Communication methods, devices, equipment, chips, and storage media

Abstract

Embodiments of the present application provide a communication method comprising receiving first information from a first network element and determining from at least one second node to participate in a VFL task, wherein the first information is used to indicate at least one second node that satisfies the conditions for performing the VFL task. The method ensures that a first node initiating a VFL task can receive first information from a first network element, and further, the first node can find a suitable node to perform the VFL task, by knowing at least one second node that satisfies the conditions for performing the VFL task.

Description

Technical Field

[0001] The embodiments of this application relate to the field of communication technologies, specifically to communication methods, devices, equipment, chips, and storage media.

Background Art

[0002] Vertical Federated Learning (VFL) enables an artificial intelligence system to efficiently and accurately use the local data of multiple nodes on the premise of meeting data privacy, security, and regulatory requirements, ensuring privacy security, eliminating data silos, and realizing multi-node data sharing between domains.

[0003] However, so far, there is no proposal that clearly shows how existing network architectures such as 5G network architectures support vertical federated learning between domains.

Summary of the Invention

[0004] The embodiments of this application provide a communication method, device, equipment, chip, and storage media.

[0005] In a first aspect, the embodiments of this application provide a communication method applicable to a first node, the method including receiving first information from a first network element and determining at least one participating node participating in a VFL task from at least one second node, where the first information is used to indicate at least one second node that meets the conditions for executing a VFL task.

[0006] In a second aspect, the embodiments of this application provide a communication method applicable to a second node, the method including transmitting the capability information of the second node to a first network element, where the capability information of the second node is used to determine whether the second node meets the conditions for executing a vertical federated learning VFL task.

[0007] In a third embodiment, an embodiment of the present application provides a communication method applicable to a first network element, the method comprising obtaining capability information of at least one second node, the capability information of the second node being used to determine whether the second node satisfies the conditions for performing a vertically federated learning VFL task.

[0008] In a fourth embodiment, an embodiment of the present application provides a communication device comprising a first receiving module configured to receive first information from a first network element, and a decision module configured to determine at least one participating node from at least one second node to participate in a VFL task, wherein the first information is used to indicate at least one second node that satisfies the conditions for performing a vertically federated learning VFL task.

[0009] In a fifth embodiment, an embodiment of the present application provides a communication device comprising a first transmitting module configured to transmit capability information of the device to a first network element, the capability information of the device being used to determine whether the device satisfies the conditions for performing a vertically federated learning VFL task.

[0010] In a sixth embodiment, an embodiment of the present application provides a communication device comprising an acquisition module configured to acquire capability information of at least one second node, the capability information of the second node being used to determine whether the second node satisfies the conditions for performing a vertically federated learning VFL task.

[0011] In a seventh embodiment, an embodiment of the present application provides a communication device comprising a memory configured to store computer-executable instructions, and a processor connected to the memory and configured to implement the method according to any one of the first to third embodiments by executing the computer-executable instructions.

[0012] In the eighth embodiment, the embodiment of the present application provides a chip, which includes a processor configured to call and execute a computer program from memory, causing a device to which the chip is installed to perform the method described in any one of the first to third embodiments.

[0013] In the ninth embodiment, an embodiment of the present application provides a computer-readable storage medium storing a computer program which performs the method described in any one of the first to third embodiments when executed by at least one processor.

[0014] In the embodiment of this application, a first node can receive first information from a first network element, the first information is used to indicate at least one second node that satisfies the conditions for performing a VFL task, and further, the first node can determine from at least one second node at least one participating node to participate in the VFL task. In this way, it is possible to ensure that the first node can find a suitable node for performing the VFL task. [Brief explanation of the drawing]

[0015] [Figure 1] This is a schematic diagram of one application scenario of the embodiment of this application. [Figure 2] This is a schematic diagram illustrating an example of the VFL training process. [Figure 3] This is a schematic diagram illustrating an example of the VFL estimation process. [Figure 4] This is a schematic diagram of an example of a 5G network architecture. [Figure 5] This is a flowchart of the communication method according to an embodiment of this application. [Figure 6] This is a flowchart of another communication method according to the embodiment of this application. [Figure 7] This is flowchart 1 of possible implementations of the communication method according to the embodiment of this application. [Figure 8] This is flowchart 2 of possible implementations of the communication method according to the embodiment of this application. [Figure 9] It is the flowchart 3 of a possible implementation of the communication method according to an embodiment of the present application. [Figure 10] It is the structural schematic diagram 1 of the communication device according to an embodiment of the present application. [Figure 11] It is the structural schematic diagram 2 of the communication device according to an embodiment of the present application. [Figure 12] It is the structural schematic diagram 3 of the communication device according to an embodiment of the present application. [Figure 13] It is the schematic structural diagram of the communication equipment according to an embodiment of the present application. [Figure 14] It is the schematic structural diagram of the chip of an embodiment of the present application. [Figure 15] It is the schematic block diagram of the communication system provided by an embodiment of the present application.

Embodiments for Carrying out the Invention

[0016] The drawings described above are for further understanding of the present application, constitute a part of the present application, and the exemplary embodiments and their descriptions of the present application are for explaining the present application and do not constitute an undue limitation of the present application.

[0017] In the following, referring to the drawings in the embodiments of the present application, the technical solutions in the embodiments of the present application will be described. It is clear that the described embodiments are only a part of the embodiments of the present application, not all of the embodiments. For other all embodiments that can be obtained by those skilled in the art without creative efforts in the application, they all belong to the protection scope of the present application.

[0018] FIG. 1 is a schematic diagram of one application scenario of an embodiment of the present application.

[0019] As shown in FIG. 1, the communication system 100 can include a terminal device 110 and a network device 120. The network device 120 can communicate with the terminal device 110 via a wireless port. Multi-service transmission is supported between the terminal device 110 and the network device 120.

[0020] It should be understood that while the embodiments of this application describe only communication system 100 as an example, the embodiments of this application are not limited to this. In other words, the technical solutions of the embodiments of this application can be applied to a variety of communication systems, such as Long Term Evolution (LTE®) systems, LTE Time Division Duplex (TDD), Universal Mobile Telecommunication System (UMTS), Ubiquitous Network (IoT) systems, Narrow Band Internet of Things (NB-IoT) systems, Enhanced Machine-Type Communications (eMTC) systems, 5G communication systems (also known as New Radio (NR) communication systems), or future communication systems (e.g., 6G communication systems).

[0021] In the communication system 100 shown in Figure 1, the network device 120 may be an access network device that communicates with terminal devices 110. The access network device provides communication coverage to a specific geographic area and can communicate with terminal devices (e.g., UEs) located within that coverage area.

[0022] The network device 120 may be an Evolutionary Node B (eNB or eNodeB) in a Long Term Evolution (LTE) system, or a Next Generation Radio Access Network (NG RAN) device, or a base station (gNB) in an NR system, or a radio controller in a Cloud Radio Access Network (CRAN), or the network device 120 may be a relay station, access point, in-vehicle equipment, wearable device, hub, switch, bridge, router, or network equipment in a future evolving Public Land Mobile Network (PLMN).

[0023] The terminal device 110 may be any terminal device, and may include, but is not limited to, a terminal device connected to the network device 120 or other terminal devices by wire or wireless connection.

[0024] For example, the terminal device 110 described above may also be called an access terminal, user equipment (UE), user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent, or user equipment. The access terminal may be a cellular phone, cordless phone, Session Initiation Protocol (SIP) phone, IoT device, satellite handheld terminal, Wireless Local Loop (WLL) station, Personal Digital Assistant (PDA), handheld device with wireless communication capabilities, computing device or other processing device connected to a wireless modem, in-vehicle device, wearable device, terminal device in a 5G network, or terminal device in a future evolving network.

[0025] Terminal device 110 can be used for device-to-device (D2D) communication.

[0026] The wireless communication system 100 includes a core network device 130 that communicates with network equipment 120, and the core network device 130 may be a 5G core network (5G Core, 5GC) device.

[0027] Figure 1 illustrates one network device, one core network device, and two terminal devices, and optionally the wireless communication system 100 may include multiple network devices and may include other numbers of terminal devices within the coverage area of ​​each network device, but is not limited to the embodiments of this application.

[0028] Figure 1 is merely an illustrative example of a system to which this application applies, and of course, the methods shown in the embodiments of this application can be applied to other systems. Furthermore, the terms “system” and “network” as used herein are always interchangeable. The term “and / or” as used herein simply describes a relationship of association between related objects, indicating that three relationships are possible. For example, A and / or B can indicate three situations: A exists independently, A and B exist simultaneously, or B exists independently. Additionally, the symbol “ / ” as used herein generally indicates an “or” relationship between preceding and succeeding related objects. Furthermore, it should be understood that “indication” as referred to in the embodiments of this application may be direct or indirect, and may also indicate an association. For example, A indicating B may mean that A directly indicates B, e.g., that B can be obtained by A; or A indirectly indicates B, e.g., that A indicates C and B can be obtained by C; and may also indicate an association between A and B. Furthermore, it should be understood that in the descriptions of the embodiments of this application, the term “correspond” may indicate a direct or indirect correspondence between two things, or a related relationship between two things, and may be a relationship such as indicating and being indicated, configuring and being configured. Furthermore, it should be understood that in the embodiments of this application, “predefined” or “predefined rule” may be implemented by pre-storing corresponding codes, tables within equipment (e.g., including terminal equipment and network equipment), or by other means for indicating related information, and this application does not limit the specific implementation methods. For example, predefined may refer to those defined in a protocol. Furthermore, it should be understood that in the embodiments of this application, “protocol” may refer to, and is not limited to, standard protocols in the field of communications, which may include, for example, LTE protocols, NR protocols, and related protocols applicable to future communication systems.

[0029] To generate Artifact Intelligence (AI) models that better meet user needs, more original user data is required for model training. In real-world scenarios, user data may be distributed across various nodes, such as terminals, base stations, core networks, and third-party OTT (Over the Top) application servers. If different feature data from the same user at each node can be linked and used for model training, the training effectiveness of the model can be significantly improved, which is of great importance. However, sharing multi-node, multi-domain data presents significant challenges to data privacy. Therefore, the VFL method can be adopted. VFL enables AI systems to efficiently and accurately use multi-node local data while ensuring privacy security, eliminating data silos, and enabling multi-node data sharing across domains, provided that data privacy, security, and regulatory requirements are met. However, to date, there have been no proposals that clearly demonstrate how existing network architectures can support VFL across domains.

[0030] Therefore, this application provides a communication method in which a first node can receive first information from a first network element and the first information can be used to know at least one second node that satisfies the conditions for performing the VFL task, so as to ensure that the first node can find a suitable node for performing the VFL task.

[0031] To facilitate understanding of the technical solutions of the embodiments of this application, the related technologies of the embodiments of this application are described below. The following related technologies can be combined with the technical aspects of the embodiments of this application as selectable technical solutions and all fall within the scope of protection of the embodiments of this application.

[0032] 1. Vertical Associative Learning (VFL) Figure 2 is a schematic diagram of an example of the VFL training process. As shown in Figure 2, Node A and Node B are nodes in different domains, and there is no exchange of original data between Node A and Node B. To train a model using multi-node data, the VFL training process may include the following steps.

[0033] In S201, the encryption samples are aligned.

[0034] VFL is suitable when there is a lot of overlap in the training sample IDs of participants, but little overlap in the data features. In VFL, it is necessary to increase the feature dimension of each sample without increasing the number of sample IDs by aligning the participant samples. For example, a UE in a certain area generates different feature data at different nodes of a communication system, and here the ID of the UE is the sample ID, and in this scenario, it is necessary to align (associate) the feature data generated by the UE at different nodes.

[0035] In S202, the encryption model is trained.

[0036] After aligning the samples, a cryptographic model can be trained on the aligned samples, and then a better performing federated model can be trained based on Model A obtained from Node A and Model B obtained from Node B. For example, step 202 may include the following steps.

[0037] In S1, the public key is distributed.

[0038] As shown in Figure 2, a third-party coordinator C can send a public key to nodes A and B in order to encrypt the data that needs to be transmitted. The encryption method can be, for example, a homomorphic encryption algorithm. In a homomorphic encryption algorithm, the sum of two samples m1 and m2 is homomorphically encrypted, i.e., it is equal to the sum of the homomorphic encrypted result of m1 and the homomorphic encrypted result of m2, and the homomorphic encryption of sample m multiplied by a constant is equal to the homomorphic encryption of sample m multiplied by the constant.

[0039] In S2, the intermediate results are exchanged.

