WO2023185788A1 - 候选成员的确定方法、装置及设备 - Google Patents

候选成员的确定方法、装置及设备 Download PDF

Info

Publication number
WO2023185788A1
WO2023185788A1 PCT/CN2023/084244 CN2023084244W WO2023185788A1 WO 2023185788 A1 WO2023185788 A1 WO 2023185788A1 CN 2023084244 W CN2023084244 W CN 2023084244W WO 2023185788 A1 WO2023185788 A1 WO 2023185788A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
federated learning
candidate
network element
indicate
Prior art date
Application number
PCT/CN2023/084244
Other languages
English (en)
French (fr)
Inventor
程思涵
崇卫微
Original Assignee
维沃移动通信有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Priority claimed from CN202210476433.9A external-priority patent/CN116866882A/zh
Application filed by 维沃移动通信有限公司 filed Critical 维沃移动通信有限公司
Publication of WO2023185788A1 publication Critical patent/WO2023185788A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/02—Arrangements for optimising operational condition

Definitions

  • This application belongs to the field of communication technology, and specifically relates to a method, device and equipment for determining candidate members.
  • Federated learning refers to a method of machine learning modeling by uniting different participants (participants, or parties, also called data owners, or clients).
  • participants do not need to expose their own data to other participants and coordinators (coordinators, also known as servers, parameter servers, or aggregation servers), so Federated learning can well protect user privacy and data security, and can solve the problem of data islands.
  • the embodiments of the present application provide a method, device and equipment for determining candidate members, which can solve the problem of low training efficiency of federated learning due to the fact that participating members of federated learning do not meet the requirements.
  • the first aspect provides a method for determining candidate members, including:
  • the first network element receives a request message sent by the second network element, where the request message includes screening information
  • the first network element determines one or more first devices as candidate members that can participate in federated learning based on the screening information
  • the first network element sends a response message to the second network element, where the response message includes the identity of the candidate member;
  • the screening information includes at least one of the following:
  • Algorithm type used to indicate the type of algorithm that needs to be supported for model training for federated learning
  • Accuracy threshold used to indicate the accuracy requirements of model training that need to be met to perform the federated learning
  • Wireless access standard used to indicate the wireless access standard to be selected for the federated learning
  • Signal quality requirements used to indicate wireless signal quality requirements when performing the federated learning
  • Traffic range used to indicate the traffic usage range requirements of the candidate member
  • Quantity information used to indicate the quantity requirements of the candidate members
  • Area of interest AOI used to indicate the area where the candidate member is located
  • the type information of federated learning is used to indicate whether the federated learning belongs to vertical federation or horizontal federation.
  • the second aspect provides a method for determining candidate members, including:
  • the second network element sends a request message to the first network element.
  • the request message includes filtering information.
  • the request message is used to instruct the first network element to determine one or more first devices according to the filtering information.
  • the second network element receives a response message sent by the first network element, where the response message includes the identity of the candidate member;
  • the screening information includes at least one of the following:
  • Algorithm type used to indicate the type of algorithm that needs to be supported for model training for federated learning
  • Accuracy threshold used to indicate the accuracy requirements of model training that need to be met to perform the federated learning
  • Wireless access standard used to indicate the wireless access standard to be selected for the federated learning
  • Signal quality requirements used to indicate wireless signal quality requirements when performing the federated learning
  • Traffic range used to indicate the traffic usage range requirements of the candidate member
  • Quantity information used to indicate the quantity requirements of the candidate members
  • Area of interest AOI used to indicate the area where the candidate member is located
  • the type information of federated learning is used to indicate whether the federated learning belongs to vertical federation or horizontal federation.
  • a device for determining candidate members including:
  • a receiving module configured to receive a request message sent by the second network element, where the request message includes screening information
  • a processing module configured to determine one or more first devices as candidate members that can participate in federated learning based on the screening information
  • a sending module configured to send a response message to the second network element, where the response message includes the identification of the candidate member
  • the screening information includes at least one of the following:
  • Algorithm type used to indicate the type of algorithm that needs to be supported for model training for federated learning
  • Accuracy threshold used to indicate the accuracy requirements of model training that need to be met to perform the federated learning
  • Wireless access standard used to indicate the wireless access standard to be selected for the federated learning
  • Signal quality requirements used to indicate wireless signal quality requirements when performing the federated learning
  • Traffic range used to indicate the traffic usage range requirements of the candidate member
  • Quantity information used to indicate the quantity requirements of the candidate members
  • Area of interest AOI used to indicate the area where the candidate member is located
  • the type information of federated learning is used to indicate whether the federated learning belongs to vertical federation or horizontal federation. state.
  • the fourth aspect provides a device for determining candidate members, including:
  • a sending module configured to send a request message to the first network element, where the request message includes screening information, and the request message is used to instruct the first network element to determine one or more first devices based on the screening information.
  • a receiving module configured to receive a response message sent by the first network element, where the response message includes the identification of the candidate member
  • the screening information includes at least one of the following:
  • Algorithm type used to indicate the type of algorithm that needs to be supported for model training for federated learning
  • Accuracy threshold used to indicate the accuracy requirements of model training that need to be met to perform the federated learning
  • Wireless access standard used to indicate the wireless access standard to be selected for the federated learning
  • Signal quality requirements used to indicate wireless signal quality requirements when performing the federated learning
  • Traffic range used to indicate the traffic usage range requirements of the candidate member
  • Quantity information used to indicate the quantity requirements of the candidate members
  • Area of interest AOI used to indicate the area where the candidate member is located
  • the type information of federated learning is used to indicate whether the federated learning belongs to vertical federation or horizontal federation.
  • a first network element in a fifth aspect, includes a processor and a memory.
  • the memory stores programs or instructions that can be run on the processor. The programs or instructions are processed by the processor. When the processor is executed, the steps of the method described in the first aspect are implemented.
  • a first network element including a processor and a communication interface, wherein the processor is configured to determine one or more first devices as candidate members that can participate in federated learning based on screening information, and the The communication interface is configured to receive a request message sent by the second network element, and send a response message to the second network element, where the response message includes the identification of the candidate member, wherein the request message includes filtering information, and the filtering
  • the information includes at least one of the following:
  • Algorithm type used to indicate the type of algorithm that needs to be supported for model training for federated learning
  • Accuracy threshold used to indicate the accuracy requirements of model training that need to be met to perform the federated learning
  • Wireless access standard used to indicate the wireless access standard to be selected for the federated learning
  • Signal quality requirements used to indicate wireless signal quality requirements when performing the federated learning
  • Traffic range used to indicate the traffic usage range requirements of the candidate member
  • Quantity information used to indicate the quantity requirements of the candidate members
  • Area of interest AOI used to indicate the area where the candidate member is located
  • the type information of federated learning is used to indicate whether the federated learning belongs to vertical federation or horizontal federation.
  • a second network element in a seventh aspect, includes a processor and a memory.
  • the memory stores programs or instructions that can be run on the processor.
  • the program or instructions are executed by the processor.
