CN116866882A

CN116866882A - Candidate member determination method, device and equipment

Info

Publication number: CN116866882A
Application number: CN202210476433.9A
Authority: CN
Inventors: 程思涵; 崇卫微
Original assignee: Vivo Mobile Communication Co Ltd
Current assignee: Vivo Mobile Communication Co Ltd
Priority date: 2022-03-28
Filing date: 2022-04-29
Publication date: 2023-10-10

Abstract

The application discloses a method, a device and equipment for determining candidate members, which belong to the technical field of communication, and the method for determining candidate members in the embodiment of the application comprises the following steps: the method comprises the steps that a first network element receives a request message sent by a second network element, wherein the request message comprises screening information; the first network element determines one or more first devices as candidate members capable of participating in federal learning according to the screening information; the first network element sends a response message to the second network element, wherein the response message comprises the identification of the candidate member, and the screening information comprises at least one of the following: time period, algorithm type, accuracy threshold, wireless access system, signal quality requirement, flow range, member type information, quantity information, region information and federal learning type information. The embodiment of the application can participate in federal learning by screening out candidate members conforming to screening information, and can improve the efficiency of federal learning.

Description

Candidate member determination method, device and equipment

Technical Field

The present application belongs to the field of communication technology, and in particular, relates to a method, an apparatus and a device for determining candidate members

Background

Federal learning (federated learning) refers to a method of machine learning modeling by joining different participants, or party, also known as data owners, or clients. In federal learning, participants do not need to expose their own data to other participants and coordinators (also called servers, parameter servers, or aggregation servers (aggregation server)), so federal learning can well protect user privacy and data security, and can solve the problem of data islanding.

After federal learning is applied to the communication field, how to select some proper members to perform federal learning to improve training efficiency is a problem to be solved at present.

Disclosure of Invention

The embodiment of the application provides a method, a device and equipment for determining candidate members, which can solve the problem of low training efficiency of federal learning due to the fact that participating members of federal learning are not in compliance with requirements.

In a first aspect, a method for determining candidate members is provided, including:

the method comprises the steps that a first network element receives a request message sent by a second network element, wherein the request message comprises screening information;

The first network element determines one or more first devices as candidate members capable of participating in federal learning according to the screening information;

the first network element sends a response message to the second network element, wherein the response message comprises the identification of the candidate member;

wherein the screening information includes at least one of:

a time period for indicating an optional time period for performing the federal learning;

the algorithm type is used for indicating the algorithm type of model training which needs to be supported by the federal learning;

the precision threshold is used for indicating the precision requirement of model training to be met in the federal learning;

the wireless access system is used for indicating the wireless access system to be selected for the federal learning;

signal quality requirements for indicating wireless signal quality requirements when performing the federal learning;

a flow range for indicating a flow usage range requirement of the candidate member;

member type information for indicating type requirements of candidate members participating in federal learning;

quantity information indicating a quantity requirement of the candidate member;

the region of interest AOI is used for indicating the region where the candidate member is located;

And the type information of the federal study is used for indicating that the federal study belongs to the longitudinal federation or the transverse federation.

In a second aspect, a method for determining candidate members is provided, including:

the second network element sends a request message to the first network element, wherein the request message comprises screening information, and the request message is used for indicating the first network element to determine one or more first devices as candidate members capable of participating in federal learning according to the screening information;

the second network element receives a response message sent by the first network element, wherein the response message comprises the identification of the candidate member;

wherein the screening information includes at least one of:

quantity information indicating a quantity requirement of the candidate member;

In a third aspect, there is provided a candidate member determination apparatus comprising:

the receiving module is used for receiving a request message sent by the second network element, wherein the request message comprises screening information;

a processing module for determining one or more first devices as candidate members capable of participating in federal learning according to the screening information;

a sending module, configured to send a response message to the second network element, where the response message includes an identifier of the candidate member;

wherein the screening information includes at least one of:

quantity information indicating a quantity requirement of the candidate member;

In a fourth aspect, there is provided a candidate member determination apparatus comprising:

a sending module, configured to send a request message to a first network element, where the request message includes screening information, and the request message is configured to instruct the first network element to determine, according to the screening information, one or more first devices as candidate members that can participate in federal learning;

a receiving module, configured to receive a response message sent by the first network element, where the response message includes an identifier of the candidate member;

Wherein the screening information includes at least one of:

quantity information indicating a quantity requirement of the candidate member;

In a fifth aspect, there is provided a first network element comprising a processor and a memory storing a program or instructions executable on said processor, said program or instructions implementing the steps of the method according to the first aspect when executed by said processor.

In a sixth aspect, a first network element is provided, including a processor and a communication interface, where the processor is configured to determine, according to screening information, one or more first devices as candidate members that can participate in federal learning, and the communication interface is configured to receive a request message sent by a second network element, send a response message to the second network element, where the response message includes an identifier of the candidate member, and the request message includes screening information, where the screening information includes at least one of:

Quantity information indicating a quantity requirement of the candidate member;

In a seventh aspect, there is provided a second network element comprising a processor and a memory storing a program or instructions executable on said processor, said program or instructions implementing the steps of the method according to the second aspect when executed by said processor.

An eighth aspect provides a second network element, including a processor and a communication interface, where the communication interface is configured to send a request message to a first network element, where the request message includes screening information, and the request message is configured to instruct the first network element to determine, according to the screening information, one or more first devices as candidate members that can participate in federal learning; receiving a response message sent by the first network element, wherein the response message comprises the identification of the candidate member;

wherein the screening information includes at least one of:

quantity information indicating a quantity requirement of the candidate member;

In a ninth aspect, a system for determining candidate members is provided, including: a first network element operable to perform the steps of the method of determining a candidate member as described in the first aspect, and a second network element operable to perform the steps of the method of determining a candidate member as described in the second aspect.

In a tenth aspect, there is provided a readable storage medium having stored thereon a program or instructions which when executed by a processor, performs the steps of the method according to the first aspect or performs the steps of the method according to the second aspect.

In an eleventh aspect, there is provided a chip comprising a processor and a communication interface coupled to the processor, the processor being for running a program or instructions to implement the method according to the first aspect or to implement the method according to the second aspect.

In a twelfth aspect, there is provided a computer program/program product stored in a storage medium, the computer program/program product being executed by at least one processor to implement the steps of the method of determining candidate members according to the first and second aspects.

In the embodiment of the application, the first network element receives the request message sent by the second network element, determines one or more first devices as candidate members capable of participating in federal learning according to screening information included in the request message, and sends identification information of the candidate members to the second network element, wherein the screening information comprises at least one of the following: time period, algorithm type, accuracy threshold, wireless access mode, signal quality requirement, flow range, member type information, quantity information, region of interest AOI and federal learning type. The first network element can screen out the participators suitable for federal learning through screening information, so that the training efficiency of federal learning is improved.

Drawings

Fig. 1 shows a block diagram of a wireless communication system to which embodiments of the present application are applicable;

FIG. 2 is a flow chart of a method for determining candidate members according to the present application;

FIG. 3 is a flow chart of another method for determining candidate members according to the present application;

FIG. 4 is a signaling diagram of a method for determining candidate members according to the present application;

FIG. 5 is a schematic diagram of a candidate member determination device according to one embodiment of the present application;

FIG. 6 is a second schematic diagram of a candidate member determination device according to the present application;

fig. 7 is a schematic structural diagram of a communication device according to an embodiment of the present application;

fig. 8 is a schematic hardware structure of a first network element according to an embodiment of the present application;

fig. 9 is a schematic hardware structure of a second network element according to an embodiment of the present application.

Detailed Description

The technical solutions of the embodiments of the present application will be clearly described below with reference to the drawings in the embodiments of the present application, and it is apparent that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which are derived by a person skilled in the art based on the embodiments of the application, fall within the scope of protection of the application.

The terms first, second and the like in the description and in the claims, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the terms so used are interchangeable under appropriate circumstances such that the embodiments of the application are capable of operation in sequences other than those illustrated or otherwise described herein, and that the "first" and "second" distinguishing between objects generally are not limited in number to the extent that the first object may, for example, be one or more. Furthermore, in the description and claims, "and/or" means at least one of the connected objects, and the character "/" generally means a relationship in which the associated object is an "or" before and after.

It should be noted that the techniques described in the embodiments of the present application are not limited to long term evolution (Long Term Evolution, LTE)/LTE evolution (LTE-Advanced, LTE-a) systems, but may also be used in other wireless communication systems, such as code division multiple access (Code Division Multiple Access, CDMA), time division multiple access (Time Division Multiple Access, TDMA), frequency division multiple access (Frequency Division Multiple Access, FDMA), orthogonal frequency division multiple access (Orthogonal Frequency Division Multiple Access, OFDMA), single carrier frequency division multiple access (Single-carrier Frequency Division Multiple Access, SC-FDMA), and other systems. The terms "system" and "network" in embodiments of the application are often used interchangeably, and the techniques described may be used for both the above-mentioned systems and radio technologies, as well as other systems and radio technologies. The following description describes a New air interface (NR) system for purposes of example and uses NR terminology in much of the description that follows, but these techniques are also applicable to applications other than NR system applications, such as generation 6 (6) ^th Generation, 6G) communication system.

