ID 5; VFL client 5: Feature ID 3; VFL client 6: Feature ID 4}. In this case, the first information indicates the information of two network elements (i.e., network element 1 and network element 2).

[0227] Network element group 1 includes VFL client 1, VFL client 2, and VFL client 3. The features corresponding to VFL client 1 include the features corresponding to event 1 and event 4. The features corresponding to VFL client 2 include the features corresponding to event 2 and event 5. The features corresponding to VFL client 3 include features 3 and 4. Therefore, the features corresponding to network element group 1 include the features corresponding to VFL client 1, VFL client 2, and VFL client 3. Network element group 2 includes VFL client 4, VFL client 5, and VFL client 6. The features corresponding to VFL client 4 include the features corresponding to event 1, event 2, event 4, and event 5. The features corresponding to VFL client 5 include feature 3. The features corresponding to VFL client 6 include feature 4. Therefore, the features corresponding to network element group 2 include the features corresponding to VFL client 4, VFL client 5, and VFL client 6.

[0228] In one possible implementation, if the first request further indicates one or more network element types, then the network element type of at least one third network element can belong to one or more network element types, and the feature corresponding to each of the at least one third network element belongs to one or more features. If the first information is the information shown in C2 above, then the network element type corresponding to at least one network element group belongs to one or more network element types. For example, the network element type corresponding to at least one network element group belongs to one of the one or more network element types, without specific limitation.

[0229] In one possible design, after the second network element sends the first information to the first network element, the first network element can include at least one third network element as another participant in the distributed model learning and training. In this case, the M third network elements are all the network elements in the at least one third network element.

[0230] For example, if the first information does not indicate the feature corresponding to at least one third network element, the first network element can also negotiate with each of the at least one third network element the features supported by each third network element.

[0231] Alternatively, if the first information also indicates a feature corresponding to at least one third network element, then the first network element can determine that the feature corresponding to the at least one third network element is a feature participating in the distributed model learning and training.

[0232] Alternatively, if the first information also indicates features corresponding to at least one third network element, and the first request indicates coarse-grained features to be discovered (e.g., the first request indicates one or more features using the methods described in A3 to A5 above (e.g., the one or more features indicated by the first request could be Event ID, NF type(s), NF Set ID(s), or capability type, etc.)), the first network element can further negotiate with each of the at least one third network element the features supported by each third network element. Optionally, the first network element can negotiate with each of the at least one third network element more fine-grained features for distributed model training, etc.

[0233] In another possible implementation, the first network element determines M third network elements from at least one third network element based on the first information. In this case, the M third network elements may be all network elements of the at least one third network element, or they may be a portion of the network elements of the at least one third network element, without specific limitation. These M third network elements can be understood as the network elements determined by the first network element that participate in the training of the distributed learning model.

[0234] The methods for determining the M third network elements differ depending on the content of the first information, which will be described below.

[0235] D1. If the first information also indicates the features corresponding to at least one third network element, the first network element can directly determine the features corresponding to at least one third network element based on the first information.

[0236] When the first information indicates information about at least one third network element and features corresponding to each of the at least one third network element, the first network element can determine M third network elements whose corresponding features include one or more features from the at least one third network element, that is, determine the M third network elements participating in the training of the distributed learning model.

[0237] Optionally, if the feature corresponding to one of the at least three third network elements is a feature used by the third network element to participate in the training of the distributed learning model, the first network element can directly determine at least one third network element as M third network elements, which can reduce the processing load of the first network element.

[0238] Alternatively, the first network element may determine multiple candidate network element groups from at least one third network element. Each candidate network element group includes some or all of the network elements in at least one third network element, and the features corresponding to each candidate network element group include one or more features. The first network element may determine one candidate network element group from multiple candidate network element groups, and the network elements included in this determined candidate network element group are M third network elements.

[0239] The determined candidate network element group (i.e., the candidate network element group corresponding to the M third network elements) can be the network element group with the smallest sum of distances to the first network element, which helps to shorten the communication time. Alternatively, the determined candidate network element group can be the network element group with the smallest overall load, which helps to achieve load balancing of the communication system. The determined candidate network element group can also be the network element group with the best overall communication quality with the first network element, which helps to improve the efficiency of distributed learning model training.

[0240] D2. If the first information includes information about at least one network element group, the first network element can directly select one network element group from the at least one network element group. The selected network element group includes M third network elements. The content of the selected network element group can be referred to the content of determining a candidate group above, and will not be listed one by one here.

[0241] D3. If the first information only indicates at least one third network element (or can be understood as the first information not indicating the feature corresponding to at least one third network element), the first network element can request the feature corresponding to at least one third network element from at least one third network element, and then determine M third network elements based on the feature corresponding to at least one third network element.

[0242] For example, a first network element sends third information to at least one third network element. This continues, with the first network element sending at least one third message in total. The third message sent to one of the at least three third network elements is used to request feedback from that third network element regarding its corresponding feature. The meaning of the feature corresponding to the third network element can be found in the previous section on features corresponding to third network elements, and will not be listed here. Thus, at least one third network element can send fourth information to the first network element. This continues, with the first network element receiving at least one fourth message in total. The fourth message received from one of the at least three third network elements indicates the feature corresponding to each third network element.

[0243] Under D3, the method for determining M third network elements can be referred to the content on determining M third network elements under D1 above, and the repetition will not be listed again.

[0244] When determining M third network elements, the first network element may optionally also determine the features of each of the M third network elements used to participate in the training of the distributed learning model. The features of each third network element used to participate in the training of the distributed learning model may be some or all of the features corresponding to each third network element.

[0245] In one possible design, the first network element can send second information to M third network elements respectively. This continues until the first network element sends a total of M pieces of second information. The first network element can send the second information directly to the M third network elements. Alternatively, the first network element can send the second information to the M third network elements through a collaborator (such as a fourth network element). Alternatively, the first network element can act as both the initiator and collaborator, in which case it can also directly send the second information to the M third network elements. The second information sent to one of the M third network elements is used to indicate the features that the third network element uses for participating in the distributed learning model training. Optionally, the features used for participating in the distributed learning model training are used to instruct the third network element to participate in the distributed learning model training. Alternatively, the second information can also instruct the third network element to participate in the distributed learning model training.

[0246] Further, optionally, the M third network elements can each send a first response message to the first network element. This continues, with the first network element receiving a total of M first response messages. One of the first response messages sent by a third network element to the first network element indicates that the third network element participates in the distributed learning model training (or is described as accepting participation in distributed learning model training), or does not participate in the distributed learning model training (or is described as not accepting participation in distributed learning model training). If the second information received by a third network element also indicates a feature used by the third network element for participating in the distributed learning model training, the first response message sent by that third network element can also be specifically understood as accepting the feature used for participating in the distributed learning model training for distributed learning model training, or not accepting participation in the distributed learning model training for the feature used for participating in the distributed learning model training.