  • a second network element including a processor and a communication interface, wherein the communication interface is used to send a request message to the first network element, the request message includes filtering information, and the request message Used to instruct the first network element to determine one or more first devices as candidate members that can participate in federated learning based on the screening information; receive a response message sent by the first network element, the response message including The identification of the candidate member;
  • the screening information includes at least one of the following:
  • Algorithm type used to indicate the type of algorithm that needs to be supported for model training for federated learning
  • Accuracy threshold used to indicate the accuracy requirements of model training that need to be met to perform the federated learning
  • Wireless access standard used to indicate the wireless access standard to be selected for the federated learning
  • Signal quality requirements used to indicate wireless signal quality requirements when performing the federated learning
  • Traffic range used to indicate the traffic usage range requirements of the candidate member
  • Quantity information used to indicate the quantity requirements of the candidate members
  • Area of interest AOI used to indicate the area where the candidate member is located
  • the type information of federated learning is used to indicate whether the federated learning belongs to vertical federation or horizontal federation.
  • a system for determining candidate members including: a first network element and a second network element.
  • the first network element can be used to perform the steps of the method for determining candidate members as described in the first aspect
  • the second network element may be configured to perform the steps of the method for determining candidate members described in the second aspect.
  • a readable storage medium In a tenth aspect, a readable storage medium is provided. Programs or instructions are stored on the readable storage medium. When the programs or instructions are executed by a processor, the steps of the method described in the first aspect are implemented, or the steps of the method are implemented as described in the first aspect. The steps of the method described in the second aspect.
  • a chip in an eleventh aspect, includes a processor and a communication interface.
  • the communication interface is coupled to the processor.
  • the processor is used to run programs or instructions to implement the method described in the first aspect. method, or implement a method as described in the second aspect.
  • a computer program/program product is provided, 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 steps of the first aspect and the third aspect.
  • the first network element receives a request message sent by the second network element and determines one or more first devices as candidate members that can participate in federated learning based on the screening information included in the request message. and send the 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 standard, signal quality requirements, traffic range, member type information, quantity information , area of interest AOI and types of federated learning.
  • the first network element can select participating members suitable for federated learning, thereby improving the training efficiency of federated learning.
  • Figure 1 shows a block diagram of a wireless communication system to which embodiments of the present application are applicable
  • Figure 2 is a schematic flow chart of a method for determining candidate members provided by this application.
  • Figure 3 is a schematic flow chart of another method for determining candidate members provided by this application.
  • Figure 4 is a signaling diagram of a method for determining candidate members provided by this application.
  • Figure 5 is one of the structural schematic diagrams of the device for determining candidate members provided by this application.
  • Figure 6 is the second structural schematic diagram of the device for determining candidate members provided by this application.
  • Figure 7 is a schematic structural diagram of a communication device provided by an embodiment of the present application.
  • Figure 8 is a schematic diagram of the hardware structure of the first network element provided by the embodiment of the present application.
  • Figure 9 is a schematic diagram of the hardware structure of the second network element provided by the embodiment of the present application.
  • first, second, etc. in the description and claims of this application are used to distinguish similar objects and are not used to describe a specific order or sequence. It is to be understood that the terms so used are interchangeable under appropriate circumstances so that the embodiments of the present application can be practiced in sequences other than those illustrated or described herein, and that "first" and “second” are distinguished objects It is usually one type, and the number of objects is not limited.
  • the first object can be one or multiple.
  • “and/or” in the description and claims indicates at least one of the connected objects, and the character “/" generally indicates that the related objects are in an "or” relationship.
  • LTE Long Term Evolution
  • LTE-Advanced, LTE-A Long Term Evolution
  • LTE-A Long Term Evolution
  • CDMA Code Division Multiple Access
  • TDMA Time Division Multiple Access
  • FDMA Frequency Division Multiple Access
  • OFDMA Orthogonal Frequency Division Multiple Access
  • SC-FDMA Single-carrier Frequency Division Multiple Access
  • system in the embodiments of this application "System” and “network” are often used interchangeably, and the techniques described can be used with the systems and radio technologies mentioned above as well as with other systems and radio technologies.
  • the following description describes the new air interface ( New Radio (NR) systems, and the term NR is used in most of the following descriptions, but these technologies can also be applied to applications other than NR system applications, such as 6th Generation (6G) communication systems.
  • 6G 6th Generation
  • FIG. 1 shows a block diagram of a wireless communication system to which embodiments of the present application are applicable.
  • the wireless communication system includes a terminal 11 and a network side device 12.
  • the terminal 11 can also be called user equipment (UE), where the terminal 11 can be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop computer (Laptop Computer), or a notebook computer or a personal digital assistant.
  • UE user equipment
  • PDA Personal Digital Assistant
  • handheld computer netbook
  • ultra-mobile personal computer UMPC
  • mobile Internet device Mobile Internet Device, MID
  • AR augmented reality
  • VR virtual reality
  • robots wearable devices
  • VUE vehicle-mounted equipment
  • PUE pedestrian terminals
  • smart home home equipment with wireless communication functions, such as refrigerators, TVs, washing machines or furniture, etc.
  • game consoles personal computers (PC)
  • teller machines or self-service machines and other terminal-side devices such as refrigerators, TVs, washing machines or furniture, etc.
  • the network side device 12 may include an access network device or a core network device, where the access network device 12 may also be called a radio access network device, a radio access network (Radio Access Network, RAN), a radio access network function or Wireless access network unit.
  • the access network device 12 may include a base station, a WLAN access point or a WiFi node, etc.
  • the base station may be called a Node B, an evolved Node B (eNB), an access point, a Base Transceiver Station (BTS), a radio Base station, radio transceiver, Basic Service Set (BSS), Extended Service Set (ESS), Home Node B, Home Evolved Node B, Transmitting Receiving Point (TRP) or all
  • eNB evolved Node B
  • BTS Base Transceiver Station
  • BSS Basic Service Set
  • ESS Extended Service Set
  • Home Node B Home Evolved Node B
  • TRP Transmitting Receiving Point
  • Core network equipment may include but is not limited to at least one of the following: 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 Function (UPF), Policy Control Function (PCF), Policy and Charging Rules Function (PCRF), Edge Application Service Discovery function (Edge Application Server Discovery Function, EASDF), Unified Data Management (UDM), Unified Data Repository (UDR), Home Subscriber Server (HSS), centralized network configuration ( Centralized network configuration (CNC), Network Repository Function (NRF), Network Exposure Function (NEF), Local NEF (Local NEF, or L-NEF), Binding Support Function (Binding Support Function, BSF), application function (Application Function, AF), etc.
  • MME mobility management entities
  • AMF Access and Mobility Management Function
  • SMF Session Management Function
  • UPF User Plane Function
  • PCF Policy Control Function
  • Federated learning includes horizontal federated learning and vertical federated learning.
  • the essence of horizontal federated learning is the union of samples. It is suitable for scenarios where participants have the same business format but reach different customers, that is, there is a lot of feature overlap and little sample overlap, such as communications.
  • the core network (CN) domain and the Radio Access Network (RAN) domain in the network serve the same service (such as mobility management (MM)) for different users (such as each terminal, that is, the sample is different) ), session management (session management, SM) business, a certain business).