Fig. 1 shows a block diagram of a wireless communication system to which an embodiment of the present application is applicable. The wireless communication system includes a terminal 11 and a network device 12. The terminal 11 may also be a User Equipment (UE) in the present application, where the terminal 11 may be a mobile phone, a tablet Computer (Tablet Personal Computer), a Laptop (Laptop Computer) or a terminal-side Device called a notebook (Personal Digital Assistant, PDA), a palm Computer, a netbook, an ultra-mobile personal Computer (ultra-mobile personal Computer, UMPC), a mobile internet Device (Mobile Internet Device, MID), an augmented reality (augmented reality, AR)/Virtual Reality (VR) Device, a robot, a Wearable Device (weather), a vehicle-mounted Device (VUE), a pedestrian terminal (PUE), a smart home (home Device with a wireless communication function, such as a refrigerator, a television, a washing machine, furniture, etc.), a game console, a personal Computer (personal Computer, PC), a teller machine, a self-service machine, etc., and the Wearable Device includes: intelligent wrist-watch, intelligent bracelet, intelligent earphone, intelligent glasses, intelligent ornament (intelligent bracelet, intelligent ring, intelligent necklace, intelligent anklet, intelligent foot chain etc.), intelligent wrist strap, intelligent clothing etc.. It should be noted that the specific type of the terminal 11 is not limited in the embodiment of the present application. The network-side device 12 may comprise an access network device or a core network device, wherein the access network device 12 may also be referred to as a radio access network device, a radio access network (Radio Access Network, RAN), a radio access network function or a radio access network element. Access network device 12 may include a base station, a WLAN access point, a WiFi node, or the like, which may be referred to as a node B, an evolved node B (eNB), an access point, a base transceiver station (Base Transceiver Station, BTS), a radio base station, a radio transceiver, a basic service set (Basic Service Set, BSS), an extended service set (Extended Service Set, ESS), a home node B, a home evolved node B, a transmission and reception point (Transmitting Receiving Point, TRP), or some other suitable terminology in the art, and the base station is not limited to a particular technical vocabulary so long as the same technical effect is achieved, and it should be noted that in the embodiment of the present application, only a base station in the NR system is described as an example, and the specific type of the base station is not limited. The core network device may include, but is not limited to, at least one of: core network nodes, core network functions, mobility management entities (Mobility Management Entity, MME), access mobility management functions (Access and Mobility Management Function, AMF), session management functions (Session Management Function, SMF), user plane functions (User Plane Function, UPF), policy control functions (Policy Control Function, PCF), policy and charging rules function units (Policy and Charging Rules Function, PCRF), edge application service discovery functions (Edge Application Server Discovery Function, EASDF), unified data management (Unified Data Management, UDM), unified data repository (Unified Data Repository, UDR), home subscriber server (Home Subscriber Server, HSS), centralized network configuration (Centralized network configuration, CNC), network storage functions (Network Repository Function, NRF), network opening functions (Network Exposure Function, NEF), local NEF (or L-NEF), binding support functions (Binding Support Function, BSF), application functions (Application Function, AF), and the like. It should be noted that, in the embodiment of the present application, only the core network device in the NR system is described as an example, and the specific type of the core network device is not limited.

The method for determining the candidate member provided by the embodiment of the application is described in detail below through some embodiments and application scenes thereof with reference to the accompanying drawings.

Federal learning includes both horizontal federal learning and vertical federal learning, wherein the nature of horizontal federal learning is a combination of samples that is applicable to scenarios where the inter-participant performance is the same but the touching customers are different, i.e., there is a large overlap in features, and there is a small overlap in samples, such as Core Network (CN) domain and radio access network (Radio Access Network, RAN) domain within a communication network serving the same service (e.g., mobility management (mobility management, MM), session management (session management, SM) service, a certain service) of different users (e.g., each terminal, i.e., sample is different). By combining the same data characteristics of different samples of the participants, the transverse federation increases the number of training samples, thereby yielding a better model.

The essence of longitudinal federal learning is feature combination, which is suitable for the scene that the sample overlap is more and the feature overlap is less, such as different services (e.g. MM, SM service, i.e. features are different) of the same user (e.g. terminal, i.e. sample is the same) served by the CN domain and the RAN domain in the communication network. By combining different data features of the common sample of the participants, the longitudinal federation increases the feature dimension of the training sample and results in a better model.

In the federal learning system, the system comprises a manager (server) and a plurality of participants, wherein the manager is used for sending a model to each participant, updating the model according to the feedback result of each participant, and sending the updated model to each participant again for the next model training. Each participant has respective data, and in order not to send local data to other people, each participant uses a model sent by a manager to train locally and returns the model to the manager for model updating.

By way of example, the process of model training may include the steps of:

step 1: the participants each download the latest model from the manager (server);

step 2: each participant trains a model by using local data, encrypts gradients and uploads the gradients to a manager (server), and the manager (server) aggregates gradient update model parameters of each participant;

step 3: the manager (server) returns the updated model to each participant;

step 4: each participant updates its own model.

And (3) performing multiple iterations/model updating in the steps 2 to 4, and completing model training under certain conditions (for example, after the number of iterations for a certain number of times is completed or the calculated value of the loss function of the model is lower than a preset value).

A network data analysis function network element (network data analytic function, NWDAF) is introduced in 5 GS. The NWDAF may collect data from each network element, network management system, etc. of the core network to perform big data statistics, analysis, or intelligent data analysis, to obtain analysis or prediction data at the network side, so as to assist each network element to more effectively control terminal access according to the data analysis result.

In the method for determining the candidate members, the data of other network elements can be collected through the NWDAF and analyzed, so that the participated members suitable for federal learning are screened out according to screening information and data analysis results, and the federal learning efficiency is improved.

In the present application, "determination of candidate members" may be understood as "determination of candidate members that can participate in federal learning".

Fig. 2 is a flow chart of a method for determining candidate members provided by the present application, and the method for determining candidate members provided by the embodiment of the present application is described below with reference to fig. 2. As shown in fig. 2, the method includes:

step 201: the first network element receives the request message sent by the second network element.

The execution main body of the candidate member determination method provided by the embodiment of the application is a first network element, and the first network element can be implemented in various forms. For example, the first network element described in the embodiment of the present application may include NWDAF, and of course, may be other network elements capable of collecting data from each network element, network management system, etc. of the core network to perform big data statistics, analysis, or intelligent data analysis. The second network element may comprise a task consumer network element, which may be, for example, an application function network element (application function, AF), a base station, a terminal, etc., or a device, which may also be a third party server.

Wherein, the request message includes screening information, and the screening information includes at least one of the following: time period, algorithm type, accuracy threshold, wireless access system, signal quality requirement, flow range, member type information, quantity information, region information and federal learning type information. Based on screening information of different dimensions, more proper participating members can be screened, so that the efficiency of federal learning is improved.

The time period is used for indicating an optional time period for performing the federal learning, and the time period may be a certain time period in the past or may be a future time period, and it is understood that if the time period is a future time period, the first network element predicts the networking state of each member in the future time period according to the acquired historical data. The networking state includes a state of a connected network and a state of an unconnected network. It should be appreciated that the time period is typically flexibly set according to the time period during which federal learning is actually performed.

The algorithm type is used for indicating the algorithm type of model training required to be supported by the federal learning, and comprises the algorithm type supported by each member and relevant to the artificial intelligence (artificial intelligence, AI) data analysis task such as 'deep learning algorithm', 'linear regression algorithm', and the like. It should be appreciated that members supporting the type of algorithm matching the function will typically be selected as participating members for federal learning based on the function of the model.

The precision threshold is used for indicating the precision requirement of model training to be met in the federal learning, and comprises precision values which can be achieved by models generated by each member after training, such as the accuracy of the models, and can be understood as the percentage of model prediction or judgment accuracy. It should be appreciated that in order to ensure accuracy of federal learning, some members with higher accuracy are typically selected as participating members of federal learning.

The wireless access system is used for indicating the wireless access system to be selected for federal learning, and comprises modes of connecting or accessing each member to a communication network, such as a WLAN (similar to WiFi) connected with non-3GPP, 5g and 4g networks connected with 3GPP, and the like. It will be appreciated that in order to ensure stability of federal learning, and to reduce traffic consumption, some members connected to WLAN or WiFi are typically selected as participating members of federal learning.

And the signal quality requirement is used for indicating the wireless signal quality requirement when the federal learning is performed. The signal quality requirements may include a threshold of network signal strength and/or a threshold of stability when the respective member connects to the communication network, etc. The signal quality requirements include WLAN network signal quality requirements, may also include 5G NR network signal quality requirements, and may also include 4G long term evolution (Long Term Evolution, LTE) network signal quality requirements. The signal quality requirement is also understood to be a minimum requirement for the proportion of time that keeps the signal strength above a certain value, e.g. the signal strength can reach the threshold requirement for at least 90% of the time. It should be appreciated that in order to ensure stability of federal learning, members with higher signal quality (e.g., stronger network signal strength and better stability) are typically selected as participating members of federal learning.