[0040] In VFL, as shown in Node B in Figure 2, the side with the sample labels can be the active side (also called the demand side). Node A can be the passive side (also called the data provider), and the passive side does not have sample labels. In this step, Node A and Node B can each use their local data to perform calculations and obtain intermediate results for the model. Here, Node A can encrypt the obtained intermediate results and send them to Node B, and furthermore, Node B can calculate and obtain the output error of the entire model using its own labels and the model output results (intermediate results) of Node A and Node B, and can encrypt this output error and send it to Node A.

[0041] In S3, calculate the gradient.

[0042] In this step, Node A and Node B each calculate their respective encrypted gradients using the output error in S2, and can then mask the calculation results before sending them to Coordinator C.

[0043] Update the model in S4.

[0044] Coordinator C can decode the gradients sent by Node A and Node B, and then return the decoded gradients to Node A and Node B, respectively. Therefore, Node A and Node B can remove the gradient mask and update their respective models with the obtained gradients.

[0045] Figure 3 is a schematic diagram of an example of the VFL estimation process. As shown in Figure 3, the VFL estimation process may include the following steps.

[0046] In S301, a model estimation request is sent.

[0047] Coordinator C can send model estimation requests to Node A and Node B respectively, which include the IDs of the models that Node A and Node B need to use, thereby indicating the models that Node A and Node B need to use.

[0048] In S302, Node A and Node B calculate and encrypt the model results.

[0049] In this step, Node A and Node B can each perform calculations using their own data and locally stored models, obtain intermediate results from the models, and encrypt those intermediate results.

[0050] In S303, Node A and Node B send the encrypted intermediate results to Coordinator C.

[0051] In S304, Coordinator C aggregates the encrypted intermediate results from each node and decrypts them.

[0052] In this step, Coordinator C aggregates the encrypted intermediate results from Node A and Node B to obtain encrypted model estimation results. Furthermore, Coordinator C can decrypt these model estimation results and send the decrypted estimation results to Node B, which is the demand side.

[0053] 2. Fifth Generation (5G) Network Architecture One of the main features of 5G network architecture is the "service architecture." Here, a core network element (service provider) can provide specific services and offer defined application programming interfaces (APIs) that can be called by other network elements (consumers).

[0054] It should be understood that the "core network element" in the embodiments of this application is also called a "core network function (NF)".

[0055] Figure 4 is a schematic diagram of an example of a 5G network architecture. As shown in Figure 4, the network architecture may include a UE, an Access Network (AN) / Radio Access Network (RAN), and core network elements. Here, the core network elements include a User Plane Function (UPF), a Data Network (DN), a Session Management Function (SMF), an Access and Mobility Management Function (AMF), a Network Slice Selection Function (NSSF), an Authentication Server Function (AUSF), a Network Exposure Function (NEF), a Network Repository Function (NRF), a Policy Control Function (PCF), a Unified Data Management (UDM), and an Application Function (AF).

[0056] Here, the UE connects to the base station at the Access Stratum (AS) for interaction with access layer messages and transmission of wireless data. The UE connects to the AMF at the Non-Access Stratum (NAS) for interaction with NAS messages. The AMF is responsible for managing the UE's mobility, and the SMF is responsible for managing the UE's sessions. In addition to managing the mobility of mobile devices, the AMF is also responsible for transferring session management-related messages between the UE and the SMF. The PCF is responsible for formulating policies related to the UE's mobility management, session management, and billing. The UPF connects to the base station and the external data network and transmits data.

[0057] Furthermore, the 5G network adds a Network Data Analytics Function (NWDAF) to the core network. This function collects data from each network element and network management within the core network, performs big data statistics, analysis, or intelligent data analysis to obtain network-side analysis and / or predictive data, and further assists each network element in more effectively controlling UE access based on the data analysis results.

[0058] Below is a brief introduction to each network element shown in Figure 4.

[0059] 1) UE may also be called terminal equipment, access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication equipment, user agent, or user device. The embodiments of this application do not limit the specific technology and specific equipment form used by the UE.

[0060] 2) A (Radio) Access Network ((R)AN) device provides authorized users in a specific area with the ability to access a communication network and may specifically include radio network devices in a 3rd Generation Partnership Project (3GPP®) network, or access points in a non-3GPP network. The embodiments of this application do not limit the specific technologies and specific forms of devices used by the (R)AN device.

[0061] 3) AMF is primarily used for functions such as access control, mobility management, and attach / detach. AMF also acts as an anchor for the connection between N1 signaling (i.e., signaling for the N1 interface, abbreviated as N1 signaling for brevity) and N2 signaling (i.e., signaling for the N2 interface, abbreviated as N2 signaling for brevity), providing SMF with routing of N1 / N2 session management (SM) messages. AMF can also maintain and manage UE status information.

[0062] 4) SMF is primarily used for selecting user plane network elements, redirecting user plane network elements, assigning Internet Protocol (IP) addresses to terminal devices, establishing, modifying, and releasing sessions, and QoS control.

[0063] 5) UPF is primarily used for receiving and forwarding user plane data. For example, UPF can receive user plane data from DN and transmit it to terminal devices via AN equipment. UPF can also receive user plane data from terminal devices via AN equipment and forward it to DN.

[0064] 6) PCF is a unified policy framework primarily for guiding network operation, providing policy rule information to control plane network elements (e.g., AMF, SMF, etc.).

[0065] 7) AF is primarily used to provide services to 3GPP networks, for example, by interacting with PCF to perform policy control.

[0066] 8) The UDM is primarily used to manage UE subscription data, which includes storing and managing UE identifiers and UE access permissions. The UDM can also generate 3GPP authentication credentials for the UE. The UDM can also register and maintain network elements that are currently serving the UE.

[0067] 9) NEF is primarily used to securely open up services and capabilities provided by 3GPP network functions to the outside world.

[0068] 10) DNs are carrier networks primarily used to provide data services to UEs. Examples include the Internet, third-party service networks, and IP Multimedia Service (IMS) networks.

[0069] 11) AUSF is primarily used for user authentication, for example, to securely authenticate a UE when it accesses the network.

[0070] 12) NSSF is primarily used to select a set of slice instances for the UE, and to determine the AMF set, permitted Network Slice Selection Assistance Information (NSSAIs) for the UE.

[0071] In the network architecture shown in Figure 4, communication between network elements can be achieved via the interfaces shown in the figure, and some interfaces can be implemented using a service-oriented interface scheme. As shown in Figure 4, communication between the UE and AMF can be achieved via the N1 interface. Communication between the RAN and AMF can be achieved via the N2 interface. Communication between the RAN and UPF can be achieved via the N3 interface, which can be used to transmit user plane data, etc. Communication between the SMF and UPF can be achieved via the N4 interface. Communication between the UPF and DN can be achieved via the N6 interface. Communication between UPFs can be achieved via the N9 interface, which can be used to transmit upstream and downstream user data streams, etc., between UPFs. The relationships between other interfaces and each network element are shown in Figure 4 and are omitted here for brevity.

[0072] It should be understood that the above naming conventions are defined simply to facilitate the distinction between different functions and are not limited to the configuration of this application. This application does not preclude the possibility of naming 6G networks and other future networks differently. For example, in a 6G network, some or all of the above network elements may borrow terminology from 5G, or adopt other names, etc.

[0073] Furthermore, it should be understood that the interface names between each network element in Figure 4 are merely examples, and the actual interface names in implementation may be different, and this application is not particularly limited to these. Also, the names of the messages (or signaling) transmitted between the above network elements are merely examples and are not limited to the original function of the messages.

[0074] The terms used in this application have been briefly explained above, but will not be repeated in the following embodiments.

[0075] To facilitate understanding of the solutions in the disclosed embodiments, the solutions of this application will be described below by specific embodiments. The above-mentioned related technologies can be combined with the technical aspects of the embodiments of this application as selectable technical solutions, and all fall within the scope of protection of the embodiments of this application. The embodiments of this application include at least some of the following:

[0076] It should be understood that in the embodiments of this application, the first node represents the node that initiates the VFL task; in other words, the first node is the initiator (also called the active / demand side) of the VFL task, meaning that the VFL task in the embodiments of this application may be initiated by the first node. Correspondingly, the second node represents a node that can participate in the VFL task as a non-initiator, or the second node may be the passive side (also called the data provider) of the VFL task. Here, the number of second nodes may be one or more (two or more), and the embodiments of this application are not limited thereto.

[0077] Furthermore, it should be understood that the first node in the embodiments of this application may be, for example, a terminal device, an access network network element, a core network element (core network NF), or an AF, and the second node may include, for example, a terminal device, an access network network element, a core network element, and at least one of the AFs, where the first network element may be, for example, an NRF. Here, the AF may be a trusted AF (e.g., a carrier's own AF) or an untrusted third-party AF (e.g., an AF from another manufacturer). Here, whether an AF is a trusted AF can be verified or identified by the core network. For example, if the core network belongs to a core network deployed by carrier A, the core network can verify or identify the AFs belonging to carrier A as trusted AFs, and verify or identify the AFs not belonging to carrier A as untrusted third-party AFs.

[0078] Figure 5 provides a communication method according to an embodiment of the present application, which may include the following steps.

[0079] In S501, the first node receives first information from the first network element, and the first information is used to indicate at least one second node that satisfies the conditions for performing the VFL task.

[0080] In one possible form, the VFL task is initiated by the first node, so "satisfying the conditions for executing a VFL task" can also be understood as satisfying the conditions for executing a VFL task initiated by the first node.

[0081] In one embodiment, a first node can initiate a VFL task by transmitting second information to a first network element, the second information being one of the following:

[0082] 1) This involves executing a VFL task; in other words, the type of task initiated by the first node is a VFL task.

[0083] 2) This is the period during which the VFL task is executed, or in other words, the period during which the first node attempts to execute the VFL task, for example, from 9:00 to 21:00.

[0084] 3) To act as the initiator of the VFL task; in other words, the first node attempts to be the initiator of the VFL task.

[0085] 4) Computing power to run VFL tasks, for example, the minimum computing power required to run a VFL task initiated by the first node.

[0086] 5) This is the service area where the VFL task is executed, or in other words, the service area where the first node attempts to execute the VFL task.

[0087] In one possible configuration, after the first network element receives second information from the first node, it can send first information to the first node, thereby indicating to the first node at least one second node that satisfies the conditions for performing the VFL task, and in turn, the first node can also receive first information from the first network element. In this way, the first node can know, by the first information, at least one second node that satisfies the conditions for performing the VFL task, thereby ensuring that the first node finds a suitable node and performs the VFL task.

[0088] As an example, the first information may include, for instance, identifiers and / or address information of at least one second node that satisfies the conditions for performing a VFL task.

[0089] In some embodiments, the first network element determines whether the second node satisfies the conditions for performing a VFL task based on the capabilities information of the second node; in other words, the first network element can determine whether the second node satisfies the conditions for performing a VFL task based on the capabilities information of the second node.

[0090] For example, the capability information of the second node can be used to demonstrate at least one of the following:

[0091] 1) Capability information indicating whether the second node supports the execution of VFL tasks. For example, the capability information for the second node may include an indication that the second node supports the execution of VFL tasks, indicating whether the second node supports VFL tasks or not.

[0092] 2) Capability information indicating the period during which the second node supports the execution of VFL tasks. For example, the capability information of the second node can be used to indicate that the second node can execute VFL tasks between 9:00 and 21:00.

[0093] 3) Capability information for a second node, which indicates whether it can function as a VFL task initiator, for example, the capability information for a second node may include an indication of whether the second node can function as a VFL task initiator, showing that the second node can or cannot function as a VFL task initiator.

[0094] 4) Computing power information of the second node, for example, the maximum computing power or range of computing power that the second node can provide.

[0095] 5) This is capability information regarding the service area of ​​the second node. Here, if the second node is an access network element, the service area of ​​the second node can also be understood as the signal coverage area of ​​the access network element.

[0096] For example, assuming that the second information indicates that the period during which the first node intends to perform the VFL task is from 9:00 to 21:00 and the service area in which the VFL task is intended to be performed is the first service area, the first network element, after receiving the second information from the first node, compares the capability information of each second node that the first network element has collected or stored with the contents of the second information, determines that the second node supports performing the VFL task between 9:00 and 21:00 and that the service area includes the first service area, and transmits the first information to the first node with information about that at least one second node (e.g., identifier and / or address information).

[0097] In some embodiments, capability information of a second node can be included in the parameter information of the second node and transmitted by the second node to the first network element, the method further comprising the second node transmitting the parameter information of the second node to the first network element, the first network element receiving and storing the parameter information of the second node, in other words, the first network element can collect or determine the parameter information of the second node. Here, the parameter information of the second node may include, for example, at least one of the capability information of the second node, the identifier (ID) of the second node, the address information of the second node, and the identifier of the application corresponding to the second node.