  • MM mobility management
  • SM session management
  • the essence of vertical federated learning is the combination of features, which is suitable for situations where there is a lot of sample overlap and little feature overlap. Scenarios, such as different services (such as MM and SM services, with different characteristics) serving the same users (such as terminals, that is, the same samples) in the CN domain and the RAN domain in the communication network. By combining different data features of the common samples of the participants, vertical federation increases the feature dimensions of the training samples and obtains a better model.
  • the federated learning system includes a manager (server) and multiple participants.
  • the manager is used to send models to each participant, update the model based on the feedback results of each participant, and send the updated model to each participant again. Participants are used for the next round of model training.
  • Each participant has its own data. In order not to send local data to others, each participant uses the model sent by the manager to train locally and then returns it to the manager for model update.
  • model training process may include the following steps:
  • Step 1 Each participant downloads the latest model from the manager (server);
  • Step 2 Each participant uses local data to train the model, and uploads the encrypted gradient to the manager (server).
  • the manager (server) aggregates the gradients of each participant to update the model parameters;
  • Step 3 The manager (server) returns the updated model to each participant;
  • Step 4 Each participant updates their respective models.
  • the above steps 2 to 4 carry out multiple iterations/model updates, and under certain conditions (such as completing a certain number of iterations or the calculated value of the model's loss function is lower than the preset value), the model training is completed.
  • NWDAF network data analytic function
  • NWDAF can collect data from various network elements of the core network, network management systems, etc. to conduct big data statistics, analysis or intelligent data analysis, and obtain network-side analysis or prediction data, thereby assisting each network element to connect terminals based on the data analysis results. for more effective control.
  • the data of other network elements can be collected through NWDAF, and these data can be analyzed to select participating members suitable for federated learning based on the screening information and data analysis results, thereby improving the federation learning efficiency.
  • Figure 2 is a schematic flow chart of a method for determining candidate members provided by this application. The following is combined with Figure 2 describes the method for determining candidate members provided by the embodiment of the present application. As shown in Figure 2, the method includes:
  • Step 201 The first network element receives the request message sent by the second network element.
  • the execution subject of the method for determining candidate members provided by the embodiment of the present application is the first network element, and the first network element can be implemented in various forms.
  • the first network element described in the embodiment of this application may include NWDAF.
  • NWDAF NWDAF
  • it may also be other network elements that can collect data from various network elements of the core network, network management systems, etc. for big data statistics, analysis, or intelligent data analysis.
  • the second network element may include a task consumer network element, for example, it may be an application function network element (AF), a base station, a terminal and other network elements or devices, and it may also be a third-party server.
  • AF application function network element
  • the request message includes filtering information, and the filtering information includes at least one of the following: time period, algorithm type, accuracy threshold, wireless access standard, signal quality requirements, traffic range, member type information, quantity information, regional information, and federated learning type information. Based on the screening information of different dimensions, more suitable participating members can be selected, thereby improving the efficiency of federated learning.
  • the time period is used to indicate an optional time period for performing the federated learning.
  • the time period can be a certain time period in the past or a future time period. It can be understood that if it is a future time period, the first A network element will predict the networking status of each member in the future based on the historical data obtained.
  • the networking status includes the status of being connected to the network and the status of not being connected to the network. It should be understood that the time period is usually set flexibly according to the actual time period for federated learning.
  • Algorithm type used to indicate the type of algorithm that needs to be supported for model training in federated learning, including the type of algorithms supported by each member related to machine learning and other artificial intelligence (AI) data analysis tasks, such as “deep learning algorithms” ", “Linear Regression Algorithm”, etc.
  • AI artificial intelligence
  • the accuracy threshold is used to indicate the accuracy requirements of model training that need to be met for federated learning, including the accuracy value that can be achieved by the model generated by each member after training, such as the accuracy of the model, which can also be understood as model prediction or judgment. Correct percentage. It should be understood that in order to ensure the accuracy of federated learning, some members with relatively high accuracy are usually selected as participating members of federated learning.
  • Wireless access format used to indicate the wireless access format to be selected for federated learning, including the way each member connects or accesses the communication network, such as connecting to non-3GPP WLAN (similar to connecting to WiFi), connecting to 3GPP's 5g, 4g Network etc. It should be understood that in order to ensure the stability of federated learning and reduce traffic consumption, some members connected to WLAN or WiFi are usually selected as participating members of federated learning.
  • Signal quality requirements used to indicate wireless signal quality requirements when performing the federated learning.
  • Signal quality requirements may include network signal strength thresholds and/or stability thresholds when each member connects to the communication network.
  • the signal quality requirements include WLAN network signal quality requirements, 5G NR network signal quality requirements, and 4G Long Term Evolution (LTE) network signal quality requirements.
  • the signal quality requirement can also be understood as the minimum requirement for the proportion of time that the signal strength remains above a certain value. For example, the signal strength can reach the threshold requirement at least 90% of the time. It should be understood that in order to ensure the stability of federated learning, some members with stronger signal quality (for example, stronger network signal strength and better stability) are usually selected as participating members of federated learning.
  • the average value, variance, etc. of the received signal strength indicator (Received Signal Strength Indication, RSSI) or the round-trip time delay (RTT) of the signal can be used to represent the signal quality.
  • the traffic range is used to indicate the traffic usage range requirements of the candidate members, which is the traffic value requirements consumed by each member within a preset time period, such as the amount of uploaded and downloaded traffic, etc., for example, it can be the accumulated usage of the UE. It should be understood that in order to reduce the pressure on each participating member when performing federated learning, some members with stable network status and low traffic consumption are usually selected as participants in federated learning.
  • the member type information is used to indicate the type requirements of the candidate members participating in the federated learning.
  • the member type information can be understood as the type of the candidate members participating in the federated learning.
  • it can be a terminal, a core network element (such as NWDAF), or a base station.
  • Area of interest is used to indicate the area where the candidate member is located.
  • the AOI can be the area where the candidate member participating in federated learning is located, or is of concern or interest.
  • Interesting area range which can be longitude and latitude, or one or more cells/tracking areas (tracking areas, TA), etc.
  • Quantity information is used to indicate the quantity requirements of the candidate members.
  • the quantity information can be understood as the number of members participating in federated learning, that is, the number of candidate members that the first network element needs to determine.
  • the first network element can determine the number of candidate network elements.
  • the quantity information may be the required minimum number of candidate members, that is, the determined number of candidate members cannot be less than the required minimum number of candidate members.
  • the quantity information may also be the required maximum number of candidate members, that is, the determined number of candidate members cannot be greater than the required maximum number of candidate members.
  • the type information of federated learning is used to indicate whether the federated learning belongs to vertical federation or horizontal federation.
  • the above request message may also include instruction information, which is used to instruct the execution of a determination task for candidate members participating in federated learning, that is, indicating that this task is for federated learning or selected by a member of federated learning.
  • the first network element can learn based on the request message or based on the instruction information in the request message that it needs to perform the determined task of the candidate members participating in the federated learning.
  • Step 202 The first network element determines one or more first devices as candidate members that can participate in federated learning based on the screening information.