[0247] This application provides a mechanism for negotiating (or aligning) network elements participating in the training of a distributed learning model. A first network element can flexibly discover other participants in the training through a second network element. Furthermore, the first network element and the determined M third network elements can negotiate or align features used for participating in the distributed learning model training, ensuring the accuracy of subsequent distributed learning model training.

[0248] The following example, using the first request indicating one or more features, is illustrated with a schematic diagram of a communication method shown in Figure 8, to illustrate the communication method involved in Figure 7. The number of at least one third network element involved in Figure 8 can be arbitrary and is not limited.

[0249] S801, the first network element sends the fifth information to the second network element. Correspondingly, the second network element receives the fifth information from the first network element. The fifth information indicates the features supported by the first network element.

[0250] The fifth piece of information can be understood as a request to register the first network element; it can also be called a registration request for the first network element. The content of the features supported by the first network element can be referenced to the content of the features corresponding to the third network element mentioned earlier, and will not be listed here. For example, the fifth piece of information indicates the name (Feature name(s)) and / or identifier of the features supported by the first network element. In addition to indicating the features supported by the first network element, the fifth piece of information may optionally also indicate at least one of the following: the identifier of the first network element, the network element type, the network element capability, the network element type (NF Type(s)) corresponding to the first network element, or the network element set ID(s) corresponding to the first network element. The content of the identifier of the first network element can be referenced to the identifier of the third network element mentioned earlier, and will not be listed here. The type of the first network element, for example, indicates that the first network element is AF. The network element capability of the first network element, for example, indicates that the network element has at least one of the following capabilities: distributed learning capability, training or inference of a distributed learning model, etc. The network element set ID refers to the identifier of the network element that the first network element supports in acquiring data.

[0251] The following is an example illustrating the content of the fifth piece of information.

[0252] [Corrected according to Rule 91 04.03.2025] G1, the fifth information includes the network element type of the first network element, the network element set identifier corresponding to the first network element, the network element type corresponding to the first network element, and the event identifier corresponding to the feature supported by the first network element.

[0253] For example, if the first network element is NWDAF and its corresponding network element type includes AMF, then it means that NWDAF supports collecting data from AMF. Alternatively, if the first network element is NWDAF and its corresponding network element type includes SMF, then it means that NWDAF supports collecting data from SMF. Furthermore, if the first network element is NWDAF and Event ID = location report, it means that NWDAF supports collecting UE location information from AMF. Another example is Event ID = Change of RAT Type, which indicates that NWDAF supports collecting RAT Type change information from AMF. Yet another example is if the network elements corresponding to network element set identifier 1 for the first network element include AMF1, AMF2, and AMF3, then it means that the first network element supports collecting data from AMF1, AMF2, and AMF3.

[0254] [Corrected according to Rule 91, 04.03.2025] G2. The fifth piece of information includes the feature name and the feature identifier. The meaning of the feature identifier can be referred to the content of the feature identifier discussed in Figure 7 above; repeated instances will not be listed again. Feature names include, for example, service experience, location, average throughput, average packet latency, etc. For example, Feature ID = 1234 indicates that the feature is average throughput, and Feature ID = 1235 indicates that the feature is average packet latency.

[0255] [Corrected according to detailed rule 91 04.03.2025] The above G1 or G2 can be applied to any type of network element, such as NWDAF, 5GC NF (e.g., AMF, SMF, UPF, PCF, etc.) or AF, without any limitation.

[0256] After receiving the fifth piece of information, the second network element can store the fifth piece of information and send a first registration response to the first network element. The first registration response indicates whether the first network element has successfully registered or has not successfully registered.

[0257] If the first network element has already been registered in the second network element, or if the first network element does not need to be registered in the second network element, then step S801 does not need to be executed. That is, step S801 is optional, and it is shown as a dashed line in Figure 8.

[0258] S802, at least one third network element sends the sixth information to the second network element. Correspondingly, the second network element receives the sixth information from at least one third network element.

[0259] One of the sixth pieces of information indicates the features supported by the third network element corresponding to that sixth piece of information. The sixth piece of information can be understood as being used to request registration of the third network element. For example, the sixth piece of information indicates the name and / or identifier of the features supported by the third network element. Optionally, in addition to indicating the features supported by the third network element corresponding to that sixth piece of information, one of the sixth pieces of information may also indicate the identifier of the third network element. Optionally, in addition to indicating the features supported by the third network element, the sixth piece of information may also indicate at least one of the following: the identifier of the third network element, the network element type, the capability, the network element type corresponding to the third network element, or the network element set identifier corresponding to the third network element. The capabilities of the third network element, the network element type corresponding to the third network element, and the network element set identifier corresponding to the third network element can be referred to respectively with reference to the capabilities of the first network element, the network element type corresponding to the first network element, and the network element set identifier corresponding to the first network element discussed above.

[0260] In cases where at least one third network element has already been registered in the second network element, step S802 may not be necessary, i.e., step S802 is optional, and is shown as a dashed line in Figure 8.

[0261] The execution order of S801 and S802 can be arbitrary, such as executing them simultaneously, or executing S801 first and then S802, or executing S802 first and then S802. There are no specific restrictions on this.

[0262] S803, the first network element sends a first request to the second network element. Correspondingly, the second network element receives the first request from the first network element. In this embodiment, the first request is used to request the discovery of network elements and indicates one or more characteristics. Of course, the first request can also indicate other information (such as one or more network element types), and this is not limited.

[0263] The content of the first request and the content of one or more features can be referred to in Figure 7 above, respectively, and repeated parts will not be listed again.

[0264] S804, the second network element sends first information to the first network element. Correspondingly, the first network element receives the first information from the second network element. The first information includes information from at least one group of network elements. The information from at least one group of network elements is used to indicate at least one third network element.

[0265] The content of the first information, the information of at least one network element group, and the content of at least one third network element can be referred to the content of the first information, the information of at least one network element group, and the content of at least one third network element discussed in Figure 7 above. Repeated parts will not be listed again.

[0266] S805. The first network element determines M third network elements from at least one network element group based on the first information.

[0267] The contents of the M third network elements and the contents for determining the M third network elements can be referred to the contents of the M third network elements and the contents for determining the M third network elements discussed in Figure 7 above. The repetitive parts will not be listed again.

[0268] In another possible implementation, the first network element can also directly use at least one third network element as a participant in the training of the distributed learning model. In this case, the first network element does not need to perform step S805, that is, S805 is an optional step, which is shown by dashed lines in Figure 8.