[0098] If there are multiple second nodes, each second node can transmit its parameter information to the first network element, and it can be understood that each second node's parameter information includes at least one of the above parameter information.

[0099] In some embodiments, each second node can transmit its parameter information to the first network element before the first node starts the VFL task, thereby allowing the first network element to collect the parameter information of each second node before the first node starts the VFL task. Furthermore, when the first node starts the VFL task, the first network element can obtain capability information for each second node from the collected parameter information of each second node. In addition, based on the capability information of each second node, the first network element can determine at least one second node that satisfies the conditions for performing the VFL task, and transmit the information of that at least one second node (e.g., identifier and / or address information) to the first node in the first information.

[0100] In S502, the first node determines, from at least one second node, at least one participating node to participate in the VFL task.

[0101] For example, the first node can determine at least one participating node to participate in the VFL task from at least one second node that satisfies the conditions for performing the VFL task.

[0102] In one embodiment, the first node determines at least one participating node to participate in a VFL task from at least one second node that satisfies the conditions for executing a VFL task, which includes determining a second node as a participating node to participate in a VFL task initiated by the first node, given that the second node agrees to participate in the VFL task initiated by the first node. In other words, if a second node satisfies the conditions for executing a VFL task and agrees to participate in a VFL task initiated by the first node, that second node can be determined as a participating node to participate in the VFL task.

[0103] In some embodiments, the method may further include the first node transmitting data requirements information to at least one second node for determining the data to perform the VFL task. Accordingly, each of the second nodes within the at least one second node may receive data requirements information from the first node.

[0104] For example, data requirements information may include, for instance, the type of data required to perform a VFL task, and / or the validity period of the required data. Here, the data type may include the data format (e.g., 8-bit quantization or 24-bit quantization) and the type of data content (e.g., application layer data, device power data, device communication performance data, etc.).

[0105] In one possible configuration, after receiving data requirements information from the first node, the second node can decide whether to agree to participate in a VFL task initiated by the first node based on its local configuration or relevant policies. For example, if the data that meets the corresponding data requirements exists locally on the second node, it can agree to participate in the VFL task; if the data that meets the corresponding data requirements does not exist locally on the second node, it can refuse to participate in the VFL task. Furthermore, the second node can send a reply to the first node indicating whether it agrees to participate in the VFL task.

[0106] For example, the second node may send back to the first node a third piece of information indicating whether it agrees to participate in the VFL task. This method may include the second node sending the third piece of information to the first node, which indicates whether it agrees to participate in the VFL task initiated by the first node. The first node may receive the third piece of information from the second node accordingly. Based on the third piece of information, the first node can know whether the second node agrees to participate in the VFL task initiated by the first node.

[0107] Figure 6 shows another communication method according to an embodiment of the present application, which may include the following steps.

[0108] In S601, the second node transmits capability information of the second node to the first network element. In response, the first network element receives the capability information of the second node.

[0109] In this step, the second node can transmit capability information of the second node to the first network element. Here, the capability information of the second node can represent the second node's ability to perform the VFL task. Based on the capability information of the second node, the first network element can determine whether the second node satisfies the conditions for performing the VFL task. In other words, whether the second node satisfies the conditions for performing the VFL task is determined by the first network element based on the capability information of the second node, and in other words, the capability information of the second node can be used by the first network element to determine whether the second node satisfies the conditions for performing the VFL task.

[0110] For example, the capability information of the second node can be used to indicate any one of the following:

[0111] 1) Capability information indicating whether the second node supports the execution of VFL tasks. For example, the capability information for the second node may include an indication that the second node supports the execution of VFL tasks, indicating whether the second node supports VFL tasks or not.

[0112] 2) Capability information indicating the period during which the second node supports the execution of VFL tasks. For example, the capability information of the second node can be used to indicate that the second node can execute VFL tasks between 9:00 and 21:00.

[0113] 3) Capability information for a second node, which indicates whether it can function as a VFL task initiator, for example, the capability information for a second node may include an indication of whether the second node can function as a VFL task initiator, showing that the second node can or cannot function as a VFL task initiator.

[0114] 4) Computing power information of the second node, for example, the maximum computing power or range of computing power that the second node can provide.

[0115] 5) This is capability information regarding the service area of ​​the second node. Here, if the second node is an access network element, the service area of ​​the second node can also be understood as the signal coverage area of ​​the access network element.

[0116] In some embodiments, capability information of a second node can be included in the parameter information of the second node and transmitted by the second node to the first network element, the method further comprising the second node transmitting the parameter information of the second node to the first network element, the first network element receiving and storing the parameter information of the second node, in other words, the first network element can collect or determine the parameter information of the second node. Here, the parameter information of the second node may include, for example, at least one of the capability information of the second node, the identifier (ID) of the second node, the address information of the second node, and the identifier of the application corresponding to the second node.

[0117] If there are multiple second nodes, it can be understood that the parameter information for each second node includes at least one of the above parameter information.

[0118] For example, the second node is a terminal, and the terminal's parameter information may include, for example, one of the following: terminal capability information, terminal identifier, terminal address information (e.g., IP address), and terminal type (e.g., mobile, vehicle, etc.). Here, terminal capability information can be used to indicate at least one of the following: whether the terminal supports the execution of VFL tasks, for what period the terminal supports the execution of VFL tasks, whether the terminal can function as a VFL task initiator, and the terminal's computing power.

[0119] As another example, the second node is an access network element, and the parameter information of the access network element may include, for example, at least one of the following: the capability information of the access network element, the identifier of the access network element (e.g., gNB ID), and the address information of the access network element. Here, the capability information of the access network element can be used to indicate at least one of the following: whether the access network element supports the execution of VFL tasks, for what period the access network element supports the execution of VFL tasks, whether the access network element can function as a VFL task initiator, the computing power of the access network element, and the service area of ​​the access network element.

[0120] As another example, the second node is a core network element, and the parameter information of the core network element may include, for example, at least one of the following: the capability information of the core network element, the address information of the core network element, and the type of the core network element (e.g., SMF, AMF, etc.). Here, the capability information of the core network element can be used to indicate at least one of the following: whether the core network element supports the execution of VFL tasks, for what period the core network element supports the execution of VFL tasks, whether the core network element can function as a VFL task initiator, the computing power of the core network element, and the service area of ​​the core network element.

[0121] As a further example, the second node is an AF, and the AF's parameter information may include, for example, one of the following: AF capability information, AF identifier, AF address information, and the identifier of the application corresponding to the AF. Here, the AF capability information can be used to indicate at least one of the following: whether the AF supports the execution of VFL tasks, for what period the AF supports the execution of VFL tasks, whether the AF can act as a VFL task initiator, and the AF's computing power.

[0122] If the second node is an AF, one possible case is that if the second node is a trusted AF, the second node can directly transmit its parameter information to the first network element. Another possible case is that if the second node is an untrusted third-party AF, the second node can transmit (transfer) its parameter information to the first network element via a NEF, and accordingly, the first network element can receive the second node's parameter information via the NEF. In this case, the second node may map its parameter information to the parameter information of an event (for example, let's call it the second event), in other words, the second node generates the parameter information of the second event based on its parameter information, and furthermore, the second node can transmit the parameter information of the second event to the NEF, and the NEF can transmit the AF's parameter information to the first network element by transferring the parameter information of the second event to the first network element.

[0123] In other words, if the second node is an untrusted third-party AF, the second node sending its parameter information to the first network element includes the second node sending the parameter information for the second event to the first network element via the NEF, and the parameter information for the second event is generated based on the parameter information of the second node. Exemplarily, the parameter information for the second event can be used to indicate at least one of the following: whether the AF supports the execution of the second event, for what period the AF supports the execution of the second event, whether the AF can act as the initiator of the second event, the computing power of the AF, and the identifier of the application corresponding to the AF, where the second event is associated with a VFL task, or in other words, executing the second event can be understood as executing a VFL task.

[0124] In one embodiment, for each second node, the parameter information of the second node can be included, for example, in the second registration request information, which is the registration request information that the second node sends to the first network element. In other words, the second node can register its parameter information with the first network element by sending the second registration request information to the first network element.

[0125] In some embodiments, each second node may transmit its parameter information to the first network element before the first node initiates the VFL task, thereby allowing the first network element to receive the parameter information of each second node before the first node initiates the VFL task. Furthermore, when the first node initiates the VFL task, the method may further include the first network element obtaining capability information for at least one second node. For example, the first network element may obtain capability information for each second node from the received parameter information for each second node. Furthermore, for each second node, the first network element can determine, based on the capability information of the second node, whether the conditions for the second node to perform the VFL task are met.

[0126] In some embodiments, the first node can transmit (register) its parameter information to a first network element, the method which may further include the first node transmitting its parameter information to the first network element. Here, the parameter information of the first node may include, for example, at least one of the following: the capability information of the first node, the identifier of the first node, the address information of the first node, and the identifier of the application corresponding to the first node.

[0127] For example, the capability information of the first node can be used to indicate at least one of the following: whether the first node supports the execution of VFL tasks, for what period the first node supports the execution of VFL tasks, whether the first node can act as a VFL task initiator, the computing power of the first node, and the service area of ​​the first node.

[0128] It is important to understand that the specific meaning of each parameter in the first node is the same as the meaning of each parameter in the second node, and they can be mutually referenced; therefore, we will not explain them again here.

[0129] If the first node is an AF, one possible case is that if the first node is a trusted AF, the first node can directly transmit its parameter information to the first network element. Another possible case is that if the first node is an untrusted third-party AF, the first node can transmit (transfer) its parameter information to the first network element via a NEF, and the first network element can receive the first node's parameter information via the NEF. In this case, the first node may map its parameter information to the parameter information of an event (for example, let's call it the first event), in other words, the first node can generate the parameter information of the first event based on its parameter information, and further transmit the AF's parameter information to the first network element by sending the parameter information of the first event to the NEF, which then forwards the parameter information of the first event to the first network element.

[0130] In other words, if the first node is an untrusted third-party AF, the first node sending its parameter information to the first network element includes the first node sending the parameter information for the first event to the first network element via the NEF, and the parameter information for the first event is generated based on the parameter information of the first node. Exemplaryly, the parameter information for the first event can be used to indicate at least one of the following: whether the AF supports the execution of the first event, for what period the AF supports the execution of the first event, whether the AF can act as the initiator of the first event, the computing power of the AF, and the identifier of the application corresponding to the AF. Here, the first event is associated with a VFL task, and in other words, executing the first event can be understood as executing a VFL task.

[0131] In one embodiment, the parameter information of the first node is included, for example, in the first registration request information, which is the registration request information that the first node sends to the first network element. In other words, the first node can register its parameter information with the first network element by sending the first registration request information to the first network element.

[0132] In S602, the first network element transmits first information to the first node. In response, the first node receives the first information.

[0133] Here, the first piece of information can be used to indicate at least one second node that satisfies the conditions for executing a VFL task initiated by the first node.

[0134] In S602, the first network element can send first information to the first node so as to indicate to the first node at least one second node that satisfies the conditions for performing the VFL task. In this way, the first node can know from the first information at least one second node that satisfies the conditions for performing the VFL task, thereby ensuring that the first node finds the appropriate node and performs the VFL task.

[0135] As an example, the first information may include, for instance, identifiers and / or address information of at least one second node that satisfies the conditions for performing a VFL task.

[0136] In some embodiments, the method may further include the first node transmitting second information to the first network element before the first network element transmits first information to the first node, the second information being for requesting the first information, and accordingly the first network element may receive the second information from the first node.

[0137] In other words, if the first node needs to find at least one second node that satisfies the conditions for executing a VFL task, the first node can send second information to the first network element to request first information. After receiving the second information from the first node, the first network element can determine, based on the capability information of each second node obtained, at least one second node that satisfies the conditions for executing a VFL task initiated by the first node, and indicate this at least one second node to the first node by feeding back the first information to the first node.

[0138] If the first node is an AF, one possible case is that if the first node is a trusted AF, the first node can directly transmit the second information to the first network element. Another possible case is that if the first node is an untrusted third-party AF, the first node can transmit (transfer) the second information to the first network element via a NEF, and accordingly, the first network element can receive the second information from the first node via the NEF.

[0139] For example, the second piece of information may be used to indicate any one of the following:

[0140] 1) This involves executing a VFL task; in other words, the type of task initiated by the first node is a VFL task.

[0141] 2) This is the period during which the VFL task is executed, or in other words, the period during which the first node attempts to execute the VFL task, for example, from 9:00 to 21:00.

[0142] 3) To act as the initiator of the VFL task; in other words, the first node attempts to be the initiator of the VFL task.