  • the first network element after receiving the request message, the first network element will filter from multiple second devices based on the filtering information in the request message, thereby determining one or more first devices that meet the filtering information. And identify these first devices as candidate members that can participate in federated learning.
  • the above-mentioned first device includes at least one of the following: a first device that is connected to the Internet within a time period; a first device that supports federated learning during the time period; a first device that supports algorithm type; a first device that supports utilization The first device that performs federated learning according to the algorithm type, the first device whose model training accuracy information is greater than the accuracy threshold; the first device that supports the wireless access format; the first device that is under the wireless access format; the first device that supports all The first device that performs federated learning under the above wireless access standard; the first device whose signal quality is greater than the signal quality threshold; the first device whose signal quality is greater than the signal quality threshold requirement; the first device whose traffic consumption is within the traffic range; which meets all requirements The first device required by the type of candidate member; the first device located within the AOI; the type of federated learning and the The type of federated learning included in the filter information is the same as the first device.
  • the first device when the screening information includes a time period, the first device is a first device that is connected to the Internet during the time period, or the first device is a first device that supports federated learning during the time period; in When the filtering information includes an algorithm type, the first device is a first device that supports the algorithm type, or the first device is a first device that supports federated learning using the algorithm type; the accuracy threshold is included in the filtering information.
  • the first device is the first device whose model training accuracy information is greater than the accuracy threshold; in the case where the filtering information includes the wireless access standard, the first device is the first device that supports the wireless access standard, or, The first device is a first device under the wireless access standard, or the first device is a first device that supports federated learning under the wireless access standard; the filtering information includes a signal quality threshold In the case of , the first device is the first device whose signal quality is greater than the signal quality threshold; in the case where the screening information includes signal quality requirements, the first device is the first device whose signal quality is greater than the signal quality requirements; in the screening information If the traffic range is included in the first device, the first device is the first device whose consumed traffic is within the traffic range; if the filtering information includes member type information, the first device is the first device that meets the type requirements of the candidate member.
  • the type of the first device is the same as the member type information indicated in the screening information; in the case where the screening information includes an AOI, the first device is the first device located within the AOI; in If the screening information includes federated learning type information, the federated learning type of the first device is the same as the federated learning type information included in the screening information.
  • the first device is the first device of the device type.
  • the filtering information includes the area information
  • the first device is the first device in the area information.
  • the first device is a device that satisfies the corresponding at least two items of filtering information.
  • the filtering information includes a time period and a supported algorithm type
  • the first device is a first device that is connected to the Internet within the time period and supports the algorithm type.
  • the filtering information includes at least two other pieces of information, it is similar to the case where the time period and the supported algorithm type are included, and will not be described again here.
  • the first device that satisfies the filtering information can be selected from at least one second device as a candidate network element through the filtering information, so that members suitable for participating in federated learning can be selected, which helps to improve the efficiency of federated learning training. efficiency.
  • the first network element may determine the data type corresponding to the filtering information, obtain corresponding attribute information of at least one device from at least one third network element based on the data type, and obtain the corresponding attribute information of at least one device based on the attribute information and Screen information and identify candidate members.
  • the above data type includes at least one of the following networking information and the corresponding time period, algorithm type, wireless access standard, signal quality and traffic.
  • the filtering information when the filtering information includes the time period, the data type corresponding to the filtering information is the networking information and the time period corresponding to the networking state.
  • the data type corresponding to the filtering information is the network information and the time period corresponding to the networking state.
  • the corresponding data type is the algorithm type.
  • the filtering information includes the precision threshold, the data type corresponding to the filtering information is precision information.
  • the filtering information includes the wireless access standard
  • the data type corresponding to the filtering information is a wireless access standard.
  • the filtering information includes a signal quality threshold, the data type corresponding to the filtering information is signal quality.
  • the filtering information includes a traffic range, the data type corresponding to the filtering information is traffic.
  • the filtering information includes member types, the data type corresponding to the filtering information is the device type.
  • the data type corresponding to the filtering information is location information.
  • the filtering information included in the request message is different, that is to say, the filtering information is related to the member type.
  • the filtering information may include at least one of a time period, a supported algorithm type, an accuracy threshold, a wireless access standard and a signal quality threshold, a traffic range, and area information.
  • the filtering information may include at least one of a supported algorithm type and an accuracy threshold.
  • the first network element After determining the data type, the first network element will obtain corresponding attribute information of at least one second device from at least one third network element based on the data type.
  • the third network element includes multiple different network elements. It should be understood that the attribute information that needs to be obtained is different, and the first network element will obtain data from different third network elements. Get attribute information from the element.
  • the first network element obtains information such as the time when the terminal connects to the WLAN and whether to connect to the WLAN from the session management function (SMF) of the third network element.
  • the first network element obtains information such as the signal quality of the connected WLAN from the third network element.
  • the third network element may be a network management device, for example, it may be an operation administration and maintenance (OAM).
  • OAM operation administration and maintenance
  • the network element unified data management function (UDM) or the third network element data collection application function (DC-AF) obtains the algorithms and accuracy supported by each candidate network element, and obtains the algorithms and accuracy supported by each candidate network element from the third network element.
  • the user plane function (UPF) obtains the terminal's traffic information, etc., or the first network element obtains the UE accumulated usage information from the SMF or the charging function (Charging Function, CHF).
  • the attribute information is the current networking status corresponding to at least one device, and the time it is in the networking status, such as whether it is connected to the network and at what time. The segment is in a networked state, etc.
  • the attribute information is the currently supported algorithm corresponding to at least one device, such as whether it supports a deep learning algorithm, whether it supports a linear regression algorithm, etc.
  • the attribute information is the accuracy currently supported by at least one device, such as the accuracy that the model can achieve after each device trains the model.
  • the attribute information is the network standard type currently accessed by at least one device, such as whether the device is currently connected to a WLAN or a 5G network.
  • the attribute information is the signal quality of the network currently accessed by at least one second device, or the signal quality of the network accessed within a preset time period.
  • the attribute information is the traffic consumed by each of the at least one second device in a preset time period.
  • the attribute information is a corresponding type of at least one second device, such as a terminal, a base station or a core network device.
  • the attribute information is the respective location of at least one second device, such as the current geographical location, or the cell or TA where it is located, etc.
  • the attribute information corresponding to at least one second device and the filter is matched to determine candidate members based on the screening information. For example, the first device that meets the screening information may be determined as a candidate member.
  • the filtering information includes that the device type is terminal, the area is cell A, the time period is A, and the wireless access standard is WiFi connected to non-3GPP. Then the first network element will obtain the corresponding attribute information based on these filtering information, and based on the obtained attribute information, select the terminal that is in the A cell, is connected to the network during the A time period, and is connected to non-3GPP WiFi as a candidate member.
  • the first network element obtains the data type corresponding to the filtering information and obtains the attribute information of each second device from at least one third network element, thereby determining the candidate member based on the attribute information and the filtering information. , because candidate members suitable for participating in federated learning can be determined from at least one second device through preset screening information, thereby improving the efficiency of federated learning.