[0269] S806. The first network element sends second information to M third network elements. Correspondingly, the M third network elements receive the second information from the first network element. The second information indicates the features used to participate in the training of the distributed learning model.

[0270] The first network element can directly send the second information to M third network elements respectively. The first network element can also send the second information to M third network elements respectively through a cooperating party (such as the fourth network element). There are no specific restrictions on this.

[0271] The content of the second information, the content of the features used to participate in the training of the distributed learning model, and the content of sending the second information can be referred to the content of the second information, the content of the features used to participate in the training of the distributed learning model, and the content of sending the second information discussed in Figure 7 above, respectively. Repeated parts will not be listed again.

[0272] In one possible implementation, the first network element determines to trigger the distributed learning model training preparation phase and determines the second information. The second information can be regarded as preparation information for distributed learning model training. For example, if the distributed learning model training is VFL model training, then the preparation information can be called the VFL Training Preparation Information.

[0273] In one possible design, the second information sent to a third network element, indicating the features used by that third network element for participating in distributed learning model training, can refer to the second information in Figure 7 above, which describes the features used by a third network element for participating in distributed learning model training; repetitions will not be listed here. Optionally, the second information sent to a third network element may include at least one of the following: a distributed learning model training type identifier, a preparation stage identifier, a distributed learning model training ID, model requirements, or alignment information for each third network element's distributed learning model training. The alignment information includes samples and / or features for each third network element's distributed learning model training. When the second information includes alignment information, and the alignment information includes features for each third network element's distributed learning model training, it is equivalent to the second information indicating the features for each third network element's distributed learning model training.

[0274] The type identifier for distributed learning model training indicates the type of distributed learning model training, such as longitudinal federated learning between NWDAF and AF, longitudinal federated learning between NWDAFs, or longitudinal federated learning between AFs assisted by NWDAF. The preparation phase identifier indicates the preparation phase of the distributed learning model training to be performed. The distributed learning model training ID represents a unique identifier for the training process. Model requirements include, for example, the model's output category, such as discrete values ​​(as associated with classification algorithms), or real values ​​(as associated with regression algorithms).

[0275] In this way, the initiator exchanges information with each participant to adjust the model training process to be performed by each participant. During the training preparation phase, all network elements receive preparation information indicating that the training of their local models (e.g., passive participants) is associated with the initiator's model training. This enables each participant to map and control the information required to achieve distributed learning model training.

[0276] As an example, S806 can be used as a standalone embodiment, and the content of Figure 8 other than S806 can be used as an optional implementation in this embodiment.

[0277] S807 and M third network elements respectively send first response messages to the first network element. Correspondingly, the first network element receives first response information from each of the M third network elements. The first response message indicates whether or not to participate in the training of the distributed learning model.

[0278] Each of the M third network elements determines, based on the received second information and locally available relevant configurations, whether it can execute the distributed learning model training process requested by the first network element. The relevant configurations include at least one of the following: the type of distributed learning model training it can participate in (e.g., in NWDAF; NWDAF-AF; or NWDAF Assisting AFs, etc.), the supported roles in distributed learning model training (e.g., master participant, slave participant, or collaborator, etc.), or the scope information of the distributed learning model. The type of distributed learning model training can be understood as the type of participant supported by the distributed learning model training process. The scope information of the distributed learning model includes, for example, interoperability metrics for distributed learning model training of participating vendors, and supported analysis filter information (e.g., area of ​​interest (AoI), slice identification, or application identification).

[0279] If a third network element can execute the distributed learning model training process requested by the first network element, then the first network element can perform the requested preparatory work, and the third network element can send a first response message indicating participation to the first network element. If it cannot execute the distributed learning model training process requested by the first network element, then the third network element can send a first response message indicating non-participation in the distributed learning model training to the first network element. Of course, the third network element can also send the first response message directly to the first network element, or it can send the first response message through a collaborating party (such as a fourth network element). The remaining content of the first response message can be referred to the content of the first response message discussed in Figure 7 above; repetitions will not be listed again.

[0280] Following S807, the first network element determines the final participants in the distributed learning model training based on the first response messages from M third network elements. For example, if a third network element indicates in its first response message that it will not participate in the distributed learning model training, the first network element will not select that third network element to participate. If a third network element indicates in its first response message that it will participate in the distributed learning model training, the first network element can select that third network element to participate. This process continues to determine the participants in the distributed learning model training.

[0281] This application provides an embodiment whereby a first network element can indicate one or more desired features to a second network element to discover participants supporting those features, thereby completing the feature alignment process. This improves the flexibility in determining participants and the flexibility in feature alignment. Furthermore, this application also provides a mechanism for participants (such as the first network element and at least one third network element) to register with the second network element.

[0282] The following example, using the first request indicating one or more features, is illustrated with a schematic diagram of a communication method shown in Figure 9, to illustrate the communication method involved in Figure 7. The number of at least one third network element involved in Figure 9 can be arbitrary and is not limited thereto.

[0283] S901, the first network element sends the fifth information to the second network element. Correspondingly, the second network element receives the fifth information from the first network element. The fifth information indicates the features supported by the first network element.

[0284] The content of the fifth information can be referred to in Figure 8 above, and the repeated parts will not be listed again.

[0285] After receiving the fifth piece of information, the second network element can store the fifth piece of information and send a first registration response to the first network element. The first registration response indicates whether the first network element has successfully registered or has not successfully registered.

[0286] If the first network element has already been registered in the second network element, or if the first network element does not need to be registered in the second network element, then step S901 does not need to be executed. That is, step S901 is optional, and it is shown as a dashed line in Figure 9.

[0287] S902, at least one third network element sends the sixth information to the second network element. Correspondingly, the second network element receives the sixth information from at least one third network element.

[0288] The content of the sixth information can be referred to the content of the sixth information discussed in Figure 8 above, and the repeated parts will not be listed again.

[0289] In cases where at least one third network element has already been registered in the second network element, step S902 may not be necessary, i.e., step S902 is optional, and is shown as a dashed line in Figure 9.

[0290] The execution order of S901 and S902 can be arbitrary, such as executing them simultaneously, or executing S901 first and then S902, or executing S902 first and then S902. There are no specific restrictions on this.

[0291] S903: The first network element sends a first request to the second network element. Correspondingly, the second network element receives the first request from the first network element. In this embodiment, the first request is used to request the discovery of network elements and indicates one or more network element types, as well as the characteristics corresponding to each network element type. The characteristics corresponding to one or more network element types are one or more features. S903 can be understood as the first network element specifying the characteristics of the network element type in the first request, or it can be described as specifying the network element type corresponding to the desired characteristics.