Where WLAN is connected, the signal quality may be represented by the mean, variance, etc. of the received signal strength indication (Received Signal Strength Indication, RSSI) or the round-trip time (RTT) of the signal.

The traffic range is used for indicating the traffic usage range requirement of the candidate member, and is information such as the traffic value requirement consumed by each member in a preset time period, the number of the uploaded and downloaded traffic, and the like, and can be accumulated usage of the UE. It should be appreciated that in order to alleviate the stress of individual participating members in federal learning, some members are typically selected to have a stable network state and a low traffic consumption as participants in federal learning.

The member type information is used for indicating the type requirement of the candidate member participating in the federal learning, and the member type information can be understood as the type of the candidate member participating in the federal learning at this time, for example, a terminal, a core network element (such as NWDAF), a base station, or the like.

An area of interest (AOI) for indicating the area in which the candidate member is located, the AOI may be the area in which the candidate member participating in federal learning is located, or an area of interest, or an area range of interest, which may be latitude and longitude, or one or more cells/Tracking Areas (TAs), etc.

And the quantity information is used for indicating the quantity requirement of the candidate members, and can be understood as the quantity of the members participating in federal learning, namely the quantity of the candidate members required to be determined by the first network element. The first network element may determine the number of candidate network elements. The number information may be the minimum number of required candidate members, i.e. the determined number of candidate members cannot be smaller than the minimum number of required candidate members. The number information may also be the maximum number of required candidate members, i.e. the number of determined candidate members cannot be greater than the maximum number of required candidate members.

And the type information of the federal study is used for indicating that the federal study belongs to the longitudinal federation or the transverse federation. Optionally, the request message may further include indication information, where the indication information is used to indicate that a determination task of a candidate member participating in federal learning is performed, that is, the task is selected by a member for federal learning or federal learning, and after the first network element receives the request message, based on the request message or based on the indication information in the request message, the first network element may learn that the determination task of the candidate member participating in federal learning needs to be performed.

Step 202: the first network element determines one or more first devices as candidate members that can participate in federal learning based on the screening information.

In this step, after receiving the request message, the first network element performs screening from a plurality of second devices according to screening information in the request message, so as to determine one or more first devices that meet the screening information, and determine the first devices as candidate members that can participate in federal learning.

Optionally, the first device includes at least one of: a first device in a networked state for a period of time; a first device supporting federal learning in the time period, a first device supporting an algorithm type; the first equipment for federal learning by utilizing the algorithm type is supported, and the model trains the first equipment with the precision information larger than the precision threshold value; a first device supporting a wireless access system; the first equipment is in the wireless access mode; the first equipment supports federal learning under the wireless access system; a first device having a signal quality greater than a signal quality threshold; a first device having a signal quality greater than a signal quality threshold requirement; a first device whose consumed flow is in a flow range; a first device that meets the type requirements of the candidate member; a first device located within the AOI; the type of federal learning is the same as the type of federal learning included in the screening information.

When the screening information comprises a time period, the first device is a first device in a networking state in the time period, or the first device is a first device supporting federal learning in the time period; under the condition that the filtering information comprises an algorithm type, the first device is a first device supporting the algorithm type, or the first device is a first device supporting federal learning by utilizing the algorithm type; under the condition that the screening information comprises an accuracy threshold value, the first equipment trains the first equipment with the accuracy information larger than the accuracy threshold value for the model; under the condition that the screening information comprises a wireless access mode, the first equipment is first equipment supporting the wireless access mode, or the first equipment is first equipment under the wireless access mode, or the first equipment is first equipment supporting federal learning under the wireless access mode; in the case that the screening information includes a signal quality threshold, the first device is a first device whose signal quality is greater than the signal quality threshold; in the case that the screening information includes a signal quality requirement, the first device is a first device with signal quality greater than the signal quality requirement; under the condition that the screening information comprises a flow range, the first device is a first device with consumed flow in the flow range; in the case that the screening information includes member type information, the first device is a first device that satisfies the type requirement of the candidate member, and it can be understood that the type of the first device is the same as the member type information indicated in the screening information; in the case that the screening information includes an AOI, the first device is a first device located within the AOI; in the case where the screening information includes federal learning type information, the federal learning type of the first device is the same as the federal learning type information included in the screening information.

In addition, when the device type is included in the filtering information, the first device is a first device of the device type, and when the region information is included in the filtering information, the first device is a first device in the region information.

It should be understood that in the case where at least two items of screening information are included, the first device is a device that satisfies the respective at least two items of screening information. For example, in the case where the time period and the supported algorithm type are included in the filtering information, the first device is a first device that is in a networking state for the time period and supports the algorithm type. The case that the filtering information includes at least two other information is similar to the case that the filtering information includes a time period and a supported algorithm type, and will not be described herein.

In the above embodiment, the first device satisfying the screening information may be screened from the at least one second device by the screening information as a candidate network element, so that a member suitable for participating in federal learning may be selected, which is helpful for improving efficiency of federal learning training.

Optionally, when determining the candidate member, the first network element may determine a data type corresponding to the screening information, acquire attribute information corresponding to each of the at least one device from the at least one third network element based on the data type, and determine the candidate member according to the attribute information and the screening information.

Optionally, the data types include at least one of the following networking information, a corresponding time period, an algorithm type, a wireless access system, a signal quality, and a traffic.

Specifically, in the case that the screening information includes a time period, the data type corresponding to the screening information is networking information and the time period corresponding to the networking state, in the case that the screening information includes a supported algorithm type, the data type corresponding to the screening information is algorithm type, in the case that the screening information includes an accuracy threshold, the data type corresponding to the screening information is accuracy information, in the case that the screening information includes a wireless access system, the data type corresponding to the screening information is wireless access system, in the case that the screening information includes a signal quality threshold, the data type corresponding to the screening information is signal quality, in the case that the screening information includes a traffic range, the data type corresponding to the screening information is traffic, and in the case that the screening information includes a member type, the data type corresponding to the screening information is a device type. In the case where the filtering information includes area information, the data type corresponding to the filtering information is location information.

It will be appreciated that in the case where the member type is included in the screening information, the member type is different, that is, the screening information included in the request message is different, that is, the screening information is related to the member type. For example, if the member type includes a terminal, the screening information may include at least one of a time period, a supported algorithm type, an accuracy threshold, a radio access scheme and signal quality threshold, a traffic range, and area information. If the member type includes a network-side device, such as a base station or a core network device, the screening information may include at least one of a supported algorithm type, an accuracy threshold.

After determining the data type, the first network element acquires attribute information corresponding to each of at least one second device from at least one third network element based on the data type. Wherein the third network element comprises a plurality of different network elements, it is understood that the attribute information to be acquired is different, and the first network element acquires the attribute information from the different third network elements. For example: the first network element obtains information such as the time when the terminal is connected to the WLAN, whether the terminal is connected to the WLAN, etc. from the third network element session management function (session management function, SMF). The first network element obtains information such as signal quality of the WLAN connection from a third network element, which may be a network management device, for example, operation management and maintenance (operation administration and maintenance, OAM), obtains algorithms and accuracy supported by each candidate network element from a third network element unified data management function (unified data management, UDM) or from a third network element data collection application function (data collection-application function, DC-AF), obtains traffic information of the terminal from a third network element user plane function (the user plane function, UPF), or the like, or the first network element obtains UE accumulated usage information from an SMF or a charging function (Charging Function, CHF).

It is understood that in the case where the data type includes networking information and corresponding time periods, the attribute information is a current networking state corresponding to each of the at least one device, and a time in the networking state, such as whether to connect to the network, in which time period the networking state is in, and the like. In the case that the data type includes an algorithm type, the attribute information is an algorithm that can be currently supported by at least one device corresponding to the data type, such as whether a deep learning algorithm is supported, whether a linear regression algorithm is supported, and the like. In the case that the data type includes precision information, the attribute information is the currently supported precision corresponding to at least one device, such as the precision that can be achieved by the model after the model is trained by each device. In the case that the data type includes a wireless access system, the attribute information is a network system type to which at least one device is currently connected, such as WLAN, 5G network, etc. that is currently connected. In case the data type comprises signal quality, the attribute information is the signal quality of the network currently accessed by the at least one second device or the signal quality of the network accessed within a preset time period. And when the data type comprises the traffic information, the attribute information is the traffic consumed in a preset time period corresponding to each of the at least one second device. In case the data type comprises a device type, the attribute information is a type to which at least one second device corresponds, such as a terminal, a base station or a core network device, respectively. In case the data type comprises location information, the attribute information is the location where the at least one second device is located, such as the geographical location where it is currently located, or the cell or TA where it is located, etc.

After the attribute information is acquired, matching the attribute information and the screening information corresponding to each of the at least one second device, so as to determine candidate members according to the screening information, for example, the first device meeting the screening information can be determined as the candidate member.

For example, assume that the filtering information includes a device type of a terminal, an area of a cell, an a period, and a wireless access system of WiFi connected to non-3 GPP. The first network element obtains the corresponding attribute information according to the screening information, and selects the terminal which is in the cell A, is in the networking state in the time period A and is connected with the WiFi of the non-3GPP as a candidate member according to the obtained attribute information.