  • the first network element can directly determine the devices that meet the screening information as candidate members, and by default, these candidate network elements that meet the screening information can participate in federated learning. The first network element can select these devices that meet the screening information. The identifiers of the candidate members of the filtered information are sent to the second network element.
  • the first network element in order to screen out more suitable participants, when determining candidate members based on attribute information and screening information, can also screen based on the willingness information of each second device. For example, the first network element may obtain the first indication information from the third network element, and determine one or more first devices from the second device as candidate members based on the filtering information and the first indication information, The above-mentioned first instruction information is used to represent the willingness information of each device to participate in federated learning, where the willingness information represents whether each device is willing to participate in federated learning.
  • the above-mentioned first device is a device that satisfies filtering information and is willing to participate in federated learning.
  • each device corresponds to its own first indication information.
  • the corresponding first indication information when the corresponding first indication information is 0, it means that the device is willing to participate in federated learning.
  • the first indication information is 1, when the first indication information is 1, it means that the device is unwilling to participate in federated learning.
  • the first indication information when the first indication information is 1, it means that the device is willing to participate in federated learning.
  • the first indication information is 0, it means that the device is willing to participate in federated learning.
  • the device is unwilling to participate in federated learning.
  • the first instruction information may also use other numerical values to represent information on the willingness of each device to participate in federated learning.
  • the first network element After obtaining the first instruction information, the first network element obtains the attribute information and filtering information of each device Matching is performed, and devices that meet the screening information and whose first instruction information indicates that they are willing to participate in federated learning are determined as candidate members.
  • the first network element selects devices that meet the screening information and are willing to participate in federated learning as candidate members based on the screening information and the first instruction information, thereby helping to improve the efficiency of subsequent federated learning.
  • the first network element when the first network element determines one or more first devices as candidate members from the second device based on the screening information and the first indication information, the first network element can also obtain each second device from the third network element. Participate in federated learning of capability information to determine candidate members based on screening information, first instruction information and capability information.
  • the capability information includes at least one of the following information: the wireless access standard participating in federated learning, the area participating in federated learning, the time of participating in federated learning, the algorithm information that can be supported by participating in federated learning, and the capabilities that can be achieved by participating in federated learning. Accuracy information and types of participating federated learning.
  • the type of participating in federated learning can include willingness to participate in horizontal federated learning, or willingness to participate in vertical federated learning.
  • the first network element determines one or more first devices as candidate members that can participate in federated learning based on the screening information, including:
  • the first network element obtains the willingness information of the first device to participate in the federated learning from the third network element;
  • the first network element determines, based on the willingness information and the screening information, the first device that is willing to participate in the federated learning and matches the screening information as a candidate member that can participate in the federated learning.
  • the first network element determines one or more first devices as candidate members that can participate in federated learning based on the screening information, including:
  • the first network element obtains federated learning capability information of the first device.
  • the capability information includes at least one of the following information: a wireless access standard participating in federated learning, a region participating in federated learning, and a network participating in federated learning. Time, algorithm information that can be supported by participating in federated learning, accuracy information that can be achieved by participating in federated learning, and the type of participating in federated learning;
  • the first network element determines from the first device according to the capability information and the screening information. Candidate members who can participate in federated learning are determined, and the ability information of the candidate members matches the screening information.
  • the first network element can obtain the federated learning capability information of the first device from the fourth network element, where the fourth network element can be at least one of the following: NRF, UDM, data collection coordination function ( Data Collection Coordination Function (DCCF), AMF.
  • NRF Network Radio Resource Management Function
  • UDM Data Collection Coordination Function
  • DCCF Data Collection Coordination Function
  • the ability information of the candidate member matches the screening information, including at least one of the following:
  • the wireless access standard of the candidate member participating in federated learning is the same as the wireless access standard included in the screening information;
  • the area where the candidate member participates in federated learning is located within the AOI included in the screening information
  • the time when the candidate member participates in federated learning is within the time period included in the screening information
  • the algorithm types that can be supported by the candidate members participating in federated learning are included in the algorithm types included in the screening information;
  • the accuracy information that the candidate members can achieve by participating in federated learning is higher than the accuracy threshold included in the screening information
  • the type of federated learning that the candidate members participate in is the same as the type of federated learning included in the screening information.
  • the first network element can also obtain the capability information of each device to participate in federated learning from the third network element, thereby matching the attribute information and filtering information of each device, and indicating the first instruction information to those who are willing to participate in federated learning.
  • Devices whose ability information to participate in federated learning satisfy the screening information are determined as candidate members, thereby improving the efficiency of subsequent federated learning.
  • the ability information to participate in federated learning that meets the screening information includes:
  • the wireless access standard of the candidate members participating in federated learning is the same as the wireless access standard included in the screening information;
  • the time when the candidate member participates in federated learning is within the time period included in the screening information
  • the algorithm types that can be supported by the candidate members participating in federated learning are included in the algorithm types included in the screening information;
  • the accuracy information that can be achieved by the candidate members participating in federated learning is higher than that contained in the screening information. Included accuracy threshold;
  • the type of federated learning that the candidate members participate in is the same as the type of federated learning included in the screening information.
  • the first network element filters out 80 devices that intend to participate in federated learning based on the first instruction information, and the first network element Based on the algorithm information that these 80 devices can support when participating in federated learning obtained from the third network element, further screening can be performed. For example, 50 of the devices can support both deep learning algorithms and Linear regression algorithm. If 30 devices only support deep learning algorithms when participating in federated learning, the first network element can further filter out those devices that can support both deep learning algorithms and linear regression based on the types of algorithms that can be supported when participating in federated learning. The first device in the regression algorithm is used as a candidate network element.
  • the implementation method is similar to the implementation method when the capability information includes the algorithm type that can be supported by participating in federated learning, and will not be described again here.
  • the first network element can obtain the willingness information and capability information of each device to participate in federated learning from third network elements such as UDM/DCAF/NRF/, where UDM/DCAF/NRF/ is a network element with storage capabilities.
  • the request message sent by the second network element also includes second indication information, where the second indication information is used to indicate the service type corresponding to federated learning.
  • the first network element may determine one or more first devices as candidate members from a plurality of second devices based on the screening information and the second indication information, wherein the first device can support The service type corresponds to the device of the service.
  • the second indication information may be an analytic ID, which is used to indicate the business type corresponding to federated learning.
  • the analytic ID can be: "UE mobility”, which is used to indicate that this task is an analysis task related to UE mobility, and "NF load”, which is used to indicate This task is an analysis task related to network element load.
  • the selected candidate members may be different. Therefore, when the first network element selects candidate members, it can also select from multiple candidates based on the filtering information and the business type indicated by the analytic ID. Filter the second device to support services corresponding to the service type.
  • the first device serves as a candidate member. For example, if the second indication information is used to indicate that the service type corresponding to federated learning is analysis tasks related to terminal mobility, then the selected first device is a device that can support analysis tasks related to terminal mobility.
  • candidate members are further screened according to the business types corresponding to federated learning, thereby improving the accuracy of the selected candidate members, improving the training efficiency of federated learning and the accuracy of the trained model.