[0292] The content of the first request, the content of one or more network element types, and the content of one or more features can be referred to in Figure 7 above, respectively, and repeated parts will not be listed again.

[0293] S904, the second network element sends first information to the first network element. Correspondingly, the first network element receives the first information from the second network element. The first information indicates at least one third network element. The content of the first information and the content of at least one third network element can be referred to in Figure 7 above for the content of the first information and the content of at least one third network element, respectively; repeated details will not be listed again.

[0294] For example, since the first request specifies the network element type corresponding to the desired feature, the second network element can also discover the network element if it meets the requirements of the corresponding network element type. In this case, the content of the first information can be referred to the content of the first information discussed in Figure 7 above, and the repetition will not be listed again. When the first information includes information of at least one network element group, the network element types of the network elements included in one network element group all belong to one or more network element types. Optionally, a network element group may include features corresponding to network elements under one or more network element types.

[0295] For example, taking the case where one or more network element types include NWDAF, and the first information includes information on 3 network element groups (such as network element group 1, network element group 2, and network element group 3). The information of network element group 1 includes {NWDAF1: Event ID 1, Event ID 4>, <NWDAF2: Event ID 2, Event ID 5}, the information of network element group 2 includes {NWDAF3: Event ID 1, Event ID 2, Event ID 4, Event ID 5}, and the information of network element group 3 includes {NWDAF4: Event ID 1, Event ID 2, Event ID 4, Event ID 5}. ""

[0296] Alternatively, taking the case where one or more network element types include AF, and the first information includes information on 3 network element groups (such as network element group 1, network element group 2, and network element group 3). The information of network element group 1 includes {AF1: Feature ID3, AF2: Feature ID4}, the information of network element group 2 includes {AF3: Feature ID3, Feature ID4}, and the information of network element group 3 includes {AF4: Feature ID3, Feature ID4}.

[0297] S905. The first network element determines M third network elements from at least one third network element according to the first information.

[0298] The content of the M third network elements and the content of determining the M third network elements can be respectively referred to the content of the M third network elements and the content of determining the M third network elements discussed in Figure 7 above. Duplicate parts will not be listed again.

[0299] In another possible implementation, the first network element can also directly use at least one third network element as a participant in the training of the distributed learning model. In this case, the first network element does not need to execute the step of S905, that is, S905 is an optional step, which is shown by a dashed line in Figure 9.

[0300] S906. The first network element sends the second information to the M third network elements. Correspondingly, the M third network elements receive the second information from the first network element. The second information indicates the features used to participate in the training of the distributed learning model.

[0301] The content of the second information, the content of the features used to participate in the training of the distributed learning model, and the content of sending the second information can be respectively referred to the content of the second information and the content of the features used to participate in the training of the distributed learning model discussed in Figure 8 above. Duplicate parts will not be listed again.

[0302] S907 and M third network elements respectively send first response messages to the first network element. Correspondingly, the first network element receives first response information from each of the M third network elements. The first response message indicates whether or not to participate in the distributed learning model training. The content of the first response message can be referred to the first response message discussed in Figure 7 above; repetitions will not be listed again.

[0303] Following S907, the first network element determines the final participants in the distributed learning model training based on the first response messages from M third network elements. For example, if a third network element indicates in its first response message that it will not participate in the distributed learning model training, the first network element will not select that third network element to participate. If a third network element indicates in its first response message that it will participate in the distributed learning model training, the first network element can select that third network element to participate. This process continues to determine the participants in the distributed learning model training.

[0304] This application provides an embodiment where a first network element can indicate one or more desired features and one or more network element types to a second network element, thereby discovering participants that support the corresponding features and conform to the corresponding network element types. This improves the flexibility in determining participants and ensures that the determined participants meet the requirements of distributed learning model training. Furthermore, the first network element can also align features with M third network elements, providing a feature alignment mechanism.

[0305] To improve the flexibility of determining participants, embodiments of this application also provide a communication method. In this method, an initiator of distributed learning (such as a first network element) can identify at least one third network element. The first network element can then request supported features from the at least one third network element, and based on the features supported by the at least one third network element, determine the final participants in the distributed learning (e.g., M network elements) from among the at least one third network element. Thus, without the need for pre-determining participants, a mechanism for online determination of distributed participants is implemented, which improves the flexibility of participant selection.

[0306] The communication method provided in the embodiments of this application will now be described in conjunction with the accompanying drawings.

[0307] Figure 10 illustrates a communication method provided by an embodiment of this application. The steps involved in Figure 10 will be described below.

[0308] S1001, The first network element determines at least one third network element.

[0309] At least one third network element can be considered as a candidate network element for participating in the training of the distributed learning model. The first network element may pre-store information about at least one third network element, such as the identifier or address of at least one third network element. The identifier of one of the third network elements can refer to the identifier of the third network element discussed in Figure 7 above, and will not be repeated. In this way, the first network element can determine the at least one third network element based on the pre-stored information of the at least one third network element. Alternatively, the first network element can receive fourth information from the second network element, which indicates the information of at least one third network element. In this way, the first network element can determine the at least one third network element based on the fourth information.

[0310] S1002, the first network element sends first information to at least one third network element respectively. Correspondingly, at least one third network element receives the first information from the first network element respectively.

[0311] The first message sent to a third network element is used to request features supported by that third network element, or can be described as a request for the third network element to provide feedback on the features it supports. The content of the features supported by the third network element can be referred to in Figure 7 above, and will not be repeated. Optionally, the first message includes a feature reporting indication, which is used to request the third network element to provide feedback on the features it supports.

[0312] S1003. At least one third network element sends second information to the first network element. Correspondingly, the first network element receives second information from at least one third network element. The second information sent by one of the third network elements to the first network element indicates the features supported by that third network element.

[0313] For example, the way in which one of the second pieces of information indicates the features supported by the third network element corresponding to that second piece of information can refer to the content of the first request indicating one or more features discussed above, and will not be listed one by one here. For example, a third network element can register coarse-grained features (e.g., NF type(s) and / or NF set ID(s), or event ID(s)) with a first network element, or a third network element can register fine-grained features with a first network element.

[0314] S1004. The first network element determines the features of M third network elements used to participate in the training of the distributed learning model.

[0315] The method for determining M third network elements by the first network element can refer to the content on determining M third network elements discussed in Figure 7 above, and the repetition will not be listed again. Alternatively, the first network element can use at least one third network element as M third network elements.

[0316] Since the second information indicates the features supported by at least one third network element, the first network element can determine the features of M third network elements to participate in the training of the distributed learning model according to the training requirements of the distributed learning model.