In this embodiment, the first network element obtains the data type corresponding to the screening information, and obtains the attribute information of each second device from at least one third network element, so that candidate members are determined based on the attribute information and the screening information, and candidate members suitable for participating in federal learning can be determined from at least one second device through preset screening information, so that the efficiency of federal learning can be improved.

In one possible implementation, the first network element may directly determine the device that satisfies the screening information as a candidate member, and default that all candidate network elements that satisfy the screening information may participate in federal learning, and the first network element may send the identification of all candidate members that satisfy the screening information to the second network element.

In another possible implementation manner, in order to screen out more suitable participants, the first network element may also screen based on willingness information of each second device when determining candidate members according to the attribute information and the screening information. The first network element may obtain first indication information from the third network element, and determine, from the second device, one or more first devices as candidate members according to the screening information and the first indication information, where the first indication information is used to indicate willingness information of each device to participate in federal learning, and the willingness information indicates whether each device is willing to participate in federal learning.

The first device is a device which satisfies screening information and is willing to participate in federal learning.

For example, each device corresponds to its own first indication information, and for a certain device, the corresponding first indication information is 0, which indicates that the device is willing to participate in federal learning, and for a case that the first indication information is 1, the device is not willing to participate in federal learning, or may indicate that the device is willing to participate in federal learning, and for a case that the first indication information is 1, the device is not willing to participate in federal learning. Of course, the first indication information may also use other values to represent willingness information of each device to participate in federal learning.

After the first network element acquires the first indication information, the attribute information of each device is matched with the screening information, and the device which satisfies the screening information and indicates the willingness to participate in federal learning by the first indication information is determined as a candidate member.

In this embodiment, the first network element screens, according to the screening information and the first indication information, the device that satisfies the screening information and is willing to participate in federal learning as a candidate member, thereby helping to improve the efficiency of federal learning in the following.

Optionally, when the first network element determines the candidate member from the second devices according to the screening information and the first indication information, the first network element may further acquire capability information of each second device participating in federal learning from the third network element, so as to determine the candidate member according to the screening information, the first indication information and the capability information.

Wherein the capability information includes at least one of the following information: wireless access system participating in federal learning, region participating in federal learning, time participating in federal learning, algorithm information capable of being supported by participating in federal learning, accuracy information capable of being achieved by participating in federal learning, and type of participating in federal learning.

Types of participation in federal learning may include, among others, willing to participate in lateral federal learning, or willing to participate in longitudinal federal learning.

In one embodiment, the first network element determines one or more first devices as candidate members capable of participating in federal learning according to the screening information, including:

the first network element acquires willingness information of the first equipment to participate in the federal learning from a third network element;

and the first network element determines the first equipment willing to participate in the federal learning and matched with the screening information as a candidate member capable of participating in the federal learning according to the willingness information and the screening information.

the first network element obtains federally learned capability information of the first device, the capability information including at least one of: the wireless access system participating in the federal learning, the region participating in the federal learning, the time participating in the federal learning, the algorithm information capable of being supported by participating in the federal learning, the precision information capable of being achieved by participating in the federal learning and the type of participating in the federal learning;

And the first network element determines candidate members which can participate in federal learning from the first equipment according to the capability information and the screening information, and the capability information of the candidate members is matched with the screening information.

It should be noted that, the first network element may obtain the capability information of federal learning of the first device from a fourth network element, where the fourth network element may be at least one of the following: NRF, UDM, data acquisition coordination function (Data Collection Coordination Function, DCCF), AMF.

The capability information of the candidate member is matched with the screening information, and the capability information comprises at least one of the following:

the wireless access modes of the candidate members participating in federal learning are the same as the wireless access modes included in the screening information;

the candidate member participating in federal learning is located in an AOI included in the screening information;

the time of participation of the candidate member in federal learning is within a time period included in the screening information;

the algorithm types supported by the candidate members participating in federal learning are contained in the algorithm types contained in the screening information;

the accuracy information which can be achieved by the candidate members participating in federal learning is higher than the accuracy threshold included in the screening information;

The candidate member participates in federal learning of the same type as federal learning included in the screening information.

Optionally, the first network element may further obtain capability information of each device for participating in federal learning from the third network element, so as to match attribute information of each device with screening information, and determine that the first indication information indicates a device willing to participate in federal learning and the capability information participating in federal learning meets the screening information, which is a candidate member, thereby improving efficiency of performing federal learning subsequently.

Wherein the ability information to participate in federal learning satisfies the screening information includes:

the wireless mode of the candidate member participating in federal learning is the same as the wireless access mode included in the screening information;

For example, if the filtering information includes a supported algorithm type, and the supported algorithm type is a deep learning algorithm, the first network element screens out 80 devices that are originally involved in federal learning based on the first indication information, and the first network element may further screen out, based on the algorithm information that is acquired from the third network element and that can be supported by the 80 devices involved in federal learning, if 50 devices can support both the deep learning algorithm and the linear regression algorithm when participating in federal learning, and 30 devices can support only the deep learning algorithm when participating in federal learning, the first network element may further screen out, as candidate network elements, the first device that can support both the deep learning algorithm and the linear regression algorithm based on the algorithm type that can be supported when participating in federal learning.

It should be appreciated that where the capability information includes other information, the implementation is similar to the implementation where the capability information includes the types of algorithms that can be supported by participating in federal learning, and will not be described in detail herein.

The first network element can acquire willingness information and capability information of each device to participate in federal learning from the third network elements, wherein the UDM/DCAF/NRF is the network element with the capability of storing energy.

Optionally, the request message sent by the second network element further includes second indication information, where the second indication information is used to indicate a service type corresponding to federal learning. When determining the candidate member, the first network element may determine one or more first devices as the candidate member from a plurality of second devices according to the screening information and the second indication information, where the first devices are devices that can support a service corresponding to the service type.

Specifically, the second indication information may be an analytical ID, which is used to indicate a service type corresponding to federal learning. Among the data analysis tasks (i.e., other tasks for which NWDAF is responsible), the analytical ID may be: "UE mobility" is used to indicate that this task is an analysis task related to UE mobility, "NF load" is used to indicate that this task is an analysis task related to network element load.

In general, in the case that the service types corresponding to federal learning are different, the selected candidate members may be different, so when the first network element selects the candidate member, the first device capable of supporting the service corresponding to the service type from the plurality of second devices may also be selected as the candidate member according to the screening information and the service type indicated by the analytical ID. Such as: the second indication information is used for indicating that the service type corresponding to federal learning is an analysis task related to the mobility of the terminal, and the selected first device is a device capable of supporting the analysis task related to the mobility of the terminal.

In this embodiment, candidate members are further screened according to the service types corresponding to federal learning, so that the accuracy of the selected candidate members can be improved, and the training efficiency of federal learning and the accuracy of the trained model are improved.

the first network element obtains network state information corresponding to the first device, wherein the network state information comprises at least one of the following items:

type information of the first device;

location information of the first device;

wireless access system information of the first equipment;

wireless signal quality information of the first device;

and the first network element determines candidate members which can participate in federal learning from the first equipment according to the network state information corresponding to the first equipment and the screening information, and the network state information corresponding to the candidate members is matched with the screening information.

It should be noted that, the first network element may obtain the network state information corresponding to the first device from a third device, where the third device may be AMF, UDM, NRF, PCF, RAN or an operation, administration and maintenance device (OAM Operation Administration and Maintenance, OAM). For example, the first network element may obtain the type information of the first device from UDM or NRF, or the first network element may obtain the radio access system information where the first device is located from AMF or PCF, or the first network element may obtain the radio signal quality information of the first device from RAN or OAM, or the first network may obtain UE accumulated usage information from SMF or CHF, etc.

The network state information corresponding to the candidate member is matched with the screening information, and the network state information comprises at least one of the following items:

the wireless access system of the candidate member is the same as the wireless access system included in the screening information;

the wireless signal quality information of the candidate member is greater than the signal quality requirement included in the screening information;

the location of the candidate member is within an AOI included in the screening information;

the type of the candidate member includes member type information included in the filtering information.

Optionally, the request message further includes ordering indication information, where the ordering indication information is used to indicate ordering of the candidate members according to second information, and the second information includes at least one of the following: the signal quality of the candidate member, the precision information of the candidate member and the flow information of the candidate member; and under the condition that the ordering indication information is included in the request message, the identification sequence of the candidate members in the response message is the sequence obtained after ordering according to the second information.

Specifically, when the request message includes ordering indication information, the first network element performs ascending or descending order ordering on the determined candidate members according to the second information in the ordering indication information after determining the candidate members. For example, the ranking may be performed in ascending or descending order according to the signal quality of the candidate members, or may be performed in ascending or descending order according to the accuracy information of the candidate members, or may be performed in ascending or descending order according to the traffic information of the candidate members.

When the first network element sends the response message to the second network element, the identification sequence of the candidate members in the response message is the sequence obtained after the second information is ordered.

In this embodiment, the first network element ranks the determined candidate members according to the second information, so that after receiving the identification information of the candidate members sent by the first network element, the second network element further screens the candidate members more meeting the requirements of itself according to the ranking information to participate in subsequent federal learning, thereby improving the efficiency of federal learning.