  • the first network element determines one or more first devices as candidate members that can participate in federated learning based on the screening information, including:
  • the first network element obtains network status information corresponding to the first device, and the network status information includes at least one of the following:
  • Type information of the first device
  • the location information of the first device is the location information of the first device
  • the first network element determines candidate members that can participate in federated learning from the first device based on the network status information corresponding to the first device and the screening information, and the network status information corresponding to the candidate members is consistent with the The above filter information matches.
  • the first network element can obtain the network status information corresponding to the first device from the third device, and the third device can be AMF, UDM, NRF, PCF, RAN or operation management and maintenance equipment (OAM Operation Administration). and Maintenance, OAM).
  • 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 wireless access standard information of the first device from the AMF or PCF, or the first network element may obtain the type information of the first device from the AMF or PCF.
  • the network element can obtain the wireless signal quality information of the first device from the RAN or OAM, or the first network can obtain the UE accumulated usage information from the SMF or CHF.
  • the network status information corresponding to the candidate member matches the screening information, including at least one of the following:
  • the wireless access system in which the candidate member is located and the wireless access system included in the screening information The formula is the same;
  • the wireless signal quality information of the candidate member is greater than the signal quality requirement included in the screening information
  • the position of the candidate member is located within the AOI included in the screening information
  • the type of the candidate member includes member type information included in the screening information.
  • the above request message also includes sorting indication information, which is used to instruct the candidate members to be sorted according to the second information, wherein the second information includes at least one of the following: signal quality of the candidate members, candidate members accuracy information and traffic information of candidate members; when the request message includes sorting indication information, the identification order of candidate members in the response message is the order obtained by sorting according to the second information.
  • the first network element when the request message includes sorting indication information, after determining the candidate members, the first network element will sort the determined candidate members in ascending or descending order according to the second information in the sorting indication information.
  • the candidate members can be sorted in ascending or descending order according to their signal quality, or they can be sorted in ascending or descending order according to their accuracy information, or they can also be sorted in ascending or descending order according to their traffic information.
  • the identification order of the candidate members in the response message is the order obtained by sorting according to the second information.
  • the first network element sorts the determined candidate members according to the second information.
  • the second network element will sort according to the sorting information.
  • the information further selects candidate members that better meet their own requirements to participate in subsequent federated learning, which can improve the efficiency of federated learning.
  • the above request message also includes grouping indication information, which is used to instruct the candidate members to be grouped according to third information, wherein the third information includes at least one of the following: the area where the candidate members are located, the candidate The time period during which the member is in the networking state, the algorithm type supported by the candidate member, the accuracy information of the candidate member, the wireless access standard of the candidate member, and the signal quality of the candidate member; if the request message includes grouping indication information, the response message The identity of the candidate member in is the identity obtained after sorting according to the third information.
  • the first network element when the request message includes grouping indication information, after determining the candidate members, the first network element will group the determined candidate members according to the third information in the grouping indication information.
  • candidate members can be grouped according to the region where they are located, such as grouping candidate members in the same region into one group.
  • the candidate members can be grouped according to the time period during which they are online, for example, candidate members who are online at the same time period are grouped into one group, or they can be grouped according to the algorithm type supported by the candidate members, for example, the candidate members can be grouped according to the algorithm type they support.
  • the network elements of the candidate members can be divided into a group, or they can be grouped according to the accuracy information of the candidate members, such as grouping candidate members with the same accuracy information into a group, or they can also be grouped according to the wireless access standard of the candidate members, such as grouping wireless
  • Candidate members with the same access mode are grouped into one group, or they can be grouped according to the signal quality of the candidate members, or they can be grouped according to the traffic information of the candidate members, or they can also be grouped according to the time period during which the candidate members can perform federated learning. For example, candidate members who can perform federated learning in the same time period are grouped into a group, for example, from 10 o'clock to 12 o'clock during the day, and so on.
  • the identifier of the candidate member in the response message is an identifier obtained by grouping according to the third information.
  • the first network element groups the determined candidate members according to the third information.
  • the second network element after receiving the identification information of the candidate members sent by the first network element, the second network element will group the determined candidate members according to the grouping information.
  • the information further selects candidate members that better meet their own requirements to participate in subsequent federated learning, which can improve the efficiency of federated learning.
  • Step 203 The first network element sends a response message to the second network element, where the response message includes the identity of the candidate member.
  • the first network element may carry the identities of these candidate members in the response message and send it to the second network element.
  • the identity of the candidate member may include the candidate member's Subscriber Permanent Identifier (SUPI) or Internet Protocol (IP).
  • the identity of the candidate member may also be a Generic Public Subscription Identifier (Generic Public Subscription Identifier). , GPSI), International Mobile Subscriber Identity (IMSI), AF specific UE ID (AF specific UE ID) or the UE IP address.
  • the above response message also includes at least one of the following information: the area where the candidate member is located, the time period during which the candidate member is in the network state, the algorithm type supported by the candidate member, the accuracy information of the candidate member, the candidate member's The wireless access standard, the signal quality of the candidate members, the traffic information of the candidate members, and the time period during which the candidate members can perform federated learning.
  • the area where the candidate member is located may include the location of the candidate member, such as longitude, latitude or cell.
  • the time period during which candidate members can perform federated learning is the prediction result analyzed by the first network element based on the willingness information and capability information of each device.
  • the signal quality of candidate members, when connected to a WLAN can be expressed by the average, variance or RTT of RSSI.
  • the traffic information of candidate members may include the traffic value consumed within a preset time period, such as the amount of uploaded and downloaded traffic, and other information.
  • the above response message may also include second indication information, and the second indication information may be an analytic ID, which is used to indicate the business type corresponding to federated learning.
  • the above response message may also include the proportion of coverage time in the network state, for example, how long a day the network is in the wifi connection state.
  • the above response message may also include address information of the candidate member, and the second network element may find the candidate member based on the address information for connection and federated learning.
  • the above response message may also include quantity information, which may be understood as the number of members participating in federated learning, that is, the number of candidate members that the first network element needs to determine.
  • the second network element After receiving the response message sent by the first network element, the second network element will select the target members who actually participate in the federated learning based on the actual federated learning training situation.
  • the second network element can determine the members to be federated learning based on the signal quality of the candidate members in the response message, and can select the top 100 members with the best signal quality for federated learning.
  • the method of determining the target member based on other information in the response message is similar to the method of determining the target member based on the signal quality, and will not be described again here.
  • the second network element can directly use the target number of candidate members as target members. For example, if the request message includes a request for 50 members for federated learning, and the response message returned by the first network element contains 50 candidate members, the second network element can directly select these 50 members as target members.
  • the second network element will connect with the target members according to the identification information of the target members to perform federated learning.
  • the second network element can further screen the target members who actually participate in federated learning based on the candidate members returned by the first network element, thereby making the screened target members more suitable for participating in federated learning and improving federated learning. s efficiency.
  • the method for determining candidate members determines one or more first devices as candidates that can participate in federated learning by receiving a request message sent by the second network element and based on the screening information included in the request message. member, and sends the identification information of the candidate member to the second network element, where the screening information includes at least one of the following: time period, algorithm type, accuracy threshold, wireless access format, signal quality requirements, traffic range, member type information, Quantity information, regional information, and type information for federated learning.