[0143] 4) Computing power to run VFL tasks, for example, the minimum computing power required to run a VFL task initiated by the first node.

[0144] 5) This is the service area where the VFL task is executed; in other words, it is the service area where the first node attempts to execute the VFL task.

[0145] For example, assuming that the second information indicates that the period during which the first node intends to perform the VFL task is from 9:00 to 21:00 and the service area in which the VFL task is intended to be performed is the first service area, the first network element, after receiving the second information from the first node, compares the capability information of each second node stored in the first network element with the contents of the second information, determines at least one second node that supports performing the VFL task between 9:00 and 21:00 and whose service area includes the first service area, and transmits the first information containing information about that at least one second node (e.g., identifier and / or address information) to the first node.

[0146] In S603, the first node determines at least one participating node that will participate in the VFL task.

[0147] For example, the first node can determine at least one participating node to participate in the VFL task from at least one second node that satisfies the conditions for performing the VFL task.

[0148] It should be understood that, in the embodiments of this application, "satisfying the conditions for executing a VFL task" means satisfying the conditions for executing a VFL task initiated by the first node. For brevity, hereafter, "at least one second node that satisfies the conditions for executing a VFL task" will be abbreviated as "at least one second node," that is, "at least one second node" as described later will mean "at least one second node that satisfies the conditions for executing a VFL task" without causing ambiguity.

[0149] In one embodiment, determining at least one participating node to participate in a VFL task from at least one second node includes determining a second node as a participating node to participate in a VFL task if the second node agrees to participate in a VFL task initiated by a first node. In other words, if a second node satisfies the conditions for executing a VFL task and agrees to participate in a VFL task initiated by a first node, that second node can be determined as a participating node to participate in the VFL task.

[0150] In some embodiments, the method may include, for each of at least one second node, the second node transmitting third information to the first node, the third information indicating whether it agrees to participate in a VFL task initiated by the first node. Accordingly, the first node can receive the third information from the at least one second node. Based on the third information, the first node can know whether each second node agrees to participate in a VFL task initiated by the first node.

[0151] In some embodiments, the method may further include the first node transmitting data requirements information to at least one second node for determining the data to perform the VFL task. Accordingly, each of the second nodes within the at least one second node may receive data requirements information from the first node.

[0152] For example, data requirements information may include, for instance, the type of data required to perform a VFL task, and / or the validity period of the required data. Here, the data type may include the data format (e.g., 8-bit quantization or 24-bit quantization) and the type of data content (e.g., application layer data, device power data, device communication performance data, etc.).

[0153] For example, if we assume that the data requirements information specifies that the type of data required to perform the VFL task is 8-bit quantized application layer data, and that the validity period of the required data is the first period, then the second node that receives the data requirements information can determine, based on the data requirements information, that the local data that is valid within the first period and of type 8-bit quantized application layer data is the data to perform the VFL task.

[0154] In one possible configuration, after receiving data requirements information from the first node, the second node can decide whether to agree to participate in a VFL task initiated by the first node based on its local configuration or relevant policies. For example, if the data that satisfies the corresponding data requirements exists locally on the second node, it can agree to participate in the VFL task; if the data that satisfies the corresponding data requirements does not exist locally on the second node, it can refuse to participate in the VFL task. Furthermore, the second node can return its decision regarding whether to agree to participate in the VFL task to the first node using third information.

[0155] If the first node is an AF, one possible case is that if the first node is a trusted AF, the first node can directly send data requirements information to at least one second node. Another possible case is that if the first node is an untrusted third-party AF, the first node can send (transfer) data requirements information to at least one second node via NEF, and accordingly, the second node can receive data requirements information from the first node via NEF.

[0156] In S604, the first node transmits the VFL model to the participating node. The participating node receives the VFL model accordingly.

[0157] Here, the participating node can be understood as a second node that has been determined to be a participating node, or as a second node that participates in the VFL task initiated by the first node. If the first node is an AF, it can be understood that the first node sends the VFL model to the participating node, and the server corresponding to the first node (i.e., the server corresponding to the AF) also sends the VFL model to the participating node.

[0158] In S603, the first node can determine from at least one second node at least one participating node to participate in the VFL task, where each participating node corresponds to one VFL model. In S604, the first node can send the corresponding VFL model to each participating node, where the VFL model can be used to perform the VFL task. In one possible form, the VFL model may include, for example, information about the corresponding participating node (e.g., identifier and / or address information of the participating node).

[0159] Here, a VFL model corresponding to a participating node can be understood as also being a VFL model required for that participating node. This VFL model can serve as the initial model when executing a VFL task at the participating node. In one embodiment, the VFL model required for the participating node can be generated by a first node. For example, if the participating node includes terminal equipment, access network elements, and core network elements, the first node can generate a VFL model required for the terminal equipment, a VFL model required for the access network elements, and a VFL model required for the core network elements. Furthermore, the first node can transmit the VFL models required for each participating node to the corresponding participating node.

[0160] In some scenarios, the first node cannot directly send the VFL model to the corresponding participant node. In this situation, the first node can transfer the VFL model corresponding to the participant node to the participant node via some (or more) intermediate devices.

[0161] For example, if the first node is an AF and the participating node is a terminal device or access network device, the AF cannot directly transmit a VFL model to the terminal device or access network device. In this situation, the first node transmitting a corresponding VFL model to the participating node may include the first node transmitting a corresponding VFL model to the terminal device or access network network element via a second network element. Here, the second network element may be a single network element, or may include two or more network elements, and the embodiments of this application are not limited thereto. In one embodiment, after receiving a VFL model, the second network element can forward the VFL model to the corresponding participating node based on the identifier and / or address information of the participating node contained in the VFL model. For example, if VFL model #1 includes the identifier and address information of a terminal device, the first node can transmit VFL model #1 to the terminal device based on the identifier and address information of the terminal device. Furthermore, for example, if VFL model #2 includes identifier and address information of an access network element, the first node can transmit VFL model #2 to the access network element based on the identifier and address information of the access network element.

[0162] In some scenarios, if the first node can directly transmit the VFL model to the corresponding participating node without forwarding it through an intermediate device, the VFL model does not need to include information about the participating node (e.g., identifier and / or address information).

[0163] In some embodiments, the method may further include the first node transmitting the same association identifier to at least one participating node, which may be used to identify different execution results of the same VFL task, or to aggregate different execution results of the same VFL task. Accordingly, the second node may receive the association identifier from the first node. In one embodiment, the first node may transmit the VFL model and the association identifier to the corresponding participating node with the same information.

[0164] Here, different execution results for the same VFL task may include, for example, execution results that occur when different nodes (e.g., the first node, or the second node determined to be a participating node) execute the same VFL task.

[0165] In one possible form, the execution results may include model training results and model reasoning results. In this case, the association identifier may include a model association identifier and a result association identifier, where the model association identifier can be used to identify and / or aggregate different model training results for the same VFL task, and the result association identifier can be used to identify and / or aggregate different model reasoning results for the same VFL task.

[0166] In the embodiments of this application, when each participating node receives a corresponding VFL model, the received VFL model can be used as the initial model for executing the VFL task, and the data for executing the VFL task determined in S603 based on the data requirements information can be used as input data. Simultaneously, the first node can generate (acquire) a VFL model corresponding to the first node and determine the local data for executing the VFL task at the first node. Furthermore, each participating node and the first node can jointly execute the VFL task initiated by the first node based on their respective VFL models and local data.

[0167] In some embodiments, after executing the VFL task, the method may further include each participating node (in other words, a second node that participated in the VFL task) sending the execution result of the VFL task executed by the participating node to the first node, where the execution result may include an association identifier. Note that if the first node is an AF, sending the execution result of the VFL task executed by the participating node to the first node can be understood as sending the execution result of the VFL task executed by the participating node to the server corresponding to the first node (i.e., the server corresponding to the AF).

[0168] One possible case is that a participating node can directly send the execution result of the VFL task it has performed to the first node. Another possible case is that a participating node (hereinafter referred to as participating node #1) cannot directly send the execution result to the first node. In this situation, participating node #1 first sends its execution result to an intermediate node (for example, participating node #2), and then the intermediate node forwards the execution result of participating node #1 to the first node.

[0169] When the intermediate node is designated as participant node #2, in the first embodiment (hereinafter referred to as embodiment #1), after receiving the execution result of participant node #1, participant node #2 can simultaneously send the execution result of participant node #1 to the first node and send its own execution result to the first node. Here, the execution results of both participant node #1 and participant node #2 include the same association identifier.

[0170] In the second embodiment (hereinafter referred to as Embodiment #2), participating node #2 can also aggregate the execution results of participating node #2 and the execution results of participating node #1 that it receives, and further transmit the obtained aggregated results to the first node. For example, the execution results of participating node #1 and participating node #2 may contain the same association identifier. In this way, when participating node #2 aggregates the execution results, by aggregating execution results (execution results of participating node #1 and participating node #2) that have the same association identifier, it is possible to ensure that the aggregated targets are the execution results of the same VFL task, and further ensure that the correct aggregation of execution results is achieved. Here, the obtained aggregated results may also contain the association identifier.

[0171] In the third embodiment (hereinafter referred to as Embodiment #3), participating node #2 can receive the execution results of other participating nodes (hereinafter referred to as Participating Node #33) in addition to the execution results of Participating Node #1, Participating Node #2 and Participating Node #3, aggregate the execution results of Participating Node #1, Participating Node #2 and Participating Node #3, and transmit the obtained aggregated result to the First Node. For example, the execution results of Participating Node #1, Participating Node #2 and Participating Node #3 may contain the same association identifier. In this way, when Participating Node #2 aggregates the execution results, by aggregating execution results that have the same association identifier (execution results of Participating Node #1, Participating Node #2 and Participating Node #3), it is possible to ensure that the aggregated targets are the execution results of the same VFL task, and furthermore, that the correct aggregation of execution results is ensured. Here, the obtained aggregated result may contain the association identifier.

[0172] In other words, with respect to a second node participating in a VFL task (e.g., participating node #2), in embodiments #2 and #3 described above, the method may further include receiving the execution result (e.g., denoted as the second execution result) of at least one participating node (e.g., participating node #1 and / or participating node #3) executing the VFL task, and sending the aggregate result to the first node, wherein the aggregate result is obtained by aggregating the execution result of at least one participating node executing the VFL task (i.e., at least one second execution result) and the execution result of the second node executing the VFL task (e.g., denoted as the third execution result), and the aggregate result may include an association identifier. In one possible form, each execution result of an execution in which at least one participating node performs a VFL task (i.e., each second execution result), and the execution result of an execution in which the second node performs a VFL task (i.e., a third execution result), all contain the same association identifier, so that the aggregate result can be obtained by aggregating based on the same association identifier.

[0173] Accordingly, the first node can receive an execution result (e.g., referred to as the first execution result) corresponding to the VFL task initiated by the first node, where the first execution result includes an execution result from at least one participating node executing the VFL task (e.g., the execution result obtained according to Embodiment #1 described above, e.g., referred to as the second execution result) and / or at least one aggregate result, where each of the at least one aggregate result is obtained by aggregating based on execution results from at least two participating nodes executing the VFL task (i.e., at least two second execution results) (e.g., obtained according to Embodiment #2 or Embodiment #3 described above). Here, each of the execution results from at least one participating node executing the VFL task (i.e., each second execution result) and / or each of the at least one aggregate result all include the same association identifier, and each of the at least one aggregate result can be obtained by aggregating based on the same association identifier.

[0174] In some embodiments, the method may further include the first node aggregating the execution results corresponding to VFL tasks having the same association identifier with the execution results of the VFL tasks executed by the first node.

[0175] As an example, after the first node receives the execution result corresponding to a VFL task initiated by the first node, it can aggregate the execution result corresponding to the VFL task with its own execution result to obtain a final aggregated result. Here, the execution result corresponding to the VFL task received by the first node and the execution result of the first node itself can both contain the same association identifier. In this way, when the first node aggregates the execution results, it can aggregate execution results that have the same association identifier (the execution result corresponding to the VFL task received by the first node and the execution result of the first node itself), thereby ensuring that the aggregated targets are the execution results of the same VFL task, and further ensuring that the correct aggregate of execution results is achieved.

[0176] The communication method according to the embodiment of this application will be described above with reference to Figures 5 and 6. To facilitate understanding of the embodiment of this application, the possible implementation flows of the communication method according to the embodiment of this application will be described below with reference to Figures 7 to 9. For the sake of explanation, in the following example, we assume that the first network element is NRF, the first node is AF, and the second node includes UE, RAN and core network NF (or referred to as core network element). Here, there may be one UE, RAN or core network NF, or multiple (two or more), and the embodiment of this application is not limited to this.