Optionally, the request message further includes grouping indication information, where the grouping indication information is used to indicate grouping of the candidate members according to third information, and the third information includes at least one of the following: the region where the candidate member is located, the time period when the candidate member is in the networking state, the algorithm type supported by the candidate member the method comprises the steps of precision information of candidate members, wireless access modes of the candidate members and signal quality of the candidate members; and when the request message comprises grouping indication information, the identification of the candidate member in the response message is the identification obtained after sorting according to the third information.

Specifically, in the case that the request message includes grouping indication information, after determining the candidate member, the first network element groups the determined candidate member according to third information in the grouping indication information. For example, the candidate members may be grouped by the region in which they are located, such as grouping candidate members in the same region into a group, or the like. The grouping may be performed according to a time period in which the candidate members are in a networking state, for example, the grouping may be performed according to an algorithm type supported by the candidate members in the same time period, for example, the grouping may be performed according to a network element supporting the same algorithm type, or the grouping may be performed according to accuracy information of the candidate members, for example, the grouping may be performed according to a radio access system of the candidate members, for example, the grouping may be performed according to a signal quality of the candidate members, or the grouping may be performed according to traffic information of the candidate members, or the grouping may be performed according to a time period in which the candidate members can perform federal learning, for example, the grouping may be performed according to the same time period in which the candidate members can perform federal learning, for example, all 10 points to 12 points in the daytime, and so on.

When the first network element sends a response message to the second network element, the identification of the candidate member in the response message is the identification obtained after grouping according to the third information.

In this embodiment, the first network element groups the determined candidate members according to the third information, so that after receiving the identification information of the candidate members sent by the first network element, the second network element further screens the candidate members more meeting the requirements of the second network element according to the grouping information to participate in subsequent federal learning, thereby improving the efficiency of federal learning.

Step 203: the first network element sends a response message to the second network element, the response message including the identification of the candidate member.

In this step, after determining candidate members that can participate in federal learning, the first network element may send the identifiers of the candidate members to the second network element with the identifiers in a response message. The candidate member identification may include a user permanent identifier (Subscriber Permanent Identifier, SUPI) or a network protocol (Internet Protocol, IP) of the candidate member, which may also be a universal public subscription identity (Generic Public Subscription Identifier, GPSI), an international mobile subscriber identity (International Mobile Subscriber Identity, IMSI), an AF-specific UE identity (AF specific UE ID), or a UE IP address.

Optionally, the response message further includes at least one of the following information: the method comprises the steps of a region where a candidate member is located, a time period when the candidate member is in a networking state, algorithm types supported by the candidate member, precision information of the candidate member, wireless access modes of the candidate member, signal quality of the candidate member, flow information of the candidate member and a time period when the candidate member can perform federal learning.

The area where the candidate member is located may include a location where the candidate member is located, such as longitude and latitude, or a cell.

The time period for the candidate member to perform federal learning is a prediction result analyzed by the first network element according to the willingness information and the capability information of each device.

The signal quality of the candidate member, in the case of a connected WLAN, may be expressed as the mean, variance or RTT of the RSSI.

The traffic information of the candidate member may include a traffic value consumed in a preset period of time, such as the number of traffic uploaded and downloaded, and the like.

Optionally, the response message may further include second indication information, where the second indication information may be an analytical ID, which is used to indicate a service type corresponding to federal learning.

Optionally, the response message may further include a coverage time proportion in the networking state, for example, how long a day is in the wifi connection state.

Optionally, the response message may further include address information of the candidate member, and the second network element may find the candidate member according to the address information, so as to perform connection and federal learning.

Optionally, the above response message may further include quantity information, where the quantity information may be understood as the number of members participating in federal learning, i.e. the number of candidate members that needs to be determined by the first network element.

Further, after receiving the response message sent by the first network element, the second network element selects a target member actually participating in federal learning according to the situation of actually performing federal learning training.

For example, the second network element may determine the member to perform federal learning according to the signal quality of the candidate member in the response message, and may select the first 100 members with the best signal quality to perform federal learning. Of course, the manner of determining the target member according to other information in the response message is similar to the manner of determining the target member according to the signal quality, and will not be described herein.

Optionally, if the response message returned by the first network element includes the target number of candidate members under the condition that the request message sent by the second network element to the first network element includes the number information, the second network element may directly use the target number of candidate members as the target members. For example, if the request message includes 50 members requiring federal learning, and the response message returned by the first network element includes 50 candidate members, the second network element may directly select the 50 members as target members.

Further, the second network element connects the target members according to the identification information of the target members after determining the target members so as to perform federal learning.

In this embodiment, the second network element may further screen the target member actually participating in federal learning according to the candidate member returned by the first network element, so that the screened target member is more suitable for participating in federal learning, and efficiency of federal learning is improved.

The method for determining candidate members provided by the embodiment of the application determines one or more first devices as candidate members capable of participating in federal learning by receiving the request message sent by the second network element and according to screening information included in the request message, and sends identification information of the candidate members to the second network element, wherein the screening information comprises at least one of the following: time period, algorithm type, accuracy threshold, wireless access system, signal quality requirement, flow range, member type information, quantity information, region information and federal learning type information. The first network element can screen out the participation members suitable for federal learning through screening information, so that the efficiency of federal learning is improved.

Fig. 3 is a flow chart of another method for determining candidate members according to the present application. The execution body in this embodiment is a second network element. As shown in fig. 3, the method includes:

step 301: the second network element sends a request message to the first network element.

The request message comprises screening information, and the request message is used for indicating the first network element to determine one or more first devices as candidate members capable of participating in federal learning according to the screening information.

Wherein the screening information includes at least one of:

Quantity information indicating a quantity requirement of the candidate member;

Step 302: the second network element receives a response message sent by the first network element, wherein the response message comprises the identification of the candidate member.

The specific implementation process and the beneficial effects in the embodiment of the present application may refer to the content of the embodiment shown in fig. 2, and are not described herein again.

Optionally, the first device comprises at least one of:

a first device in a networked state for the period of time;

a first device supporting federal learning during the time period;

a first device supporting the algorithm type;

a first device supporting federal learning using the algorithm type;

the model training precision information is greater than the first equipment of the precision threshold value;

the first equipment is in the wireless access mode;

the first equipment supports federal learning under the wireless access system;

a first device having a signal quality greater than the signal quality requirement;

a first device having a consumed flow rate in the flow rate range;

A first device that meets the type requirements of the candidate member;

a first device located within the AOI;

and the type of the federal study is the same as the type information of the federal study included in the screening information.

Optionally, the request message further includes: and the indication information is used for indicating to execute the determination task of the candidate member participating in the federation learning or indicating that the task is selected by the member for federation learning or federation learning.

Optionally, the request message further includes ordering indication information, where the ordering indication information is used to indicate ordering of the candidate members according to second information, and the second information includes at least one of the following: the signal quality of the candidate member, the precision information of the candidate member and the flow information of the candidate member;

the identification of the candidate member in the response message is the identification obtained after sorting according to the second information.

Optionally, the request message further includes grouping indication information, where the grouping indication information is used to indicate grouping the candidate members according to third information, and the third information includes at least one of the following information: the region where the candidate member is located, the time period when the candidate member is in the networking state, the algorithm type supported by the candidate member the method comprises the steps of precision information of candidate members, wireless access modes of the candidate members and signal quality of the candidate members;

The identification of the candidate member in the response message is the identification obtained after grouping according to the third information.

Optionally, the response message further includes an area where the candidate member is located, a time period where the candidate member is in a networking state, an algorithm type supported by the candidate member, accuracy information of the candidate member, a wireless access format of the candidate member, signal quality of the candidate member, traffic information of the candidate member, and a time period where the candidate member can perform federal learning.

Optionally, the method further comprises:

and the second network element determines target members participating in federal learning according to the identification of the candidate members.

The method for determining candidate members provided by the embodiment of the application determines one or more first devices as candidate members capable of participating in federal learning by receiving a request message sent by a second network element and according to screening information included in the request message, and sends identification information of the candidate members to the second network element, wherein the screening information comprises at least one of the following: time period, algorithm type, accuracy threshold, wireless access system, signal quality requirement, flow range, member type information, quantity information, region information and federal learning type information. The first network element can screen out the participation members suitable for federal learning through screening information, so that the efficiency of federal learning is improved. In addition, the second network element can further screen and obtain members truly suitable for participating in federal learning based on the candidate members sent by the first network element, so that the efficiency of federal learning is improved.

The specific implementation process and technical effects of the method of the present embodiment are similar to those of the first network element side method embodiment, and specific reference may be made to the detailed description of the first network element side method embodiment, which is not repeated herein.

Fig. 4 is a signaling diagram of a method for determining candidate members according to the present application. As shown in fig. 4, the method includes:

step 401: the second network element sends a request message to the first network element.

The request message includes screening information, where the screening information includes at least one of the following: time period, algorithm type, accuracy threshold, wireless access system, signal quality requirement, flow range, member type information, quantity information, region information and federal learning type information.