  • the first network element can select participating members suitable for federated learning, thereby improving the efficiency of federated learning.
  • Figure 3 is a schematic flowchart of another method for determining candidate members provided by this application.
  • the execution subject is the second network element.
  • the method includes:
  • Step 301 The second network element sends a request message to the first network element.
  • the request message includes screening information, and the request message is used to instruct the first network element to determine one or more first devices as candidate members that can participate in federated learning based on the screening information.
  • the screening information includes at least one of the following:
  • Algorithm type used to indicate the type of algorithm that needs to be supported for model training for federated learning
  • Accuracy threshold used to indicate the accuracy requirements of model training that need to be met to perform the federated learning
  • Wireless access standard used to indicate the wireless access standard to be selected for the federated learning
  • Signal quality requirements used to indicate wireless signal quality requirements when performing the federated learning
  • Traffic range used to indicate the traffic usage range requirements of the candidate member
  • Quantity information used to indicate the quantity requirements of the candidate members
  • Area of interest AOI used to indicate the area where the candidate member is located
  • the type information of federated learning is used to indicate whether the federated learning belongs to vertical federation or horizontal federation.
  • Step 302 The second network element receives the response message sent by the first network element, where the response message includes the identity of the candidate member.
  • the first device includes at least one of the following:
  • the first device that is connected to the Internet during the said time period
  • the first device to support federated learning during the stated time period
  • the first device to support federated learning utilizing the described algorithm type
  • the first device in the wireless access standard is the first device in the wireless access standard
  • the first device that supports federated learning under the wireless access standard
  • the first device whose signal quality is greater than the signal quality requirement
  • the first device whose flow rate is within the flow range
  • the first device located within the AOI
  • the first device has the same federated learning type information as the federated learning type information included in the screening information.
  • the request message also includes: indication information, the indication information is used to instruct to perform a certain task of candidate members participating in federated learning, or to indicate that this task is for federated learning or members of federated learning. Selected.
  • the request message also includes sorting indication information.
  • the sorting indication information is used to indicate sorting the candidate members according to the second information.
  • the second information includes at least one of the following: signal quality of the candidate members, accuracy information of the candidate members, and Traffic information of candidate members;
  • the identifier of the candidate member in the response message is the identifier obtained after sorting according to the second information.
  • the request message also includes grouping indication information.
  • the grouping indication information is used to instruct the candidate members to be grouped according to the third information.
  • the third information includes at least one of the following information: the area where the candidate members are located, the candidate members' The time period during which the candidate member is in the network state, the algorithm type supported by the candidate member, the accuracy information of the candidate member, the wireless access standard of the candidate member, and the signal quality of the candidate member;
  • the identification of the candidate members in the response message is the identification obtained after grouping according to the third information.
  • the response message also includes the area where the candidate member is located, the time period during which the candidate member is in the networking state, the algorithm type supported by the candidate member, the accuracy information of the candidate member, the wireless access standard of the candidate member, and the Signal quality, traffic information of candidate members, and the time period during which candidate members are available for federated learning.
  • the method also includes:
  • the second network element determines the target member to participate in federated learning based on the identification of the candidate member.
  • the method for determining candidate members receives a request message sent by the second network element, and determines one or more first devices as candidate members that can participate in federated learning based on the screening information included in the request message. , and sends the 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 standard, signal quality requirements, traffic range, member type information, quantity information, regional information, and type information for federated learning.
  • the first network element can select participating members suitable for federated learning, thereby improving the efficiency of federated learning.
  • the second network element can further screen out the members that are truly suitable for participating in federated learning based on the candidate members sent by the first network element, thereby improving the efficiency of federated learning.
  • Figure 4 is a signaling diagram of a method for determining candidate members provided by this application. As shown in Figure 4, the method includes:
  • Step 401 The second network element sends a request message to the first network element.
  • the request message includes filtering information, and the filtering information includes at least one of the following: time period, algorithm type, accuracy threshold, wireless access standard, signal quality requirements, traffic range, member type information, quantity information, regional information, and federated learning. Type information.
  • Step 402 The first network element determines one or more first devices as candidate members that can participate in federated learning based on the screening information.
  • Step 403 The first network element sends a response message to the second network element, where the response message includes the identity of the candidate member.
  • the method for determining candidate members receives a request message sent by the second network element, and determines one or more first devices as candidate members that can participate in federated learning based on the screening information included in the request message. , and sends the 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 standard, signal quality requirements, traffic range, member type information, quantity Types of information, area of interest AOI and federated learning.
  • the first network element can select participating members suitable for federated learning, thereby improving the training efficiency of federated learning.
  • the execution subject may be a device for determining candidate members.
  • the device for determining candidate members performed by the device for determining candidate members is used as an example to illustrate the device for determining candidate members provided by the embodiment of the present application.
  • FIG. 5 is one of the structural schematic diagrams of a device for determining candidate members provided by this application. As shown in Figure 5, the device for determining candidate members provided by this embodiment includes:
  • the receiving module 11 is configured to receive a request message sent by the second network element, where the request message includes screening information;
  • the processing module 12 is configured to determine one or more first devices as candidate members that can participate in federated learning based on the screening information;
  • Sending module 13 configured to send a response message to the second network element, where the response message includes the identity of the candidate member
  • the screening information includes at least one of the following:
  • Algorithm type used to indicate the type of algorithm that needs to be supported for model training for federated learning
  • Accuracy threshold used to indicate the accuracy requirements of model training that need to be met to perform the federated learning
  • Wireless access standard used to indicate the wireless access standard to be selected for the federated learning
  • Signal quality requirements used to indicate wireless signal quality requirements when performing the federated learning
  • Traffic range used to indicate the traffic usage range requirements of the candidate member
  • Quantity information used to indicate the quantity requirements of the candidate members
  • Area of interest AOI used to indicate the area where the candidate member is located
  • the type information of federated learning is used to indicate whether the federated learning belongs to vertical federation or horizontal federation.
  • the receiving module receives a request message sent by the second network element, and the request message includes screening information; the processing module determines one or more first devices as candidate members that can participate in federated learning based on the screening information; send The module sends a response message to the second network element, where the response message includes the identification of the candidate member, wherein the screening information includes at least one of the following: time period, algorithm type, accuracy threshold, wireless access standard, signal quality Requirements, traffic range, member type information, quantity information, area of interest AOI and type of federated learning.
  • the device for determining candidate members can select participating members suitable for federated learning by filtering information, thereby improving the training efficiency of federated learning.
  • the first device includes at least one of the following:
  • the first device that is connected to the Internet during the said time period
  • the first device to support federated learning during the stated time period
  • the first device to support federated learning utilizing the described algorithm type
  • the first device in the wireless access standard is the first device in the wireless access standard
  • the first device that supports federated learning under the wireless access standard
  • the first device whose signal quality is greater than the signal quality requirement
  • the first device whose flow rate is within the flow range
  • the first device located within the AOI
  • the first device has the same federated learning type information as the federated learning type information included in the screening information.
  • the processing module 12 is specifically used for:
  • the candidate members are determined based on the attribute information and the screening information.