[0177] To execute VFL, the initiating node must first find a node capable of executing VFL, in other words, a node that meets the conditions for executing the VFL task. For this reason, each node must register its information with the NRF before executing the VFL task. When the AF is the initiating node, the UE, RAN, and core network elements register their ability to support VFL with the NRF, allowing the NRF to find a suitable node for the AF to execute the VFL task.

[0178] Figure 7 is a flowchart 1 of a possible implementation of the communication method according to an embodiment of this application, and the implementation flow may include the following steps.

[0179] In S701, the UE sends a registration request to the NRF.

[0180] A UE can register its information with the NRF by sending a registration request to the NRF. For example, such a registration request includes a UE profile (corresponding to the terminal parameter information in the above embodiment), and the UE profile includes, but is not limited to, at least one of the following: UE ID, whether the UE supports VFL, the period during which the UE can support VFL, the type of UE, UE address information (e.g., IP address), an indication of whether the UE supports acting as a VFL initiator, and UE computing power information.

[0181] In S702, the RAN sends a registration request to the NRF.

[0182] A RAN can register its information with the NRF by sending a registration request to the NRF. For example, such a registration request includes a RAN profile (corresponding to the parameter information of the access network element in the above embodiment), and the RAN profile includes, but is not limited to, at least one of the following: the RAN's ID (e.g., gNB ID), whether the RAN supports VFL, the period during which the RAN can support VFL, the RAN's address information, an indication of whether it supports functioning as a VFL initiator, the RAN's computing power information, and the RAN's service area (coverage area) information.

[0183] In S703, the core network NF sends a registration request to the NRF.

[0184] A core network NF can register its information with the NRF by sending a registration request to the NRF. For example, such a registration request includes a profile of the core network NF (corresponding to the parameter information of the core network element in the above embodiment), and the profile of the core network NF includes, but is not limited to, at least one of the following: the type of NF, whether the NF supports VFL, the period during which the NF can support VFL, the address information of the NF, an indication of whether the NF supports functioning as a VFL initiator, the computing power information of the NF, and the service area information of the NF.

[0185] In S704, the AF sends a registration request to the NRF.

[0186] An AF can register its information with the NRF by sending a registration request to the NRF. For example, such a registration request includes an AF profile (corresponding to the AF parameter information in the above embodiment), and the AF profile includes, but is not limited to, at least one of the following: AF ID, Application ID corresponding to the AF, whether the AF supports VFL, the period during which the AF can support VFL, AF address information, an indication of whether the AF supports functioning as a VFL initiator, and AF computing power information.

[0187] Furthermore, if the AF is an untrusted third-party AF, the AF can register its information with the NRF via the NEF. For example, the AF can first send its profile to the NEF, the NEF can generate an NEF profile, and the NEF can register the NEF profile with the NRF. The NEF can map the AF's information to the information of a single event, in other words, to a single event ID, thereby allowing the NEF to register the information of that event (corresponding to the parameter information of the first event in the above embodiment) as the NEF profile with the NRF. For example, the information of the event includes, but is not limited to, at least one of the following: the NEF ID, whether the AF supports VFL, the application ID corresponding to the AF, the period for which the event is supported, an indication of whether the AF supports initiating the event, and the AF's computing power information.

[0188] It should be understood that this embodiment does not limit the execution order of S701 to S704.

[0189] In S705, the NRF stores the profiles of each node.

[0190] After receiving registration requests from each node (UE, RAN, core network NF and AF), the NRF can store a profile of each node and mark each node as an available, active node. The NRF can then send a registration response to each node in response to the acceptance of its registration. The step of the NRF sending registration responses to each node may include S706-S709.

[0191] In S706, the NRF sends a registration response to the network NF.

[0192] In S707, the NRF sends a registration response to the RAN.

[0193] In S708, the NRF sends a registration response to the UE.

[0194] In S709, the NRF sends a registration response to the AF.

[0195] It should be understood that this embodiment does not limit the execution order of S706 to S709.

[0196] As shown in the implementation flow in Figure 7, each node can register its information in the NRF, so that when executing a VFL task, the initiating node can use VFLNRF to find the appropriate node and perform the task.

[0197] Figure 8 is a flowchart 2 of a possible implementation of the communication method according to an embodiment of this application. In one embodiment, the implementation flow shown in Figure 8 can be executed after the implementation flow shown in Figure 6, that is, in the implementation flow shown in Figure 8, it is assumed that the information of each node has already been registered in the NRF. As shown in Figure 8, the implementation flow may include the following steps.

[0198] In S801, AF sends a node discovery request to NRF (corresponding to the second piece of information in the above embodiment).

[0199] When an AF acts as the initiator of a VFL task, the AF may send a node discovery request to the NRF when it is attempting to find a node capable of running the VFL task. For example, a node discovery request may include, but is not limited to, an instruction to run the VFL, the period for which the VFL is to be run, an instruction to act as the initiator of the VFL, and the service area for which the VFL is to be run.

[0200] In some embodiments, if the AF is a third-party AF, the AF can send a node discovery request to the NRF via the NEF, and the parameters included in the node discovery request are the same.

[0201] In S802, the NRF authorizes the AF's node discovery request.

[0202] In this step, the NRF may authorize the AF's node discovery request. In one embodiment, before authorizing the AF's node discovery request, the NRF may determine, based on the AF's profile information, whether the AF can support VFL and whether it can support functioning as a VFL initiator. Furthermore, if the AF can support VFL and can support functioning as a VFL initiator, the NRF authorizes the AF's node discovery request.

[0203] In S803, the NRF sends a node discovery request response (corresponding to the first information in the above embodiment) to the AF.

[0204] Since the NRF stores profile information of the relevant nodes, based on the information included in the node discovery request S801, the NRF can determine the nodes that meet the criteria for the AF (e.g., UEs, RANs, and core network NFs that meet the criteria) and return information about the nodes that meet the criteria (e.g., node ID and / or address information) to the AF in the node discovery request response.

[0205] Here, "nodes that meet the conditions" can be understood as nodes that meet the conditions included in the node discovery request, or as nodes that meet the conditions for executing the VFL task initiated by AF. For the sake of explanation, we will assume below that UR, RAN, and the core network NF are all nodes that meet the conditions.

[0206] In some embodiments, if the AF is a third-party AF, the NRF sends a node discovery request response to the AF via the NEF.

[0207] In S804, the AF sends a VFL preparation request to the NEF.

[0208] If the AF is a third-party AF, the AF can send a VFL preparation request to the NEF, which can then forward the VFL preparation request to each node. The VFL preparation request may include the type of data required to perform the VFL task and the data's validity period (corresponding to the data requirements information in the above embodiment).

[0209] In S805, the NEF sends a VFL preparation request to each node.

[0210] After the NEF receives a VFL preparation request from the AF, it can send the VFL preparation request to each node that meets the conditions (as shown in Figure 8, the NEF can send the VFL preparation request to the UE, RAN, and core network NF respectively), thereby allowing each node to pre-prepare the data necessary to execute the VFL task based on the type of data required to execute the VFL task and the data's validity period, as included in the VFL preparation request.

[0211] In S806, each node sends a VFL preparation request response message (corresponding to the third piece of information in the above embodiment) to the NEF.

[0212] After receiving a VFL ready request, each node may decide whether to join the VFL task based on its local configuration and send a decision to the NEF indicating whether it agrees to join, which may be included, for example, in the VFL ready request response message, or the VFL ready request message may be used to indicate whether each node agrees to join the VFL task initiated by the AF.

[0213] In S807, the NEF selects a node.

[0214] Based on the VFL preparation request response messages fed back by each node, the NEF can select VFL nodes based on the AF's request. For example, the NEF can select nodes that meet the conditions for executing a VFL task initiated by the AF and that agree to join the VFL task.

[0215] In S808, the NEF sends the transmission node selection result to the AF.

[0216] In this step, the NEF can return the node selection result in S807 to the AF, where the node selection result may include information about the selected node (e.g., the node's ID and / or address information), allowing the AF to determine that the selected node is a participating node that will participate in the VFL task.

[0217] Note that S804-S808 will be explained using the example that the AF is a third-party AF. If the AF is a trusted AF, the AF can directly send VFL readiness requests to each node that meets the conditions (e.g., UE, RAN, and core network NF), and each node that meets the conditions can directly send a VFL readiness request response message to the AF, which can be used to indicate whether each node agrees to join the VFL task initiated by the AF, and furthermore, the AF can determine which nodes that meet the conditions and agree to join the VFL task will be participating nodes in the VFL task.

[0218] As shown in the implementation flow in Figure 8, the AF can discover and determine a suitable node for the VFL task by sending a node discovery request to the NRF.

[0219] Figure 9 is a flowchart 3 of a possible implementation of the communication method according to an embodiment of this application. In one embodiment, the implementation flow shown in Figure 9 can be executed after the implementation flow shown in Figure 7, that is, in the implementation flow shown in Figure 9, it is assumed that the AF has already determined each node to participate in the VFL task. In the following implementation flow, the participating nodes determined by the AF are the UE, RAN, and core network NF. As shown in Figure 9, the implementation flow may include the following steps.

[0220] In S901, the server corresponding to AF sends the VFL model and association identifier for each node to the core network NF.

[0221] Here, the VFL model corresponding to each node can be used as the initial model for executing the VFL task of each node.

[0222] In this step, the AF-enabled server can first send the required (corresponding) VFL model for each node (UE, RAN, core network NF) participating in the VFL task to the core network NF. Here, the model can include information about the corresponding node. For example, if the required model for the UE is model #1 and the required model for the RAN is model #2, then model #1 may include information about the UE (such as the UE's ID and / or address information) to indicate that model #1 needs to be sent to the UE, and model #2 may include information about the RAN (such as the RAN's ID and / or address information) to indicate that model #2 needs to be sent to the RAN.

[0223] In this step, AF can also send the same association identifier (association ID) to each node. In VFL, different nodes need to employ local models and data to train the models and infer results, and then aggregate them at the initiator. Therefore, to facilitate the association of VFL execution results (including model training results and / or inference results) generated at each node, AF can send the same association identifier to each node. In this way, after each node obtains a VFL execution result through computation, the association identifier can be added to that node's VFL execution result, thereby ensuring that the results generated at different nodes for the same VFL task have the same association identifier, and further ensuring that the aggregate node performs the correct aggregation on the VFL execution results obtained by each node. Here, "aggregate node" can be understood as a node that aggregates the VFL execution results of at least two nodes.

[0224] In one embodiment, the association identifier can be included, for example, in the VFL model corresponding to each node.

[0225] In S902, the core network NF receives the VFL model and association identifier corresponding to the RAN from the RAN.

[0226] After obtaining the corresponding VFL model and association identifier required by the RAN, the core network NF can transmit the VFL model and association identifier to the RAN based on the RAN information contained in the VFL model (e.g., the RAN's ID and / or address information).

[0227] In S903, the core network NF sends the VFL model corresponding to the UE and an association identifier to the UE.

[0228] After obtaining the corresponding VFL model and association identifier required by the UE, the core network NF can transmit the VFL model and association identifier to the UE based on the UE information contained in the VFL model (e.g., the UE's ID and / or address information). In one embodiment, the VFL can transmit the VFL model and association identifier corresponding to the UE to the UE via the RAN.

[0229] In some embodiments, the server corresponding to the AF can also directly send the VFL model corresponding to the UE and its association identifier to the UE via the application layer. In this case, the flow method of the implementation further includes S904.

[0230] In S904, the server corresponding to AF sends the VFL model and association identifier corresponding to UE to UE via the application layer.

[0231] When executing S904, the server corresponding to AF in S901 does not need to send the VFL model corresponding to UE to the core network NF, and in that case, it can be understood that there is no need to execute S903.

[0232] In S905, each node executes the VFL task.

[0233] After the distribution of the VFL model is complete, each node can perform federated learning on the same task based on its association identifier. For example, during the execution of a VFL task, each node may interact with intermediate results (e.g., gradients, losses, etc.). Since the results produced by each node will have the same association identifier, processing the results with the same association identifier ensures that each node performs federated learning on the same task.

[0234] In S906, the UE sends the execution result of the VFL task performed by the UE to the core network NF.

[0235] After the UE executes the VFL task, it can send the execution results, including the obtained association identifier, to the core network NF.

[0236] In S907, the RAN sends the execution results of the VFL task that the RAN has performed to the core network NF.

[0237] After the RAN executes the VFL task, the execution results, including the obtained association identifier, can be sent to the core network NF.

[0238] In some embodiments, the implementation flow further includes S908.

[0239] In S908, the core network NF aggregates the execution results.