[0317] For example, when the first network element determines at least one third network element, it can already clearly identify the fine-grained features supported by the third network element. Therefore, the first network element can determine M features from the fine-grained features supported by the M third network elements that the third network element will use for training the distributed learning model. Alternatively, if the second information indicates that the third network element needs to support coarse-grained features, the first network element can negotiate with the third network element through S1002 and S1003 to determine the fine-grained features (e.g., feature(s), feature ID(s), or event ID(s)) needed for model training, thereby determining M features for the third network element to use for training the distributed learning model.

[0318] In one possible implementation, M third network elements each send a first response message to a first network element. Correspondingly, the first network element receives first response information from each of the M third network elements. The first response message indicates whether or not to participate in the training of the distributed learning model.

[0319] Each of the M third network elements determines, based on the received second information and the relevant local configuration, whether it can execute the distributed learning model training process requested by the first network element. The relevant configuration includes at least one of the following: the type of distributed learning model training it can participate in, the supported VFL roles, or the scope information of the distributed learning model.

[0320] If a third network element can execute the distributed learning model training process requested by the first network element, then the first network element can perform the requested preparatory work, and the third network element can send a first response message indicating participation to the first network element. If it cannot execute the distributed learning model training process requested by the first network element, then the third network element can send a first response message indicating non-participation in the distributed learning model training to the first network element. Of course, the third network element can also send the first response message directly to the first network element, or it can send the first response message through a collaborating party (such as a fourth network element).

[0321] In this embodiment, the first network element can directly determine at least one third network element, and from these third network elements, determine the participants ultimately involved in the training of the distributed learning model, thus improving the flexibility of participant determination. Furthermore, at least one third network element can register in the second network element, but it does not register the features supported by this at least one third network element, reducing the risk of leakage of features supported by at least one third network element.

[0322] The following example illustrates the communication method involved in Figure 10, using the first network element determining at least one third network element based on the fourth information. Figure 11 shows a schematic diagram of such a communication method.

[0323] S1101, the first network element sends the fifth information to the third network element. The fifth information indicates the information of the first network element. Correspondingly, the third network element receives the fifth information from the first network element. The fifth information indicates the information of the first network element.

[0324] The information of the first network element includes its identifier or address. Optionally, the information of the first network element may also include its network element type and / or network element capabilities. The content of the network element capabilities can be referred to the content of the network element capabilities discussed in Figure 8 above, and repeated parts will not be listed again.

[0325] S1102. At least one third network element sends a sixth message to the second network element. Correspondingly, the second network element receives the sixth message from at least one third network element. The sixth message sent by one of the third network elements to the second network element indicates the information of that third network element.

[0326] The information of the third network element includes its identifier or address. Optionally, the information of the third network element may also include its network element type and / or network element capabilities.

[0327] S1103, the second network element sends fourth information to the first network element. Correspondingly, the first network element receives the fourth information from the second network element. The fourth information indicates information from at least one third network element. The information from at least one third network element includes the identifier or address of at least one third network element, etc.

[0328] S1104. The first network element determines at least one third network element based on the fourth information.

[0329] Since the fourth information indicates information about at least one third network element, the first network element can directly determine at least one third network element based on the fourth information.

[0330] S1105. The first network element sends first information to at least one third network element. Correspondingly, at least one third network element receives the first information from the first network element. The first information sent to a third network element is used to request a feature supported by the third network element, or can be described as requesting the third network element to send (or provide feedback) a feature supported by the third network element, or can be described as requesting to obtain a feature supported by the third network element.

[0331] S1106. At least one third network element sends second information to the first network element. Correspondingly, the first network element receives the second information from at least one third network element. The second information indicates the features supported by the third network element.

[0332] For example, the second information includes the identifier or address of the third network element, as well as the features supported by the third network element.

[0333] S1107. The first network element determines the features of M third network elements to participate in the training of the distributed learning model based on the second information.

[0334] The content regarding the first network element determining the features of M third network elements used in the training of the distributed learning model can be referred to in Figure 10 above, where the content regarding the first network element determining the features of M third network elements used in the training of the distributed learning model is discussed. Repeated parts will not be listed again.

[0335] For example, at least one third network element can register coarse-grained features (e.g., NF type(s) and / or NF set ID(s), or event ID(s)) with the first network element. The first network element can discover at least one third network element through the second network element. The network element discovery request indicates the coarse-grained features that the third network element needs to support. After discovering the third network element, the first network element negotiates with the third network element through S1105 and S1106 for the fine-grained features (e.g., feature(s), feature ID(s), or event ID(s)) needed for model training. Then, M features for the third network elements to participate in the distributed learning model training are determined.

[0336] S1108. The first network element sends third information to M third network elements among at least one third network element. Correspondingly, the M third network elements receive the third information from the first network element. The third information sent to one of the third network elements indicates the features that the third network element uses to participate in the training of the distributed learning model. This achieves feature alignment.

[0337] The first network element can directly send third information to each of the M third network elements, or it can send third information to each of the M third network elements through a collaborating party (such as a fourth network element), without any specific limitation. For example, the first network element can determine the features used by each of at least one third network element for participating in the training of the distributed learning model, and indicate the features used by that third network element for participating in the training of the distributed learning model in the third information sent by one of the third network elements.

[0338] In one possible implementation, the first network element determines to trigger the distributed learning model training preparation phase and determines the third information. This third information can be considered preparation information for distributed learning model training; for example, if the distributed learning model training is VFL model training, then the preparation information can be called the VFL Training Preparation Information.

[0339] In one possible design, the way a third piece of information sent to a third network element indicates the features used by that third network element for participating in distributed learning model training can refer to the content of a second piece of information indicating the features used by a third network element for participating in distributed learning model training, as discussed in Figure 7 above, and will not be listed here. Optionally, the third information may include at least one of the following: a distributed learning model training type identifier, a preparation stage identifier, a distributed learning model training ID, model requirements, or alignment information for each third network element's distributed learning model training, etc. The content of the distributed learning model training type identifier, preparation stage identifier, distributed learning model training ID, model requirements, and alignment information for each third network element's distributed learning model training can refer to the content of the distributed learning model training type identifier, preparation stage identifier, distributed learning model training ID, model requirements, and alignment information for each third network element's distributed learning model training, as discussed in Figure 8 above, and will not be listed here. When the third information includes alignment information, and the alignment information includes the features of each third network element's distributed learning model training, it is equivalent to the third information indicating the features of each third network element's distributed learning model training.

[0340] As an example, S1108 can be used as a standalone embodiment, and the content of Figure 11 other than S1108 can be used as an optional implementation in this embodiment.