Step 402: the first network element determines one or more first devices as candidate members that can participate in federal learning based on the screening information.

Step 403: the first network element sends a response message to the second network element, the response message including the identification of the candidate member.

The method for determining candidate members provided by the embodiment of the application determines one or more first devices as candidate members capable of participating in federal learning by receiving a request message sent by a second network element and according to screening information included in the request message, and sends identification information of the candidate members to the second network element, wherein the screening information comprises at least one of the following: time period, algorithm type, accuracy threshold, wireless access mode, signal quality requirement, flow range, member type information, quantity information, region of interest AOI and federal learning type. The first network element can screen out the participators suitable for federal learning through screening information, so that the training efficiency of federal learning is improved.

According to the candidate member determining method provided by the embodiment of the application, the execution subject can be the candidate member determining device. In the embodiment of the present application, a method for determining a candidate member by using a candidate member determining device is taken as an example, and the candidate member determining device provided in the embodiment of the present application is described.

Fig. 5 is a schematic structural diagram of a candidate member determination device provided by the present application. As shown in fig. 5, the candidate member determining apparatus provided in this embodiment includes:

a receiving module 11, configured to receive a request message sent by a second network element, where the request message includes screening information;

a processing module 12 configured to determine one or more first devices as candidate members that may participate in federal learning based on the screening information;

a sending module 13, configured to send a response message to the second network element, where the response message includes an identifier of the candidate member;

wherein the screening information includes at least one of:

quantity information indicating a quantity requirement of the candidate member;

In the device of this embodiment, the receiving module receives a request message sent by the second network element, where the request message includes screening information; the processing module determines one or more first devices as candidate members capable of participating in federal learning according to the screening information; the sending module sends a response message to the second network element, wherein the response message comprises the identification of the candidate member, and the screening information comprises at least one of the following: time period, algorithm type, accuracy threshold, wireless access mode, signal quality requirement, flow range, member type information, quantity information, region of interest AOI and federal learning type. The candidate member determining device can screen out the participating members suitable for federal learning through screening information, so that the training efficiency of federal learning is improved.

Optionally, the first device comprises at least one of:

a first device in a networked state for the period of time;

a first device supporting federal learning during the time period;

a first device supporting the algorithm type;

a first device supporting federal learning using the algorithm type;

the first equipment is in the wireless access mode;

the first equipment supports federal learning under the wireless access system;

a first device having a consumed flow rate in the flow rate range;

a first device that meets the type requirements of the candidate member;

a first device located within the AOI;

and the type of the federal study is the same as the type information of the federal study included in the screening information. Optionally, the processing module 12 is specifically configured to:

determining the data type corresponding to the screening information;

acquiring attribute information corresponding to each of at least one device from at least one third network element based on the data type;

and determining the candidate member according to the attribute information and the screening information.

Optionally, the data type includes at least one of:

networking information, corresponding time period, algorithm type, precision information, wireless access system, signal quality and flow.

Optionally, the processing module 12 is specifically configured to:

acquiring first indication information from a third network element, wherein the first indication information is used for representing willingness information of each second device to participate in federal learning; wherein the willingness information indicates whether the second devices are willing to participate in federal learning;

and determining one or more first devices as the candidate members according to the screening information and the first indication information, wherein the first devices are devices which are willing to participate in federal learning and meet the screening information in the second devices.

Optionally, the processing module 12 is specifically configured to:

acquiring willingness information of the first equipment to participate in the federal learning from a third network element;

and determining the first equipment willing to participate in the federal learning and matched with the screening information as a candidate member capable of participating in the federal learning according to the willingness information and the screening information.

Optionally, the processing module 12 is specifically configured to:

acquiring federally learned capability information of the first device, the capability information including at least one of: the wireless access system participating in the federal learning, the region participating in the federal learning, the time participating in the federal learning, the algorithm information capable of being supported by participating in the federal learning, the precision information capable of being achieved by participating in the federal learning and the type of participating in the federal learning;

And determining candidate members which can participate in federal learning from the first equipment according to the capability information and the screening information, wherein the capability information of the candidate members is matched with the screening information.

Wherein the capability information of the candidate member matches the screening information, including at least one of:

Optionally, the processing module 12 is specifically configured to:

acquiring network state information corresponding to the first device, wherein the network state information comprises at least one of the following items:

Type information of the first device;

location information of the first device;

wireless access system information of the first equipment;

wireless signal quality information of the first device;

and determining candidate members which can participate in federal learning from the first equipment according to the network state information corresponding to the first equipment and the screening information, wherein the network state information corresponding to the candidate members is matched with the screening information.

Optionally, the processing module is specifically configured to:

acquiring capability information of each second device participating in federal learning, wherein the capability information comprises at least one of the following information: the wireless access system participating in the federal learning, the region participating in the federal learning, the time participating in the federal learning, the algorithm information capable of being supported by participating in the federal learning, the precision information capable of being achieved by participating in the federal learning and the type of participating in the federal learning; and determining one or more first devices as the candidate members according to the screening information, the first indication information and the capability information, wherein the first devices are devices which are willing to participate in federal learning and have the capability information meeting the screening information in the second devices.

the candidate member participates in federal learning of the same type as federal learning included in the screening information. Optionally, the request message further includes second indication information, where the second indication information is used to indicate a service type corresponding to federal learning;

the processing module is specifically configured to:

determining one or more first devices as the candidate members according to the screening information and the second indication information;

the first device is a device capable of supporting a service corresponding to the service type.

Optionally, the request message further includes ordering indication information, where the ordering indication information is used to indicate that the candidate member is ordered according to second information, and the second information includes at least one of the following: the signal quality of the candidate member, the precision information of the candidate member and the flow information of the candidate member;

The identification sequence of the candidate members in the response message is the sequence obtained after the second information is ordered.

Optionally, the request message further includes grouping indication information, where the grouping indication information is used to indicate that the candidate members are grouped according to third information, and the third information includes at least one of the following information:

the method comprises the steps of locating a candidate member, locating a region where the candidate member is located, locating a time period where the candidate member is in a networking state, locating an algorithm type supported by the candidate member, locating precision information of the candidate member, locating a wireless access mode of the candidate member and locating a signal quality of the candidate member;

Optionally, the response message further includes at least one of the following information: the region where the candidate member is located, the time period when the candidate member is in a networking state, the algorithm type supported by the candidate member, the precision information of the candidate member the wireless access mode of the candidate member, the signal quality of the candidate member, the flow information of the candidate member and the time period for federal learning of the candidate member.

The apparatus of this embodiment may be used to execute the method of any one of the foregoing first network element side method embodiments, and specific implementation processes and technical effects of the method are similar to those of the first network element side method embodiment, and specific reference may be made to detailed description of the first network element side method embodiment, which is not repeated herein.

Fig. 6 is a second schematic structural diagram of a candidate member determination device according to the present application. As shown in fig. 6, the apparatus for determining candidate members provided in this embodiment includes:

a sending module 21, configured to send a request message to a first network element, where the request message includes screening information, and the request message is configured to instruct the first network element to determine, according to the screening information, one or more first devices as candidate members that can participate in federal learning;

a receiving module 22, configured to receive a response message sent by the first network element, where the response message includes an identifier of the candidate member;

wherein the screening information includes at least one of:

quantity information indicating a quantity requirement of the candidate member;

Optionally, the first device comprises at least one of:

a first device in a networked state for the period of time;

a first device supporting federal learning during the time period;

a first device supporting the algorithm type;

a first device supporting federal learning using the algorithm type;

the first equipment is in the wireless access mode;

The first equipment supports federal learning under the wireless access system;

a first device having a consumed flow rate in the flow rate range;

a first device that meets the type requirements of the candidate member;

a first device located within the AOI;

and the type of the federal study is the same as the type information of the federal study included in the screening information. Optionally, the request message further includes: and the indication information is used for indicating to execute the determination task of the candidate member participating in the federation learning or indicating that the task is selected by the member for federation learning or federation learning.

Optionally, the request message further includes grouping indication information, where the grouping indication information is used to indicate that the candidate members are grouped according to third information, and the third information includes at least one of the following information: the method comprises the steps of locating a candidate member, locating a region where the candidate member is located, locating a time period where the candidate member is in a networking state, locating an algorithm type supported by the candidate member, locating precision information of the candidate member, locating a wireless access mode of the candidate member and locating a signal quality of the candidate member;

Optionally, the response message further includes an area where the candidate member is located, a time period where the candidate member is in a networking state, an algorithm type supported by the candidate member, accuracy information of the candidate member, a wireless access mode of the candidate member, signal quality of the candidate member, flow information of the candidate member and a time period where the candidate member can perform federal learning.

Optionally, the apparatus further comprises: a processing module 23;

a processing module 23, configured to determine a target member participating in federal learning according to the identification of the candidate member.

The apparatus of this embodiment may be used to execute the method of any one of the foregoing second network element side method embodiments, and specific implementation processes and technical effects of the method are similar to those of the second network element side method embodiment, and specific reference may be made to detailed description of the second network element side method embodiment, which is not repeated herein.