[0240] After receiving the execution results of VFL tasks by UN and RAN, the core network NF aggregates the execution results with the same association identifier (UN, RAN, and the core network NF's own execution results) based on the association identifier, and then, in S909, can send the aggregated results to the server corresponding to AF.

[0241] It should be understood that, in some embodiments, after the core network NF receives execution results from the UN and RAN, it can directly send the received execution results (execution results from the UE and RAN) and the execution results generated by the core network NF to the server corresponding to the AF without aggregating the execution results.

[0242] In S909, the core network NF sends the execution results corresponding to the VFL task to the server that supports AF.

[0243] One possible case is when the core network NF receives the execution results of the VFL task from the UN and RAN and then aggregates the execution results from each node. In this step, the execution result corresponding to the VFL task may be an aggregated result obtained by aggregating the execution results from each node, and this aggregated result includes an association identifier.

[0244] In another possible case, if the core network NF does not aggregate the execution results of each node after receiving the execution results of the VFL task from the UN and RAN, the execution results corresponding to the VFL task may include the execution results of each node (e.g., the execution results of the UN, RAN, and core network NF), and furthermore, each node's execution results may include the same association identifier.

[0245] In some embodiments, the UE can send the execution results of the UE directly to the server corresponding to the AF via the application layer, and the execution results include an association identifier, in which case the flow of the implementation may further include S710.

[0246] In S910, the UE sends the execution results of the VFL task performed by the UE to the AF-compatible server via the application layer.

[0247] When running S910, it is not necessary to run S906, and in S908, when the core network NF aggregates the execution results of each node, it is not necessary to aggregate the execution results of the UE.

[0248] In S911, the server supporting AF aggregates the execution results.

[0249] In this step, the server corresponding to the AF aggregates the results generated at all nodes to obtain the final VFL result. For example, if the core network NF aggregates the execution results of the UR, RAN and the core network NF, the server corresponding to the AF aggregates the aggregate result obtained by the core network NF aggregation with the execution result generated by the server corresponding to the AF based on the association identifier to obtain the final VFL task execution result. Here, the aggregate result obtained by the core network NF aggregation and the execution result generated by the server corresponding to the AF have the same association identifier.

[0250] As shown in the implementation flow in Figure 9, the AF sends an initial VFL model to the nodes participating in each VFL task, associates the execution results of the VFL tasks by each node using association identifiers, aggregates the execution results of each related node, and obtains the final execution result of the VFL task.

[0251] It should be understood that in the process described above, the AF being the initiator of the VFL task is only one example; in the solutions of the embodiments of this application, any node can also be the initiator of the VFL task. In one embodiment, in actual application scenarios of VFL, a label may exist on the UE or AF, and therefore, the UE or AF can typically be the initiator of the VFL task.

[0252] The embodiment of this application provides support for vertical federated learning across domains in a network architecture by describing the process in which each node registers, the initiator discovers and determines participating nodes, submits an initial model, and aggregates the execution results.

[0253] In the solution of the embodiment of this application, first, each node registers information related to its VFL with the NRF to ensure that the initiator can find the appropriate node. Then, the initiator can find each node that can participate in the VFL via the NRF. Furthermore, by sending an initial model and association identifier to each node, the initiator can ensure that each node performs the same VFL task and that the correct combination (aggregate) of results is obtained. This solution is, 1) The challenge of how the initiating node (e.g., UE or AF) selects appropriate nodes in other domains to run VFL, 2) The challenge of how the initiator sends the initial model to each node to train the model and / or infer results, and 3) We solved previously unsolvable technical challenges, including the issue of how the initiator relates the training / inference results generated at each node.

[0254] The solution of the embodiment of this application allows the initiator to link nodes in multiple domains and perform vertical federated learning to prevent data leakage from those domains, thereby linking higher-dimensional data to comprehensively analyze and predict the research subject (e.g., the UE experience) and optimize the research content within each domain (e.g., optimizing the UE experience).

[0255] According to the above embodiment, the embodiment of this application provides a corresponding communication device.

[0256] Figure 10 is a schematic diagram of the configuration of a communication device according to an embodiment of this application, which is applied to the first node, and as shown in Figure 10, the communication device 1000 is The system includes a first receiving module 1001 configured to receive first information from a first network element, and a decision module 1002 configured to determine at least one participating node from the at least one second node to participate in the VFL task, wherein the first information is used to indicate at least one second node that satisfies the conditions for performing the VFL task.

[0257] In some embodiments, whether the second node satisfies the conditions for performing a VFL task is determined by the first network element based on the capability information of the second node, which can be used to indicate at least one of the following: whether the second node supports the execution of a VFL task, for what period the second node supports the execution of a VFL task, whether the second node can act as the initiator of a VFL task, the computing power of the second node, and the service area of ​​the second node.

[0258] In some embodiments, parameter information for at least one second node is collected or determined by a first network element, where for each second node, the parameter information for the second node includes at least one of the following: capability information for the second node, identifier for the second node, address information for the second node, and identifier for the application corresponding to the second node.

[0259] In some embodiments, for each second node, the parameter information of the second node is included in the second registration request information, which is the registration request information that the second node sends to the first network element.

[0260] In some embodiments, the device 1000 further comprises a first transmit module configured to transmit parameter information of the device 1000 to a first network element before receiving first information from the first network element, the parameter information of the device 1000 including at least one of the following: capability information of the device 1000, identifier of the device 1000, address information of the device 1000, and identifier of an application corresponding to the device 1000, wherein the capability information of the device 1000 is used to indicate at least one of the following: whether the device 1000 supports the execution of VFL tasks, for what period of time the device 1000 supports the execution of VFL tasks, whether the device 1000 can function as a VFL task initiator, the computing power of the device 1000, and the service area of ​​the device 1000.

[0261] In some embodiments, device 1000 is an application function AF, and the first transmission module is configured to transmit parameter information of a first event to a first network element by a network exposure function (NEF), the first network element being generated based on the parameter information of device 1000, and the parameter information of the first event is used to indicate at least one of the following: whether the AF supports the execution of the first event, for what period the AF supports the execution of the first event, whether the AF can act as the initiator of the first event, the computing power of the AF, and an identifier for the application corresponding to the AF, where the first event is associated with a VFL task.

[0262] In some embodiments, the parameter information of the device 1000 is included in the first registration request information, which is the registration request information that the device 1000 transmits to the first network element.

[0263] In some embodiments, the device 1000 further includes a second transmitting module configured to send second information to the first network element to request first information before receiving first information from the first network element.

[0264] In some embodiments, the device 1000 is an AF, and the second transmission module is specifically configured to transmit second information to the first network element via the NEF.

[0265] In some embodiments, the second information is used to indicate at least one of the following: the execution of the VFL task, the duration for which the VFL task is executed, the role of the VFL task initiator, the computing power needs of the VFL task, and the service area in which the VFL task is executed.

[0266] In some embodiments, the decision module 1002 specifically determines a second node that agrees to participate in the VFL task as a participating node in the VFL task.

[0267] In some embodiments, the device 1000 further comprises a third transmission module configured to transmit data requirements information to at least one second node for determining the data required to perform a VFL task, the data requirements information including the type of data required to perform the VFL task and / or the validity period of the required data.

[0268] In some embodiments, the device 1000 further comprises a second receiving module configured to receive third information from at least one second node, the third information being used to indicate whether the second node agrees to participate in the VFL task.

[0269] In some embodiments, the device 1000 is an AF, and the third transmission module is configured to transmit data requirements information to at least one second node via the NEF.

[0270] In some embodiments, the device 1000 further comprises a fourth transmitting module configured to transmit a corresponding VFL model to at least one participating node, the VFL model being used for the execution of a VFL task, wherein the VFL model includes identifier and / or address information of the corresponding participating node.

[0271] In some embodiments, the device 1000 is an AF, the participating node is a terminal device or an access network network element, and the fourth transmitting module is specifically configured to transmit a corresponding VFL model to the terminal device or access network network element via the second network element.

[0272] In some embodiments, the device 1000 further comprises a fifth transmission module configured to transmit an association identifier to at least one participating node, the association identifier being used to aggregate different execution results of the same VFL task.

[0273] In some embodiments, the device 1000 further comprises a third receiving module configured to receive a first execution result corresponding to a VFL task, the first execution result including a second execution result in which at least one participating node performed the VFL task, and / or at least one aggregate result, where each of the at least one aggregate result is obtained by aggregating at least two second execution results.

[0274] In some embodiments, each second execution result and / or each aggregate result among at least one aggregate result contains the same association identifier.

[0275] In some embodiments, each aggregate result among at least one aggregate result is obtained by aggregating based on the same association identifier.

[0276] In some embodiments, the device 1000 further includes an aggregate module configured to aggregate execution results corresponding to VFL tasks having the same association identifier with execution results from when the device 1000 performed the VFL tasks.

[0277] In some embodiments, the device 1000 is a terminal device, an access network element, a core network element, or an AF.

[0278] In some embodiments, the second node includes at least one of terminal equipment, access network elements, core network elements, and AF.

[0279] In some embodiments, the first network element is a network memory function (NRF).

[0280] Figure 11 is a schematic diagram of the structure of a communication device according to an embodiment of this application, which is applied to the second node, and as shown in Figure 11, the communication device 1100 is The system includes a first transmitting module 1101 configured to transmit capability information of the device 1100 to a first network element, and the capability information of the device 1100 is used to determine whether the device 1100 meets the conditions for performing a vertically federated learning VFL task.

[0281] In some embodiments, capability information for device 1100 is used to indicate at least one of the following: whether device 1100 supports the execution of VFL tasks, for what period of time device 1100 supports the execution of VFL tasks, whether device 1100 can function as a VFL task initiator, the computing power of device 1100, and service area information of device 1100.

[0282] In some embodiments, the device 1100 further comprises a second transmitting module configured to transmit parameter information of the device 1100 to a first network element, the parameter information of the device 1100 including at least one of the following: capability information of the device 1100, an identifier of the device 1100, address information of the device 1100, and an identifier of an application corresponding to the device 1100.

[0283] In some embodiments, the apparatus 1100 is an application function AF, and the second transmission module is specifically configured to transmit the parameter information of the second event to the first network element by a network exposer function (NEF). The parameter information of the second event is generated based on the parameter information of the apparatus 1100. The parameter information of the second event is used to indicate at least one of whether the AF supports the execution of the second event, the period during which the AF supports the execution of the second event, whether the AF can function as the initiator of the second event, the computing power of the AF, and the identifier of the application corresponding to the AF. Here, the second event is associated with a VFL task.

[0284] In some embodiments, the parameter information of the apparatus 1100 is included in the second registration request information, and the second registration request information is the registration request information transmitted by the apparatus 1100 to the first network element.

[0285] In some embodiments, the apparatus 11 is further provided with a first receiving module configured to receive data requirement information from the first node. The data requirement information is used to determine the data for executing the VFL task, and the data requirement information includes the type of data required for the execution of the VFL task and / or the valid time of the required data.

[0286] In some embodiments, the apparatus 1100 is further provided with a third transmission module configured to transmit third information to the first node. The third information is used to indicate whether the apparatus 1100 agrees to participate in the VFL task.

[0287] In some embodiments, the first node is an AF, and the first receiving module is specifically configured to receive data requirement information from the first node by the NEF.

[0288] In some embodiments, the device 1100 further comprises a second receiving module configured to receive a VFL model from a first node, the VFL model being used for performing a VFL task, where the VFL model corresponds to the device 1100 and includes identifier and / or address information of the device 1100.

[0289] In some embodiments, the device 1100 further comprises a third receiving module configured to receive association identifiers from a first node, which are used to aggregate different execution results of the same VFL task.

[0290] In some embodiments, the device 1100 further comprises a third transmission module configured to send the execution results of the device 1100 performing a VFL task to a first node, the execution results including an association identifier.

[0291] In some embodiments, the device 1100 further comprises a fourth receiving module configured to receive a second execution result in which at least one participating node has performed a VFL task, and a fourth transmitting module that transmits an aggregate result to a first node, wherein the aggregate result is obtained by aggregating at least one second execution result and a third execution result in which the device 1100 has performed a VFL task, and the aggregate result includes an association identifier.

[0292] In some embodiments, the second and third execution results each have the same association identifier, and the aggregated result is obtained by aggregating based on the same association identifier.

[0293] In some embodiments, the first node is a terminal device, an access network element, a core network element, or an AF.

[0294] In some embodiments, the device 1100 is a terminal device, an access network network element, a core network element, or an AF.

[0295] In some embodiments, the first network element is a network memory function (NRF).