[0341] S1109, M third network elements send first response messages to the first network element. Correspondingly, the first network element receives first response messages from at least one third network element. The first response message indicates whether or not to participate in the distributed learning model training. The content of the first response message can be referred to in Figure 10 above; repetitions will not be listed again.

[0342] Following step S1109, the first network element determines the final participants in the distributed learning model training based on the first response messages from the M third network elements. For example, if a third network element indicates in its first response message that it will not participate in the distributed learning model training, the first network element will not select that third network element to participate. If a third network element indicates in its first response message that it will participate in the distributed learning model training, the first network element can select that third network element to participate. This process continues to determine the participants in the distributed learning model training.

[0343] In this embodiment, participants are not required to register their supported features in the second network element, reducing the risk of feature leakage. The first network element can discover at least one third network element through the second network element and request at least one third network element to report its supported features. The first network element determines M third network elements to participate in vertical federated learning based on the features reported by these at least one third network element, improving the flexibility in determining which third network elements participate in the training of the distributed learning model. Furthermore, the first network element can also determine the features of the M third network elements used to participate in the distributed learning model, thereby completing the feature alignment process.

[0344] It is understood that, in order to achieve the functions in the above embodiments, each network element includes corresponding hardware structures and / or software modules for performing each function. Those skilled in the art should readily recognize that, based on the units and method steps described in conjunction with the embodiments disclosed in this application, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is executed in hardware or by computer software driving hardware depends on the specific application scenario and design constraints of the technical solution.

[0345] Please refer to Figure 12, which is a schematic diagram of a communication device provided in an embodiment of this application. This communication device can be used to implement the functions of the first network element, the second network element, or any third network element in the above method embodiments, and therefore can also achieve the beneficial effects of the above method embodiments. In the embodiments of this application, the communication device can be the main participant or slave participant involved in Figure 2, the initiator or first network element involved in Figure 4, the initiator or AF involved in Figure 6, the registrant or second network element involved in Figure 4, any network element involved in Figure 5 (such as VFLSF), the registrant or VFLSF involved in Figure 6, the main participant or slave participant involved in Figure 2, the registrant or second network element involved in Figure 4, any network element involved in Figure 5, or other participants or any third network element involved in Figure 6, etc.

[0346] As shown in Figure 12, the communication device 1200 includes a processing module 1210 and a transceiver module 1220. The communication device 1200 is used to implement the function of the first network element in any of the method embodiments shown in Figures 7 to 9, the function of the second network element in any of the method embodiments shown in Figures 7 to 9, the function of the first network element in the method embodiment shown in Figure 10 or Figure 11, or the function of any of the third network elements in the method embodiment shown in Figure 10 or Figure 11.

[0347] In the first embodiment, the communication device 1200 is used to implement the function of the first network element in any of the method embodiments shown in Figures 7 to 9.

[0348] For example, the transceiver module 1220 is used to send a first request and receive first information under the control of the processing module 1210. Optionally, the processing module 1210 is also used to determine M third network elements.

[0349] In the second embodiment, the communication device 1200 is used to implement the function of the second network element in any of the method embodiments shown in Figures 7 to 9.

[0350] For example, the transceiver module 1220 is used to receive a first request and send first information under the control of the processing module 1210.

[0351] In the third embodiment, the communication device 1200 is used to implement the function of the first network element in the method embodiment shown in FIG10 or FIG11 above.

[0352] For example, the transceiver module 1220 is used to send first information and receive second information; the processing module 1210 is used to determine at least one third network element and determine the features of M third network elements for participating in the training of the distributed learning model.

[0353] In the fourth embodiment, the communication device 1200 is used to implement the function of the third network element in the method embodiment shown in FIG10 or FIG11 above.

[0354] For example, the transceiver module 1220 is used to receive first information and send second information under the control of the processing module 1210.

[0355] A more detailed description of the above-mentioned processing module 1210 and transceiver module 1220 can be obtained directly from the relevant descriptions in any of the method embodiments shown in Figures 7 to 11, and will not be repeated here.

[0356] Please refer to Figure 13, which is a schematic diagram of the structure of a communication device provided in an embodiment of this application. This communication device can be used to implement the functions of the first network element, the second network element, or any third network element in the above method embodiments, and therefore can also achieve the beneficial effects of the above method embodiments. In the embodiments of this application, the communication device can be the main participant or slave participant involved in Figure 2, the initiator or first network element involved in Figure 4, the initiator or AF involved in Figure 6, the registrant or second network element involved in Figure 4, any network element involved in Figure 5 (such as VFLSF), the registrant or VFLSF involved in Figure 6, the main participant or slave participant involved in Figure 2, the registrant or second network element involved in Figure 4, any network element involved in Figure 5, or other participants or any third network element involved in Figure 6, etc. As shown in Figure 13, the communication device 1300 includes a processing circuit 1310 and an interface circuit 1320. The processing circuit 1310 and the interface circuit 1320 are coupled to each other. The implementation of the processing circuit 1310 can be referred to the above discussion, and repeated details will not be listed again. The interface circuit 1320 can be a transceiver or an input / output interface. Optionally, the communication device 1300 may also include a memory 1330 for storing instructions executed by the processing circuit 1310, or storing input data required for the processing circuit 1310 to run instructions, or storing data generated after the processing circuit 1310 runs instructions.

[0357] [Correction 04.03.2025 based on Rule 91] Communication device 1300 can be used to implement any of the method embodiments shown in Figures 7 to 11. Optionally, communication device 1300 can be used to implement the functions of communication device 1200 shown in Figure 12 above. For example, processing circuit 1310 is used to implement the functions of processing module 1210, and interface circuit 1320 is used to implement the functions of transceiver module 1220.

[0358] When the aforementioned communication device is a chip applied to a network element (such as a first network element, a second network element, or a third network element), the network element chip implements the function of a network element in the above method embodiments. The network element chip receives information from other modules (such as a radio frequency module or antenna) within a network element, the information being sent from one network element to another; or, the network element chip sends information to other modules (such as a radio frequency module or antenna) within a network element, the information being sent from one network element to another.

[0359] The memory involved in the various embodiments of this application may include volatile memory, such as random access memory (RAM). The memory may also include non-volatile memory, such as read-only memory (ROM), flash memory, hard disk drive (HDD), or solid state drive (SSD).

[0360] This application provides another example of a communication device, which includes at least one processor and at least one memory coupled together. The at least one processor and the at least one memory are used to store instructions. When the instructions are executed by the at least one processor, the communication device performs the methods described in the above embodiments. Taking a communication device including a processor and a memory as an example, as shown in FIG14, the communication device 1400 includes a processor 1410 and a memory 1420. The processor 1410 and the memory 1420 are coupled together. The memory 1420 stores instructions. When the instructions stored in the memory 1420 are executed by the processor 1410, the communication device 1400 performs the methods described in any of the above embodiments.