The candidate member determining device in the embodiment of the present application may be an electronic device, for example, an electronic device with an operating system, or may be a component in an electronic device, for example, an integrated circuit or a chip. The electronic device may be a terminal, or may be other devices than a terminal. By way of example, terminals may include, but are not limited to, the types of terminals 11 listed above, other devices may be servers, network attached storage (Network Attached Storage, NAS), etc., and embodiments of the application are not specifically limited.

The candidate member determining device provided in the embodiment of the present application can implement each process implemented by the method embodiments of fig. 2 to fig. 4, and achieve the same technical effects, and in order to avoid repetition, a detailed description is omitted here.

Optionally, as shown in fig. 7, the embodiment of the present application further provides a communication device 700, including a processor 701 and a memory 702, where the memory 702 stores a program or instructions that can be executed on the processor 701, for example, when the communication device 700 is a terminal, the program or instructions implement, when executed by the processor 701, the steps of the above embodiment of the candidate member determining method, and achieve the same technical effects. When the communication device 700 is a network side device, the program or the instruction, when executed by the processor 701, implements the steps of the above embodiment of the method for determining candidate members, and the same technical effects can be achieved, so that repetition is avoided, and no further description is given here.

The embodiment of the application also provides a first network element, which comprises a processor and a communication interface, wherein the processor is used for determining one or more first devices as candidate members capable of participating in federal learning according to screening information, the communication interface is used for receiving a request message sent by a second network element and sending a response message to the second network element, the response message comprises the identification of the candidate members, the request message comprises the screening information, and the screening information comprises at least one of the following components: time period, algorithm type, accuracy threshold, wireless access system, signal quality requirement, flow range, member type information, quantity information, region information and federal learning type information. The terminal embodiment corresponds to the terminal-side method embodiment, and each implementation process and implementation manner of the method embodiment can be applied to the terminal embodiment, and the same technical effects can be achieved. Specifically, fig. 8 is a schematic hardware structure of a first network element for implementing an embodiment of the present application.

The first network element 800 includes, but is not limited to: at least part of the components of the radio frequency unit 801, the network module 802, the audio output unit 803, the input unit 804, the sensor 805, the display unit 806, the user input unit 807, the interface unit 808, the memory 809, and the processor 810, etc.

Those skilled in the art will appreciate that the first network element 800 may further include a power source (e.g., a battery) for powering the various components, and the power source may be logically connected to the processor 810 by a power management system, thereby implementing functions such as charge, discharge, and power consumption management by the power management system. The first network element structure shown in fig. 8 does not constitute a limitation of the first network element, and the first network element may include more or less components than those shown in the drawings, or may be combined with some components, or different component arrangements, which are not described herein.

It should be appreciated that in embodiments of the present application, the input unit 804 may include a graphics processing unit (Graphics Processing Unit, GPU) 8041 and a microphone 8042, with the graphics processor 8041 processing image data of still pictures or video obtained by an image capturing device (e.g., a camera) in a video capturing mode or an image capturing mode. The display unit 806 may include a display panel 8061, and the display panel 8061 may be configured in the form of a liquid crystal display, an organic light emitting diode, or the like. The user input unit 807 includes at least one of a touch panel 8071 and other input devices 8072. Touch panel 8071, also referred to as a touch screen. The touch panel 8071 may include two parts, a touch detection device and a touch controller. Other input devices 8072 may include, but are not limited to, a physical keyboard, function keys (e.g., volume control keys, switch keys, etc.), a trackball, a mouse, a joystick, and so forth, which are not described in detail herein.

In the embodiment of the present application, after receiving downlink data from the network side device, the radio frequency unit 801 may transmit the downlink data to the processor 810 for processing; in addition, the radio frequency unit 801 may send uplink data to the network side device. In general, the radio frequency unit 801 includes, but is not limited to, an antenna, an amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, and the like.

The memory 809 may be used to store software programs or instructions and various data. The memory 809 may mainly include a first storage area storing programs or instructions and a second storage area storing data, wherein the first storage area may store an operating system, application programs or instructions (such as a sound playing function, an image playing function, etc.) required for at least one function, and the like. Further, the memory 809 may include volatile memory or nonvolatile memory, or the memory 809 may include both volatile and nonvolatile memory. The nonvolatile Memory may be a Read-Only Memory (ROM), a Programmable ROM (PROM), an Erasable PROM (EPROM), an Electrically Erasable EPROM (EEPROM), or a flash Memory. The volatile memory may be random access memory (Random Access Memory, RAM), static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double Data Rate SDRAM (ddr SDRAM), enhanced SDRAM (Enhanced SDRAM), synchronous DRAM (SLDRAM), and Direct RAM (DRRAM). Memory x09 in embodiments of the application includes, but is not limited to, these and any other suitable types of memory.

The processor 810 may include one or more processing units; optionally, the processor 810 integrates an application processor that primarily processes operations involving an operating system, user interface, application programs, etc., and a modem processor that primarily processes wireless communication signals, such as a baseband processor. It will be appreciated that the modem processor described above may not be integrated into the processor 810.

The radio frequency unit 801 is configured to receive a request message sent by a second network element, where the request message includes screening information, and send a response message to the second network element, where the response message includes an identifier of the candidate member.

A processor 810 configured to determine one or more first devices as candidate members that may participate in federal learning based on screening information comprising at least one of:

quantity information indicating a quantity requirement of the candidate member;

In the above embodiment, the first network element receives the request message sent by the second network element, determines, according to the screening information included in the request message, one or more first devices as candidate members capable of participating in federal learning, and sends identification information of the candidate members to the second network element, where the screening information includes at least one of the following: time period, algorithm type, accuracy threshold, wireless access mode, signal quality requirement, flow range, member type information, quantity information, region of interest AOI and federal learning type. The first network element can screen out the participators suitable for federal learning through screening information, so that the training efficiency of federal learning is improved.

Optionally, the first device comprises at least one of:

a first device in a networked state for the period of time;

a first device supporting federal learning during the time period;

a first device supporting the algorithm type;

a first device supporting federal learning using the algorithm type;

the first equipment is in the wireless access mode;

the first equipment supports federal learning under the wireless access system;

a first device having a consumed flow rate in the flow rate range;

a first device that meets the type requirements of the candidate member;

a first device located within the AOI;

Optionally, the processor 810 is further configured to determine a data type corresponding to the screening information;

acquiring attribute information corresponding to each of at least one second device from at least one third network element based on the data type;

Optionally, the data type includes at least one of:

Optionally, the processor 810 is further configured to obtain first indication information from the third network element, where the first indication information is used to represent willingness information of each second device to participate in federal learning; wherein the willingness information indicates whether the devices are willing to participate in federal learning;

and according to the screening information and the first indication information, determining one or more first devices as the candidate members, wherein the first devices are the devices which are willing to participate in federal learning and meet the screening information in the second devices.

Optionally, the processor 810 is configured to obtain willingness information of the first device to participate in the federal learning from a third network element; and determining the first equipment willing to participate in the federal learning and matched with the screening information as a candidate member capable of participating in the federal learning according to the willingness information and the screening information.

Optionally, the processor 810 is configured to obtain federally learned capability information of the first device, where the capability information includes at least one of the following information: the wireless access system participating in the federal learning, the region participating in the federal learning, the time participating in the federal learning, the algorithm information capable of being supported by participating in the federal learning, the precision information capable of being achieved by participating in the federal learning and the type of participating in the federal learning; and determining candidate members which can participate in federal learning from the first equipment according to the capability information and the screening information, wherein the capability information of the candidate members is matched with the screening information.

Optionally, the processor 810 is configured to obtain network status information corresponding to the first device, where the network status information includes at least one of the following:

type information of the first device;

location information of the first device;

wireless access system information of the first equipment;

wireless signal quality information of the first device;

Optionally, the processor 810 is configured to obtain capability information of each second device participating in federal learning, where the capability information includes at least one of the following information: the wireless access system participating in the federal learning, the region participating in the federal learning, the time participating in the federal learning, the algorithm information capable of being supported by participating in the federal learning, the precision information capable of being achieved by participating in the federal learning and the type of participating in the federal learning; and determining one or more first devices as the candidate members according to the screening information, the first indication information and the capability information, wherein the first devices are devices which are willing to participate in federal learning and have the capability information meeting the screening information in the second devices.

Optionally, the request message further includes second indication information, where the second indication information is used to indicate a service type corresponding to federal learning;

a processor 810 further configured to determine one or more first devices as the candidate members based on the screening information and the second indication information;

In the above embodiment, the first network element receives the request message sent by the second network element, determines, according to the screening information included in the request message, one or more first devices as candidate members that can participate in federal learning, and sends identification information of the candidate members to the second network element, where the screening information includes at least one of the following: time period, algorithm type, accuracy threshold, wireless access mode, signal quality requirement, flow range, member type information, quantity information, region of interest AOI and federal learning type. The first network element can screen out the participators suitable for federal learning through screening information, so that the training efficiency of federal learning is improved.

Specifically, the embodiment of the application also provides a second network element. As shown in fig. 9, the second network element 900 includes: a processor 901, a network interface 902, and a memory 903. The network interface 902 is, for example, a common public radio interface (common public radio interface, CPRI).