[0296] Figure 12 is a schematic diagram of the structure of a communication device according to an embodiment of this application, which is applied to the first network element, and as shown in Figure 12, the communication device 1200 is The system includes an acquisition module 1201 configured to acquire capability information from at least one second node, the capability information from the second node being used to determine whether the second node meets the conditions for performing a vertically federated learning VFL task.

[0297] In some embodiments, the capability information of the second node can be used to indicate at least one of the following: whether the second node supports the execution of row VFL tasks, for what period the second node supports the execution of VFL tasks, whether the second node can act as the initiator of VFL tasks, the computing power of the second node, and the service area information of the second node.

[0298] In some embodiments, the device 1200 further comprises a first receiving module configured to receive parameter information of at least one second node before acquiring capability information of at least one second node, wherein for each second node, the parameter information of the second node includes at least one of the following: capability information of the second node, identifier of the second node, address information of the second node, and identifier of the application corresponding to the second node.

[0299] In some embodiments, for the second node, the parameter information of the second node is included in the second registration request information, which is the registration request information that the second node sends to the device 1200.

[0300] In some embodiments, the device 1200 further comprises a second receiving module configured to receive parameter information of a first node, the parameter information of the first node including at least one of the following: capability information of the first node, identifier of the first node, address information of the first node, and identifier of an application corresponding to the first node, wherein the capability information of the first node is used to indicate at least one of the following: whether the first node supports the execution of a VFL task, for what period the first node supports the execution of a VFL task, whether the first node can function as an initiator of a VFL task, the computing power of the first node, and the service area of ​​the first node.

[0301] In some embodiments, the parameter information of the first node is included in the first registration request information, which is the registration request information that the first node sends to the device 1200.

[0302] In some embodiments, the device 1200 further comprises a transmission module configured to transmit first information to a first node, the first information being used to indicate at least one second node that satisfies the conditions for performing a VFL task.

[0303] In some embodiments, the device 1200 further comprises a third receiving module configured to receive second information from the first node before transmitting first information to the first node, the second information being used to request the first information.

[0304] In some embodiments, the first node is a terminal device, an access network element, a core network element, or an application function AF.

[0305] In some embodiments, the second node includes at least one of terminal equipment, access network elements, core network elements, and AF.

[0306] In some embodiments, the device 1200 is a network memory function (NRF).

[0307] It should be understood that those skilled in the art can understand the related description of the communication device in the embodiments of this application by referring to the related description of the communication method in the embodiments of this application.

[0308] FIG. 13 is a schematic structural diagram of a communication device 1300 according to an embodiment of this application. The communication device may be a terminal device, a network device, or a core network device. The communication device 1300 shown in FIG. 13 includes a processor 1310, and the processor 1310 is configured to call and execute a computer program from a memory to implement the method in the embodiments of this application.

[0309] Optionally, as shown in FIG. 13, the communication device 1300 may further include a memory 1320. Here, the processor 1310 is configured to call and execute a computer program from the memory 1320 to implement the method in the embodiments of this application.

[0310] Here, the memory 1320 may be an independent device separate from the processor 1310, or may be integrated into the processor 1310.

[0311] Optionally, as shown in FIG. 13, the communication device 1300 may further include a transceiver 1330, and the processor 1310 can control the transceiver 1330 to communicate with other devices. Specifically, it can transmit information or data to other devices, or receive information or data transmitted from other devices.

[0312] Here, the transceiver 1330 may include a transmitter and a receiver. The transceiver 1330 may further include an antenna, and the number of antennas may be one or more.

[0313] Selectively, the communication device 1300 may be the first node of the embodiment of this application, and the communication device 1300 can implement the corresponding process implemented by the transmitting device in each method of the embodiment of this application, which will not be described again here for the sake of brevity.

[0314] Selectively, the communication device 1300 may be specifically the second node of the embodiment of this application, and the communication device 1300 can implement the corresponding process implemented by the second node in each method of the embodiment of this application, which will not be described again here for the sake of brevity.

[0315] Selectively, the communication device 1300 may be specifically the first network element of the embodiment of this application, and the communication device 1300 can implement the corresponding process implemented by the first network element in each method of the embodiment of this application, which will not be described again here for the sake of brevity.

[0316] Figure 14 is a schematic diagram of the chip of an embodiment of this application. The chip 1400 shown in Figure 14 comprises a processor 1410, which is configured to call and execute a computer program from memory to realize the method of the embodiment of this application.

[0317] Selectively, as shown in Figure 14, the chip 1400 may further include a memory 1420, where the processor 1410 is configured to call and execute a computer program from the memory 1420 to implement the method in the embodiment of this application.

[0318] Here, the memory 1420 may be a standalone device independent of the processor 1410, or it may be integrated into the processor 1410.

[0319] Selectively, the chip 1400 may further include an input interface 1430, which the processor 1410 can control to communicate with other devices or chips, specifically to acquire information or data transmitted by other devices or chips.

[0320] Selectively, the chip 1400 may further include an output interface 1440, where the processor 1410 can control the output interface 1440 to communicate with other devices or chips, specifically, to output information or data to other devices or chips.

[0321] Selectively, the chip may be applied to the first node in the embodiments of this application, and the chip can implement the corresponding process implemented by the transmitting equipment in each method of the embodiments of this application, which will not be described again here for the sake of brevity.

[0322] Selectively, the chip may be applied to the second node in the embodiments of this application, which can implement the corresponding processes implemented by the transmitting equipment in each method of the embodiments of this application, which will not be described again here for the sake of brevity.

[0323] Selectively, the chip may be applied to the first network element in the embodiments of this application, and the chip can implement the corresponding processes implemented by the transmitting equipment in each method of the embodiments of this application, which will not be described again here for the sake of brevity.

[0324] It should be understood that the chips relating to the embodiments of this application may also be called system-level chips, system chips, chip systems, or on-chip system chips, etc.

[0325] Embodiments of this application provide a computer storage medium for storing one or more programs that are executed by one or more processors.

[0326] Figure 15 is a schematic block diagram of a communication system 1500 according to an embodiment of the present application. As shown in Figure 15, the communication system 1500 comprises a first node 1510, a second node 1520, and a first network element 1530.

[0327] Here, the first node 1510 can be used to implement the corresponding function realized by the first node in the above method, the second node 1520 can be used to implement the corresponding function realized by the second node in the above method, and the first network element 1530 can be used to implement the corresponding function realized by the first network element in the above method, which will not be explained again here for the sake of brevity.

[0328] It should be understood that the processor in the embodiments of this application may be an integrated circuit chip having signal processing capabilities. In the implementation process, each step of the embodiments of the above method may be completed via hardware integrated logic circuits within the processor or via instructions in the form of software. The above processor may be a general-purpose processor, a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA), or other programmable logic devices, discrete gates, or transistor logic devices, discrete hardware components, etc. Each method, step and logic block diagram disclosed in the embodiments of this application can be implemented or executed. The general-purpose processor may be a microprocessor, or the processor may be any conventional processor, etc. The steps of the methods disclosed in the embodiments of this application may be directly implemented as being executed and completed by a hardware decoding processor, or by a combination of hardware and software modules within a decoding processor. The software module may reside in a mature storage medium in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, or electrically erasable programmable memory, or registers. The storage medium is located in memory, and the processor reads the information in memory and combines it with its hardware to complete the steps of the method described above.

[0329] Understandably, the memory in the embodiments of this application may be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. Here, non-volatile memory may be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (Erasable PROM, EPROM), electrically erasable programmable read-only memory (Electrically EPROM, EEPROM), or flash memory. Volatile memory may be random access memory (RAM) used as an external cache. By illustrative rather than limiting description, many forms of RAM are available, such as static random access memory (Static RAM, SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, ESDRAM), synchronous-connected dynamic random access memory (Synchlink DRAM, SLDRAM), and direct memory bus random access memory (Direct Rambus RAM, DR RAM). It should be noted that the memory of the systems and methods described herein includes, but is not limited to, these and any other suitable types of memory.

[0330] It should be understood that the above-mentioned memories are illustrative but not limiting. For example, the memories in the embodiments of this application may include static random access memory (static RAM, SRAM), dynamic random access memory (dynamic RAM, DRAM), synchronous dynamic random access memory (synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (double data rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (enhanced SDRAM, ESDRAM), synchronously connected dynamic random access memory (synch-link DRAM, SLDRAM), and direct memory bus random access memory (Direct Rambus RAM, DR RAM). In other words, the memories in the embodiments of this application are intended to include, but not be limited to, these and any other suitable types of memory.

[0331] Embodiments of this application further provide a computer-readable storage medium configured to store computer programs.

[0332] Selectively, the computer-readable storage medium may be applied to the first node in the embodiments of this application, and the computer program causes the computer to execute the corresponding process implemented by the first node in each method of the embodiments of this application, which for the sake of brevity will not be repeated here.

[0333] Selectively, the computer-readable storage medium may be applied to the second node in the embodiments of this application, and the computer program causes the computer to execute the corresponding process implemented by the second node in each method of the embodiments of this application, which for the sake of brevity will not be repeated here.

[0334] Selectively, the computer-readable storage medium may be applied to the first network element in the embodiments of this application, and the computer program causes the computer to execute the corresponding process implemented by the first network in each method of the embodiments of this application, which for the sake of brevity will not be repeated here.

[0335] Embodiments of this application further provide computer program products including computer program instructions.

[0336] Selectively, the computer program product may be applied to the first node in the embodiments of this application, and the computer program instructions cause the computer to execute the corresponding process implemented by the first node in each method of the embodiments of this application, which for the sake of brevity will not be repeated here.

[0337] Selectively, the computer program product may be applied to the second node in the embodiments of this application, and the computer program instructions cause the computer to execute the corresponding process implemented by the second node in each method of the embodiments of this application, which for the sake of brevity will not be repeated here.

[0338] Selectively, the computer program product may be applied to the first network element in the embodiments of this application, and the computer program instructions cause the computer to execute the corresponding processes implemented by the first network element in each method of the embodiments of this application, which will not be described again here for the sake of brevity.

[0339] Embodiments of this application further provide computer programs.

[0340] Selectively, the computer program may be applied to the first node in the embodiments of this application, and when the computer program is executed on the computer, it causes the computer to execute the corresponding process implemented by the first node in each method of the embodiments of this application, which for the sake of brevity will not be repeated here.

[0341] Selectively, the computer program may be applied to the second node in the embodiments of this application, and when the computer program is executed on the computer, it causes the computer to execute the corresponding process implemented by the second node in each method of the embodiments of this application, which for the sake of brevity will not be repeated here.

[0342] Selectively, the computer program may be applied to the first network element in the embodiments of this application, and when the computer program is executed on the computer, it causes the computer to execute the corresponding process implemented by the first network element in each method of the embodiments of this application, which for the sake of brevity will not be repeated here.

[0343] As will be obvious to those skilled in the art, the units and algorithmic steps of each example described with reference to the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware form or software form will depend on the specific application and design constraints of the technical solution. Skilled technicians may implement the described functions using different methods depending on each specific application, but such implementations should not be considered beyond the scope of this application.

[0344] Those skilled in the art will understand this clearly, but for the sake of convenience and brevity, the specific working processes of the systems, apparatus, and units described above can be referenced to the corresponding processes in the embodiments of the methods described above and will not be repeated here.

[0345] In some embodiments provided in this application, it should be understood that the disclosed systems, apparatus and methods can be implemented in other ways. For example, the embodiments of the apparatus described above are illustrative only, and for example, the division of the units is only a logical functional division, and other divisional modes may be possible in actual implementation, and for example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not performed. Also, the mutual coupling, direct coupling or communication connection shown or discussed may be an indirect coupling or communication connection via some communication interface, apparatus or unit, and may be in an electrical, mechanical or other form.

[0346] The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed among multiple network units. Depending on the actual needs, some or all of these units can be selected to achieve the objectives of the solution of this embodiment.

[0347] Furthermore, each functional unit in each embodiment of this application may be integrated into a single processing unit, the individual units may exist physically independently, and two or more units may be integrated into a single unit.

[0348] If the aforementioned functions are implemented in the form of a software function unit and sold or used as an independent product, they may be stored on a computer-readable storage medium. Based on this understanding, the technical solutions of this application may be essentially, or contribute to the prior art, or a part of such technical solutions may be embodied in the form of a software product, which is stored on a storage medium and includes several instructions for causing a computer device (which may be a personal computer, server, or network device, etc.) to perform all or part of the steps of the method described in each embodiment of this application. The storage medium includes various media capable of storing program code, such as USB flash disks, mobile hard disks, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0349] The above descriptions are merely specific embodiments of the present application, and the scope of protection of this application is not limited thereto. Any person skilled in the art will readily conceive of any variations or substitutions within the technical scope disclosed herein, and all such variations or substitutions should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be subject to the scope of protection of the claims.