[0361] The processors involved in the various embodiments of this application can be central processing units (CPUs), or other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. A general-purpose processor can be a microprocessor or any conventional processor.

[0362] The method steps in the various embodiments of this application can be implemented in hardware or in software instructions executable by a processor. The software instructions can consist of corresponding software modules, which can be stored in random access memory, flash memory, read-only memory, programmable read-only memory, erasable programmable read-only memory, electrically erasable programmable read-only memory, registers, hard disks, portable hard disks, CD-ROMs, or any other form of storage medium known in the art. An exemplary storage medium is coupled to a processor, enabling the processor to read information from and write information to the storage medium. The storage medium can also be a component of the processor. The processor and storage medium can reside in an ASIC. Alternatively, the ASIC can reside in a base station or terminal. The processor and storage medium can also exist as discrete components in a base station or terminal.

[0363] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer programs or instructions. When the computer program or instructions are loaded and executed on a computer, the processes or functions described in the embodiments of this application are performed entirely or partially. The computer can be a general-purpose computer, a special-purpose computer, a computer network, a network device, a user equipment, or other programmable device. The computer program or instructions can be stored in a computer-readable storage medium or transferred from one computer-readable storage medium to another. For example, the computer program or instructions can be transferred from one website, computer, server, or data center to another website, computer, server, or data center via wired or wireless means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium, such as a floppy disk, hard disk, or magnetic tape; it can also be an optical medium, such as a digital video optical disc; or it can be a semiconductor medium, such as a solid-state drive. The computer-readable storage medium may be a volatile or non-volatile storage medium, or may include both types of storage media.

[0364] This application provides a communication system comprising a first network element and a second network element. The first network element can perform the functions of the first network element in any of the method embodiments shown in Figures 7 to 9, and the second network element can perform the functions of the second network element in any of the method embodiments shown in Figures 7 to 9. Optionally, the communication system further includes at least one third network element, which can perform the functions of at least one third network element in any of the method embodiments shown in Figures 7 to 9.

[0365] This application provides a communication system comprising a first network element and at least one third network element. The first network element can perform the functions of the first network element in the method embodiment of Figure 10 or Figure 11 above, and the at least one third network element can perform the functions of at least one third network element in the method embodiment of Figure 10 or Figure 11 above. Optionally, the communication system further comprises a second network element, which can perform the functions of the second network element in the method embodiment of Figure 10 or Figure 11 above.

[0366] This application provides a chip system comprising a processor and an interface. The processor is configured to call and execute instructions from the interface, and when the processor executes the instructions, it implements the method described in any one of Figures 7 to 11.

[0367] This application provides a computer-readable storage medium for storing computer programs or instructions that, when run, implement the method described in any one of Figures 7 to 11 above.

[0368] This application provides a computer program product containing instructions that, when run on a computer, implements the method described in any one of Figures 7 to 11 above.

[0369] In the various embodiments of this application, unless otherwise specified or in case of logical conflict, the terminology and / or descriptions of different embodiments are consistent and can be referenced by each other. The technical features of different embodiments can be combined to form new embodiments according to their inherent logical relationship.

[0370] It is understood that the various numerical designations used in the embodiments of this application are merely for descriptive convenience and are not intended to limit the scope of the embodiments of this application. The order of the process numbers described above does not imply the order of execution; the execution order of each process should be determined by its function and internal logic.

Claims

1. A communication method, characterized in that, Applications to vertical federated learning servers include: Send a first request to the network storage function element. The first request is used to request the discovery of a vertical federated learning client. The first request indicates one or more feature identifiers, and the one or more feature identifiers indicate the features of the vertical federated learning client to be discovered for training the vertical federated learning model. Receive first information from the network storage function element, the first information including information from at least one vertical federated learning client, wherein the union of feature identifiers supported by the at least one vertical federated learning client includes the one or more feature identifiers.

2. The method according to claim 1, characterized in that, The first request further indicates one or more network element types, and a feature identifier corresponding to each of the one or more network element types, wherein the feature identifier corresponding to each network element type belongs to the one or more feature identifiers; Wherein, the network element type of the at least one vertical federated learning client belongs to one or more network element types.

3. The method according to claim 1 or 2, characterized in that, The first information also indicates the feature identifier corresponding to each of the at least one vertical federated learning clients.

4. The method according to claim 3, characterized in that, The feature identifier corresponding to each of the at least one vertical federated learning clients is the feature identifier among the one or more feature identifiers.

5. The method according to claim 3 or 4, characterized in that, The method further includes: Based on the at least one vertical federated learning client, M vertical federated learning clients are determined, and the M vertical federated learning clients are used to participate in the training of the vertical federated learning model.

6. The method according to claim 5, characterized in that, The method further includes: Determine the feature identifier for each of the M longitudinal federated learning clients used to participate in the training of the longitudinal federated learning model; The second information is sent to each of the M vertical federated learning clients, wherein the second information sent to one of the M vertical federated learning clients is used to indicate the feature identifier used by the vertical federated learning client to participate in the training of the vertical federated learning model.

7. The method according to any one of claims 1-6, characterized in that, The first information includes: Information on at least one vertical federated learning client group, wherein the vertical federated learning client group includes some or all of the vertical federated learning clients in the at least one vertical federated learning client group, and the union of feature identifiers supported by the vertical federated learning clients in the vertical federated learning client group includes the one or more feature identifiers.

8. The method according to claim 7, characterized in that, The first request further indicates one or more network element types, and a feature identifier corresponding to each of the one or more network element types, wherein the feature identifier corresponding to each network element type belongs to the one or more feature identifiers; Wherein, the network element type of any vertical federated learning client in the vertical federated learning client group is a first network element type, the first network element type is one of the one or more network element types, and the union of the feature identifiers supported by the vertical federated learning clients in the vertical federated learning client group includes the feature identifier corresponding to the first network element type indicated by the first request.

9. The method according to claim 7, characterized in that, The first request further indicates one or more network element types, and a feature identifier corresponding to each of the one or more network element types, wherein the feature identifier corresponding to each network element type belongs to the one or more feature identifiers; Wherein, the union of network element types of the vertical federated learning clients in the vertical federated learning client group includes the one or more network element types, the network element type of any vertical federated learning client in the vertical federated learning client group is one of the one or more network element types, and the union of feature identifiers supported by all vertical federated learning clients in the vertical federated learning client group whose network element types are all the first network element type includes the feature identifier corresponding to the first network element type indicated by the first request, where the first network element type is one of the one or more network element types.