Specifically, the network side device 900 of the embodiment of the present application further includes: instructions or programs stored in the memory 903 and executable on the processor 901, the processor 901 invokes the instructions or programs in the memory 903 to execute the method executed by each module shown in fig. 6, and achieve the same technical effects, so that repetition is avoided and thus a description thereof is omitted.

The embodiment of the application also provides a readable storage medium, on which a program or an instruction is stored, which when executed by a processor, implements each process of the above embodiment of the candidate member determination method, and can achieve the same technical effects, so that repetition is avoided, and no further description is given here.

Wherein the processor is a processor in the terminal described in the above embodiment. The readable storage medium includes computer readable storage medium such as computer readable memory ROM, random access memory RAM, magnetic or optical disk, etc.

The embodiment of the application further provides a chip, which comprises a processor and a communication interface, wherein the communication interface is coupled with the processor, and the processor is used for running a program or instructions to realize the processes of the embodiment of the candidate member determination method, and the same technical effects can be achieved, so that repetition is avoided, and the description is omitted here.

It should be understood that the chips referred to in the embodiments of the present application may also be referred to as system-on-chip chips, or the like.

The embodiment of the present application further provides a computer program/program product, where the computer program/program product is stored in a storage medium, and the computer program/program product is executed by at least one processor to implement the respective processes of the above embodiment of the candidate member determination method, and achieve the same technical effects, so that repetition is avoided, and details are not repeated herein.

The embodiment of the application also provides a system for determining the candidate members, which comprises the following steps: a first network element operable to perform the steps of the method of determining candidate members as described above, and a second network element operable to perform the steps of the method of determining candidate members as described above.

It should be noted that, in this document, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one … …" does not exclude the presence of other like elements in a process, method, article, or apparatus that comprises the element. Furthermore, it should be noted that the scope of the methods and apparatus in the embodiments of the present application is not limited to performing the functions in the order shown or discussed, but may also include performing the functions in a substantially simultaneous manner or in an opposite order depending on the functions involved, e.g., the described methods may be performed in an order different from that described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.

From the above description of the embodiments, it will be clear to those skilled in the art that the above-described embodiment method may be implemented by means of software plus a necessary general hardware platform, but of course may also be implemented by means of hardware, but in many cases the former is a preferred embodiment. Based on such understanding, the technical solution of the present application may be embodied essentially or in a part contributing to the prior art in the form of a computer software product stored in a storage medium (e.g. ROM/RAM, magnetic disk, optical disk) comprising instructions for causing a terminal (which may be a mobile phone, a computer, a server, an air conditioner, or a network device, etc.) to perform the method according to the embodiments of the present application.

The embodiments of the present application have been described above with reference to the accompanying drawings, but the present application is not limited to the above-described embodiments, which are merely illustrative and not restrictive, and many forms may be made by those having ordinary skill in the art without departing from the spirit of the present application and the scope of the claims, which are to be protected by the present application.

Claims

1. A method of determining candidate members, comprising:

wherein the screening information includes at least one of:

Quantity information indicating a quantity requirement of the candidate member;

2. The method according to claim 1, wherein the request message further comprises:

and the indication information is used for indicating to execute the determination task of the candidate member participating in the federation learning or indicating that the task is selected by the member for federation learning or federation learning.

3. The method of claim 1, wherein the first device comprises at least one of:

a first device in a networked state for the period of time;

a first device supporting federal learning during the time period;

a first device supporting the algorithm type;

a first device supporting federal learning using the algorithm type;

the first equipment is in the wireless access mode;

the first equipment supports federal learning under the wireless access system;

a first device having a consumed flow rate in the flow rate range;

a first device that meets the type requirements of the candidate member;

a first device located within the AOI;

4. A method according to claim 3, wherein the first network element determining one or more first devices as candidate members for participation in federal learning based on the screening information, comprising:

the first network element acquires first indication information from the third network element, wherein the first indication information is used for representing willingness information of each second device to participate in federal learning; wherein the willingness information indicates whether the second devices are willing to participate in federal learning;

5. A method according to any one of claims 1 to 3, wherein the first network element determining one or more first devices as candidate members that can participate in federal learning based on the screening information, comprising:

6. The method according to any one of claims 1 to 5, wherein the first network element determining one or more first devices as candidate members that can participate in federal learning based on the screening information, comprising:

7. The method of claim 6, wherein the capability information of the candidate member matches the screening information, comprising at least one of:

8. The method according to any one of claims 1 to 7, wherein the first network element determining one or more first devices as candidate members that can participate in federal learning based on the screening information, comprising:

Type information of the first device;

location information of the first device;

wireless access system information of the first equipment;

wireless signal quality information of the first device;

9. The method of claim 8, wherein the network state information corresponding to the candidate member matches the screening information, comprising at least one of:

10. The method of claim 4, wherein the determining one or more first devices as the candidate members based on the screening information and the first indication information comprises:

The first network element acquires capability information of each second device participating in federal learning, wherein the capability information comprises at least one of the following information: the wireless access system participating in the federal learning, the region participating in the federal learning, the time participating in the federal learning, the algorithm information capable of being supported by participating in the federal learning, the precision information capable of being achieved by participating in the federal learning and the type of participating in the federal learning;

and determining one or more first devices as the candidate members according to the screening information, the first indication information and the capability information, wherein the first devices are devices which are willing to participate in federal learning and have the capability information meeting the screening information in the second devices.

11. The method according to any one of claims 1 to 10, wherein the request message further includes second indication information, where the second indication information is used to indicate a service type corresponding to federal learning;

the first network element determines one or more first devices as candidate members capable of participating in federal learning according to the screening information, and the method comprises the following steps:

the first network element determines one or more first devices as the candidate members according to the screening information and the second indication information;

12. The method according to any one of claims 1 to 10, wherein the request message further includes ordering indication information, where the ordering indication information is used to indicate that the candidate member is ordered according to second information, where the second information includes at least one of: the signal quality of the candidate member, the precision information of the candidate member and the flow information of the candidate member;

13. The method according to any one of claims 1 to 10, wherein the request message further includes grouping indication information, the grouping indication information being used for indicating that the candidate members are grouped according to third information, the third information including at least one of the following information:

14. The method according to any one of claims 1 to 10, wherein the response message further comprises at least one of the following information: the region where the candidate member is located, the time period when the candidate member is in a networking state, the algorithm type supported by the candidate member, the precision information of the candidate member the wireless access mode of the candidate member, the signal quality of the candidate member, the flow information of the candidate member and the time period for federal learning of the candidate member.

15. A method of determining candidate members, comprising:

wherein the screening information includes at least one of:

quantity information indicating a quantity requirement of the candidate member;

16. The method of claim 15, wherein the request message further comprises:

17. The method of claim 15, wherein the first device comprises at least one of:

a first device in a networked state for the period of time;

a first device supporting federal learning during the time period;

a first device supporting the algorithm type;

a first device supporting federal learning using the algorithm type;

the first equipment is in the wireless access mode;

the first equipment supports federal learning under the wireless access system;

a first device having a consumed flow rate in the flow rate range;

a first device that meets the type requirements of the candidate member;

a first device located within the AOI;

18. The method according to any one of claims 15 to 17, wherein the request message further comprises ordering indication information, the ordering indication information being used to indicate ordering of the candidate members according to second information, the second information comprising at least one of: the signal quality of the candidate member, the precision information of the candidate member and the flow information of the candidate member;

19. The method according to any one of claims 15 to 17, wherein the request message further includes grouping indication information, the grouping indication information being used for indicating that the candidate members are grouped according to third information, the third information including at least one of the following information: the method comprises the steps of locating a candidate member, locating a region where the candidate member is located, locating a time period where the candidate member is in a networking state, locating an algorithm type supported by the candidate member, locating precision information of the candidate member, locating a wireless access mode of the candidate member and locating a signal quality of the candidate member;

20. The method according to any one of claims 15 to 17, wherein the response message further includes an area in which the candidate member is located, a period of time in which the candidate member is in a networking state, a type of algorithm supported by the candidate member, accuracy information of the candidate member, a wireless access format of the candidate member, a signal quality of the candidate member, traffic information of the candidate member, and a period of time in which the candidate member can perform federal learning.

21. The method according to claim 17 or 18, characterized in that the method further comprises:

22. A candidate member determination device, comprising:

wherein the screening information includes at least one of:

quantity information indicating a quantity requirement of the candidate member;

23. A candidate member determination device, comprising:

wherein the screening information includes at least one of:

quantity information indicating a quantity requirement of the candidate member;

24. A first network element comprising a processor and a memory storing a program or instructions executable on the processor, which when executed by the processor, implement the steps of the method of determining candidate members as claimed in any one of claims 1 to 14.

25. A second network element comprising a processor and a memory storing a program or instructions executable on the processor, which when executed by the processor, implement the steps of the method of determining candidate members as claimed in any one of claims 15 to 21.

26. A readable storage medium, characterized in that it has stored thereon a program or instructions which, when executed by a processor, implements the method of determining a candidate member according to any of claims 1 to 14, or the steps of the method of determining a candidate member according to any of claims 15 to 21.