Claims

1. A communication method applicable to the first node, Receiving first information from a first network element, wherein the first information is used to indicate at least one second node that satisfies the conditions for performing a VFL task, This includes determining from the at least one second node which will participate in the VFL task, Communication method.

2. Whether the second node satisfies the conditions for executing the VFL task is determined by the first network element based on the capability information of the second node. The capability information of the aforementioned second node is: Whether the second node supports the execution of VFL tasks, During the period during which the second node supports the execution of the VFL task, Whether the second node can function as the initiator of the VFL task, The computing power of the second node, and Any of the service areas of the second node mentioned above is used to indicate one of the terms, The communication method according to claim 1.

3. The parameter information of at least one second node is collected or determined by the first network element. For each second node, the parameter information of the second node is as follows: Capability information of the aforementioned second node, Identifier of the second node, The address information of the second node, and Includes at least one of the identifiers of the application corresponding to the second node, The communication method according to claim 2.

4. For each second node, the parameter information of the second node is included in the second registration request information, and the second registration request information is the registration request information that the second node transmits to the first network element. The communication method according to claim 3.

5. Prior to receiving the first information from the first network element, the communication method further includes transmitting parameter information of the first node to the first network element. The parameter information of the first node is: Capability information of the first node, Identifier of the first node, The address information of the first node, and Includes at least one of the application identifiers corresponding to the first node, The capability information of the aforementioned first node is: Whether the first node supports the execution of VFL tasks, During the period during which the first node supports the execution of the VFL task, Whether the first node can function as the initiator of the VFL task, The computing power of the first node, and Used to indicate at least one of the service areas of the first node, The communication method according to any one of claims 1 to 4.

6. The first node is an application function (AF), and transmitting the parameter information of the first node to the first network element includes transmitting the parameter information of the first event to the first network element by a network exposure function (NEF). The parameter information for the first event is generated based on the parameter information for the first node. The parameter information for the first event is: Whether the AF supports the execution of the first event, During the period during which the aforementioned AF supports the execution of the first event, Whether the aforementioned AF can function as the initiator of the first event, The computing power of the aforementioned AF, and The communication method according to claim 5, used to indicate at least one of the identifiers of an application corresponding to the AF, wherein the first event is associated with a VFL task.

7. The parameter information of the first node is included in the first registration request information, and the first registration request information is the registration request information that the first node transmits to the first network element. The communication method according to claim 5 or 6.

8. Prior to receiving first information from the first network element, the communication method further includes transmitting second information to the first network element, the second information being used to request the first information. The communication method according to any one of claims 1 to 7.

9. The first node is an AF, and transmitting the second information to the first network element is: The communication method according to claim 8, comprising transmitting the second information to the first network element by NEF.

10. The second information mentioned above is, Execution of VFL task, During the period in which the VFL task is executed, To act as the initiator of the VFL task, Computing power to run VFL tasks, and The communication method according to claim 8 or 9, used to indicate at least one of the service areas on which a VFL task is performed.

11. Determining at least one participating node to participate in the VFL task from the at least one second node is: A communication method according to any one of claims 1 to 10, comprising determining a second node that agrees to participate in the VFL task as a participating node to participate in the VFL task.

12. The aforementioned communication method is, The communication method according to any one of claims 1 to 11, further comprising transmitting data requirements information to at least one second node, the data requirements information being used to determine the data required to perform the VFL task, and the data requirements information including the type of data required to perform the VFL task and / or the validity period of the required data.

13. The communication method further includes receiving third information from at least one second node, the third information being used to indicate whether the second node agrees to participate in the VFL task. The communication method according to any one of claims 1 to 12.

14. The first node is AF, and transmitting data requirement information to the at least one second node includes transmitting the data requirement information to the at least one second node by NEF. The communication method according to claim 12 or 13.

15. The communication method further includes transmitting a VFL model corresponding to the at least one participating node, the VFL model being used to perform the VFL task, The communication method according to any one of claims 1 to 14, wherein the VFL model includes identifiers and / or address information of the corresponding participating node.

16. The first node is an AF, the participating node is a terminal device or an access network network element, and transmitting the corresponding VFL model to the participating node is: This includes transmitting a VFL model corresponding to the terminal device or the access network element via the second network element. The communication method according to claim 15.

17. The communication method further includes transmitting an association identifier to the at least one participating node, the association identifier being used to aggregate different execution results of the same VFL task. The communication method according to any one of claims 1 to 16.

18. The communication method further includes receiving a first execution result corresponding to the VFL task, The first execution result includes a second execution result and / or at least one aggregate result in which at least one participating node performs the VFL task, wherein each of the at least one aggregate result is obtained by aggregating based on at least two second execution results. The communication method according to any one of claims 1 to 17.

19. Each second execution result and / or each aggregate result among the at least one aggregate result includes the same association identifier. The communication method according to claim 18.

20. Each of the at least one aggregate result is obtained by aggregating based on the same association identifier. The communication method according to claim 18 or 19.

21. The communication method further includes aggregating the execution result corresponding to the VFL task having the same association identifier and the execution result of the first node executing the VFL task. A communication method according to any one of claims 18 to 20.

22. The first node is a terminal device, an access network element, a core network element, or an AF. A communication method according to any one of claims 1 to 5, 7 to 8, 10 to 13, 15, or 17 to 21.

23. The second node includes at least one of terminal equipment, access network network elements, core network elements, and AF. A communication method according to any one of claims 1 to 22.

24. The first network element is a network memory function (NRF). A communication method according to any one of claims 1 to 23.

25. A communication method applicable to the second node, The communication method includes transmitting capability information of the second node to a first network element, and the capability information of the second node is used to determine whether the second node satisfies the conditions for performing a vertically federated learning VFL task. Communication method.

26. The capability information of the aforementioned second node is: Whether the second node supports the execution of VFL tasks, During the period during which the second node supports the execution of the VFL task, Whether the second node can function as the initiator of the VFL task, The computing power of the second node, and Any of the service area information of the second node is used to indicate one item, The communication method according to claim 25.

27. The aforementioned communication method is, The process further includes transmitting parameter information of the second node to the first network element, wherein the parameter information of the second node is Capability information of the aforementioned second node, Identifier of the second node, The address information of the second node, and Includes at least one of the identifiers of the application corresponding to the second node, The communication method according to claim 25 or 26.

28. The second node is an application function (AF), and transmitting the parameter information of the second node to the first network element includes transmitting the parameter information of the second event to the first network element by a network exposure function (NEF). The parameter information for the second event is generated based on the parameter information for the second node. The parameter information for the second event is: Whether the aforementioned AF supports the execution of the second event, During the period during which the aforementioned AF supports the execution of the second event, Whether the aforementioned AF can function as the initiator of the second event, The computing power of the aforementioned AF, and Used to indicate at least one of the application identifiers corresponding to the aforementioned AF, The aforementioned second event is associated with the VFL task. The communication method according to claim 27.

29. The parameter information of the second node is included in the second registration request information, and the second registration request information is the registration request information that the second node transmits to the first network element. The communication method according to claim 27 or 28.

30. The communication method further includes receiving data requirement information from the first node, and the data requirement information is used to determine the data for performing the VFL task. The aforementioned data requirements information includes the type of data required to perform the VFL task, and / or the validity period of the required data. The communication method according to any one of claims 25 to 29.

31. The communication method further includes transmitting third information to the first node, the third information being used to indicate whether the second node agrees to participate in the VFL task. A communication method according to any one of claims 25 to 30.

32. The first node is AF, and receiving data requirement information from the first node includes receiving the data requirement information from the first node via NEF. The communication method according to claim 30 or 31.

33. The communication method further includes receiving a VFL model from the first node, and the VFL model is used to perform the VFL task. The VFL model corresponds to the second node, and the VFL model includes identifier and / or address information of the second node. A communication method according to any one of claims 25 to 32.

34. The communication method further includes receiving an association identifier from the first node, the association identifier being used to aggregate different execution results of the same VFL task. A communication method according to any one of claims 25 to 33.

35. The communication method further includes transmitting to the first node the execution result of the VFL task performed by the second node, the execution result including the association identifier. The communication method according to claim 34.

36. The aforementioned communication method is, At least one participating node receives a second execution result from executing the VFL task, The process includes sending the aggregate result to the first node, wherein the aggregate result is obtained by aggregating at least one second execution result and a third execution result obtained when the second node performs the VFL task, The aggregate result includes the association identifier, The communication method according to claim 34.

37. Each of the second execution results and the third execution result has the same association identifier. The aggregated result is obtained by aggregating based on the same association identifier. The communication method according to claim 36.

38. The first node is a terminal device, an access network element, a core network element, or an AF. A communication method according to any one of claims 25 to 31 or 33 to 37.

39. The second node is a terminal device, an access network element, a core network element, or an AF. A communication method according to any one of claims 25 to 27 or 29 to 38.

40. The first network element is a network memory function (NRF). A communication method according to any one of claims 25 to 39.

41. A method applied to a first network element, wherein the communication method is This includes obtaining capability information for at least one second node, the capability information for the second node being used to determine whether the second node satisfies the conditions for performing a vertically federated learning VFL task. Communication method.

42. The capability information of the aforementioned second node is: Whether the second node supports the execution of VFL tasks, During the period during which the second node supports the execution of the VFL task, Whether the second node can function as the initiator of the VFL task, The computing power of the second node, and Any of the service area information of the second node is used to indicate one item, The communication method according to claim 41.

43. Before acquiring capability information for at least one second node, the communication method, The process further includes receiving parameter information of at least one second node, and for each second node, the parameter information of the second node is: Capability information of the aforementioned second node, Identifier of the second node, The address information of the second node, and Includes at least one of the identifiers of the application corresponding to the second node, The communication method according to claim 41 or 42.

44. For each second node, the parameter information of the second node is included in the second registration request information, and the second registration request information is the registration request information that the second node transmits to the first network element. The communication method according to claim 43.

45. The communication method further includes receiving parameter information of the first node, The parameter information of the first node is: Capability information of the first node, Identifier of the first node, The address information of the first node, and Includes at least one of the application identifiers corresponding to the first node, The capability information of the aforementioned first node is: Whether the first node supports the execution of VFL tasks, During the period during which the first node supports the execution of the VFL task, Whether the first node can function as the initiator of the VFL task, The computing power of the first node, and Used to indicate at least one of the service areas of the first node, The communication method according to any one of claims 41 to 44.

46. The parameter information of the first node is included in the first registration request information, and the first registration request information is the registration request information that the first node transmits to the first network element. The communication method according to claim 45.

47. The communication method further includes transmitting first information to the first node, the first information being used to indicate at least one second node that satisfies the conditions for performing the VFL task. A communication method according to any one of claims 40 to 46.

48. Before transmitting the first information to the first node, the communication method, The process further includes receiving second information from the first node, the second information being used to request the first information. The communication method according to claim 47.

49. The first node is a terminal device, an access network network element, a core network element, or an application function (AF). The communication method according to any one of claims 41 to 48.

50. The second node includes at least one of terminal equipment, access network network elements, core network elements, and AF. The communication method according to any one of claims 41 to 49.

51. The first network element is a network memory function (NRF). A communication method according to any one of claims 41 to 50.

52. A first receiving module configured to receive first information from a first network element, The system comprises a decision module configured to determine at least one participating node to participate in the VFL task from the at least one second node, wherein the first information is used to indicate at least one second node that satisfies the conditions for executing the VFL task. Communication device.

53. The system includes a first transmission module configured to transmit capability information of the device to a first network element, the capability information of the device being used to determine whether the device satisfies the conditions for performing a vertically federated learning VFL task. Communication device.

54. The system includes an acquisition module configured to acquire capability information of at least one second node, the capability information of the second node being used to determine whether the second node satisfies the conditions for performing a vertically federated learning VFL task. Communication device.

55. It includes memory and a processor, the memory being configured to store computer-executable instructions, The processor is connected to the memory and is configured to implement the communication method described in any one of claims 1 to 24, or the communication method described in any one of claims 25 to 40, or the communication method described in any one of claims 41 to 51, by executing the computer-executable instructions. Communication equipment.

56. A chip including a processor, wherein the processor calls and executes a computer program from memory to cause a device on which the chip is installed to execute the communication method described in any one of claims 1 to 24, or the communication method described in any one of claims 25 to 40, or the communication method described in any one of claims 41 to 51. Tip.

57. At least one processor is made to execute the communication method described in any one of claims 1 to 24, or the communication method described in any one of claims 25 to 40, or the communication method described in any one of claims 41 to 51. A computer-readable storage medium on which computer programs are stored.

Images