10. The method according to any one of claims 1-9, characterized in that, The method further includes: A third message is sent to each of the at least one vertical federated learning clients, wherein the third message sent to one of the at least one vertical federated learning clients instructs the vertical federated learning client to provide the feature identifier supported by the vertical federated learning client; Each receives a fourth message from the at least one vertical federated learning client, wherein the fourth message received from one of the at least one vertical federated learning clients indicates the feature identifier supported by the one vertical federated learning client.

11. The method according to any one of claims 1-10, characterized in that, The method further includes: The network storage function element sends a fifth message, which indicates the feature identifiers supported by the vertical federated learning server.

12. A communication method, characterized in that, The method, applied to network storage function network elements, includes: Receive a first request from a longitudinal federated learning server, the first request being for requesting the discovery of a longitudinal federated learning client, the first request indicating one or more feature identifiers, the one or more feature identifiers indicating features of the longitudinal federated learning client to be discovered for training the longitudinal federated learning model; Send first information to the vertical federated learning server, the first information including information of at least one vertical federated learning client, wherein the union of feature identifiers supported by the at least one vertical federated learning client includes the one or more feature identifiers.

13. The method according to claim 12, characterized in that, The first request further indicates one or more network element types, and a feature identifier corresponding to each of the one or more network element types, wherein the feature identifier corresponding to each network element type belongs to the one or more feature identifiers; Wherein, the network element type of the at least one vertical federated learning client belongs to one or more network element types.

14. The method according to claim 12 or 13, characterized in that, The first information also indicates the feature identifier corresponding to each of the at least one vertical federated learning clients.

15. The method according to claim 14, characterized in that, The feature identifier corresponding to each of the at least one vertical federated learning clients is the feature identifier among the one or more feature identifiers.

16. The method according to any one of claims 12-15, characterized in that, The first information includes: Information on at least one longitudinal federated learning client group, wherein the information on one longitudinal federated learning client group includes some or all of the longitudinal federated learning clients in the one longitudinal federated learning client group, and the union of feature identifiers supported by the longitudinal federated learning clients in the one longitudinal federated learning client group includes the one or more feature identifiers.

17. The method according to claim 16, characterized in that, The first request further indicates one or more network element types, and a feature identifier corresponding to each of the one or more network element types, wherein the feature identifier corresponding to each network element type belongs to the one or more feature identifiers; Wherein, the network element type of any vertical federated learning client in the vertical federated learning client group is a first network element type, the first network element type is one of the one or more network element types, and the union of the feature identifiers supported by the vertical federated learning clients in the vertical federated learning client group includes the feature identifier corresponding to the first network element type indicated by the first request.

18. The method according to claim 16, characterized in that, The first request further indicates one or more network element types, and a feature identifier corresponding to each of the one or more network element types, wherein the feature identifier corresponding to each network element type belongs to the one or more feature identifiers; The union of network element types of the vertical federated learning clients in the vertical federated learning client group includes the one or more network element types. The network element type of any vertical federated learning client in the vertical federated learning client group is one of the one or more network element types. The union of feature identifiers supported by all vertical federated learning clients in the vertical federated learning client group whose network element types are all the first network element type includes the feature identifier corresponding to the first network element type indicated by the first request. The first network element type is one of the one or more network element types.

19. The method according to any one of claims 12-18, characterized in that, The method further includes: Receive fifth information from the vertical federated learning server, the fifth information indicating the feature identifiers supported by the vertical federated learning server; and / or, Receive a sixth message from the at least one vertical federated learning client, and receive a sixth message from one of the at least one vertical federated learning clients to indicate the feature identifiers supported by the one vertical federated learning client.

20. A communication method, characterized in that, Applied to the first network element, the method includes: Identify at least one third network element; Send first information to each of the at least one third network element, wherein the first information sent to one of the at least one third network elements is used to request features supported by the one third network element, wherein some or all of the features supported by the at least one third network element are features corresponding to the data trained by the distributed learning model; Each receives second information from the at least one third network element, wherein receiving second information from one of the at least one third network elements indicates the features supported by the third network element. Based on the second information received from the at least one third network element, M features of the third network elements used to participate in the training of the distributed learning model are determined, wherein the M third network elements are some or all of the at least one third network element, and M is a positive integer.

21. The method according to claim 20, characterized in that, The method further includes: Third information is sent to each of the M third network elements, wherein the third information sent to one of the M third network elements indicates a feature used to participate in the training of the distributed learning model.

22. The method according to claim 20 or 21, characterized in that, Identify at least one third network element, including: Receive fourth information from the second network element, the fourth information indicating information from the at least one third network element; Based on the fourth information, the at least one third network element is determined.

23. A communication method, characterized in that, Applied to a third network element, the method includes: Receive first information from a first network element, the first information being used to request features supported by the third network element; Send a second message to the first network element, the second message indicating the features supported by the third network element.

24. The method according to claim 23, characterized in that, Receive third information from the first network element, wherein the third information indicates the features of the third network element used to participate in the training of the distributed learning model.

25. The method according to claim 23 or 24, characterized in that, The method further includes: Send a fifth message to the second network element, the fifth message indicating the information of the third network element.

26. A communication device, characterized in that, include: A module for performing the method as described in any one of claims 1-11; or, A module for performing the method as described in any one of claims 12-19; ​​or, A module for performing the method as described in any one of claims 20-22; or, A module for performing the method as described in any one of claims 23-25.

27. A communication device, characterized in that, include: A processor and a memory, wherein the memory is used to store program instructions, and the processor is used to execute the program instructions in the memory to implement the method as claimed in any one of claims 1-11, the method as claimed in any one of claims 12-19, the method as claimed in any one of claims 20-22, or the method as claimed in any one of claims 23-25.

28. A communication device, characterized in that, include: Processing circuits and interface circuits; among which: The interface circuit is used to couple with a memory external to the communication device and to provide a communication interface for the processing circuit to access the memory; the processing circuit is used to execute program instructions in the memory to implement the method as described in any one of claims 1-11, the method as described in any one of claims 12-19, the method as described in any one of claims 20-22, or the method as described in any one of claims 23-25.

29. A computer program product containing instructions, characterized in that, When the instruction is executed by the communication device, the communication device performs the method as described in any one of claims 1-11, implements the method as described in any one of claims 12-19, implements the method as described in any one of claims 20-22, or implements the method as described in any one of claims 23-25.

30. A computer-readable storage medium, characterized in that, The storage medium stores a computer program or instructions that, when executed by a communication device, implement the method as described in any one of claims 1-11, the method as described in any one of claims 12-19, the method as described in any one of claims 20-22, or the method as described in any one of claims 23-25.

Images