  • the data type includes at least one of the following:
  • Networking information and corresponding time period algorithm type, accuracy information, wireless access standard, signal quality and traffic.
  • processing module 12 is specifically used for:
  • the first indication information is used to indicate the willingness information of each second device to participate in federated learning; wherein the willingness information indicates whether each second device is willing to participate in federated learning. ;
  • One or more first devices are determined as the candidate members according to the screening information and the first instruction information, and the first devices are devices among the second devices that are willing to participate in federated learning and satisfy the screening information. .
  • processing module 12 is specifically used for:
  • the first device that is willing to participate in the federated learning and matches the screening information is determined as a candidate member that can participate in the federated learning.
  • processing module 12 is specifically used for:
  • the capability information includes at least one of the following information: a wireless access standard participating in federated learning, a region participating in federated learning, a time participating in federated learning, and a location participating in federated learning.
  • the algorithm information that can be supported, the accuracy information that can be achieved by participating in federated learning, and the type of participating in federated learning;
  • Candidate members who can participate in federated learning are determined from the first device according to the capability information and the screening information, and the capability information of the candidate members matches the screening information.
  • the capability information of the candidate member matches the screening information, including at least one of the following:
  • the wireless access standard of the candidate member participating in federated learning is the same as the wireless access standard included in the screening information;
  • the area where the candidate member participates in federated learning is located within the AOI included in the screening information
  • the time when the candidate member participates in federated learning is within the time period included in the screening information
  • the algorithm types that can be supported by the candidate members participating in federated learning are included in the algorithm types included in the screening information;
  • the accuracy information that the candidate members can achieve by participating in federated learning is higher than the accuracy threshold included in the screening information
  • the type of federated learning that the candidate members participate in is the same as the type of federated learning included in the screening information.
  • processing module 12 is specifically used for:
  • the network status information includes at least one of the following:
  • Type information of the first device
  • the location information of the first device is the location information of the first device
  • candidate members that can participate in federated learning are determined from the first device, and the network status information corresponding to the candidate members matches the screening information.
  • the network status information corresponding to the candidate member matches the screening information, including at least one of the following:
  • the wireless access standard of the candidate member is the same as the wireless access standard 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 position of the candidate member is located within the AOI included in the screening information
  • the type of the candidate member includes member type information included in the screening information.
  • processing module is specifically used for:
  • the capability information includes at least one of the following information: the wireless access standard participating in federated learning, the area participating in federated learning, the time of participating in federated learning, and the capabilities of participating in federated learning. Supported algorithm information, accuracy information that can be achieved by participating in federated learning, and types of participating in federated learning; one or more first devices are determined to be the required ones based on the screening information, the first instruction information, and the capability information.
  • the first device is a device among the second devices that is willing to participate in federated learning and whose federated learning capability information satisfies the screening information.
  • the ability information to participate in federated learning that meets the screening information includes:
  • the wireless access standard of the candidate members participating in federated learning is the same as the wireless access standard included in the screening information;
  • the time when the candidate member participates in federated learning is within the time period included in the screening information
  • the algorithm types that can be supported by the candidate members participating in federated learning are included in the algorithm types included in the screening information;
  • the accuracy information that can be achieved by the candidate members participating in federated learning is higher than that contained in the screening information. Included accuracy threshold;
  • the request message also includes second indication information, where the second indication information is used to indicate the service type corresponding to federated learning;
  • the processing module is specifically used for:
  • the first device is a device that can support services corresponding to the service type.
  • the request message also includes sorting indication information, the sorting indication information is used to instruct the candidate members to be sorted according to second information, and the second information includes at least one of the following: the candidate members The signal quality, the accuracy information of the candidate member and the traffic information of the candidate member;
  • the identification order of the candidate members in the response message is the order obtained by sorting according to the second information.
  • the request message also includes grouping indication information, and the grouping indication information is used to instruct the candidate members to be grouped according to third information, where the third information includes at least one of the following information:
  • the identifiers of the candidate members in the response message are identifiers obtained after grouping according to the third information.
  • the response message also includes at least one of the following information: the area where the candidate member is located, the time period during which the candidate member is in the networking state, the algorithm type supported by the candidate member, the The accuracy information of the candidate member, the wireless access standard of the candidate member, the signal quality of the candidate member, the traffic information of the candidate member, and the time period during which the candidate member can perform federated learning.
  • the device of this embodiment can be used to execute the method of any of the foregoing first network element side method embodiments. Its specific implementation process and technical effects are similar to those in the first network element side method embodiment. For details, please refer to Chapter 1 The detailed introduction of the method embodiment on the network element side will not be described again here.
  • FIG. 6 is the second structural schematic diagram of the device for determining candidate members provided by this application. As shown in Figure 6, the device for determining candidate members provided by this embodiment includes:
  • the sending module 21 is configured to send a request message to the first network element.
  • the request message includes filtering information.
  • the request message is used to instruct the first network element to send one or more first devices to the first network element according to the filtering information. Identified as candidate members who can participate in federated learning;
  • the receiving module 22 is configured to receive a response message sent by the first network element, where the response message includes the identification of the candidate member;
  • the screening information includes at least one of the following:
  • Algorithm type used to indicate the type of algorithm that needs to be supported for model training for federated learning
  • Accuracy threshold used to indicate the accuracy requirements of model training that need to be met to perform the federated learning
  • Wireless access standard used to indicate the wireless access standard to be selected for the federated learning
  • Signal quality requirements used to indicate wireless signal quality requirements when performing the federated learning
  • Traffic range used to indicate the traffic usage range requirements of the candidate member
  • Quantity information used to indicate the quantity requirements of the candidate members
  • Area of interest AOI used to indicate the area where the candidate member is located
  • the type information of federated learning is used to indicate whether the federated learning belongs to vertical federation or horizontal federation.
  • the first device includes at least one of the following:
  • the first device that is connected to the Internet during the said time period
  • the first device to support federated learning during the stated time period
  • the first device to support federated learning utilizing the described algorithm type
  • the first device in the wireless access standard is the first device in the wireless access standard
  • the first device that supports federated learning under the wireless access standard
  • the first device whose signal quality is greater than the signal quality requirement
  • the first device whose flow rate is within the flow range
  • the first device located within the AOI
  • the first device has the same federated learning type information as the federated learning type information included in the screening information.
  • the request message also includes: indication information, the indication information is used to instruct to perform a certain task of candidate members participating in federated learning, or to indicate that this task is for federated learning or members of federated learning. Selected.
  • the request message also includes sorting indication information, the sorting indication information is used to instruct the candidate members to be sorted according to second information, and the second information includes at least one of the following: the candidate members The signal quality, the accuracy information of the candidate member and the traffic information of the candidate member;
  • the identifier of the candidate member in the response message is the identifier obtained by sorting according to the second information.
  • the request message also includes grouping instruction information, and the grouping instruction information is used to instruct the candidate members to be grouped according to third information, where the third information includes at least one of the following information: The area where the candidate member is located, the time period during which the candidate member is in the networking state, the algorithm type supported by the candidate member, the accuracy information of the candidate member, the wireless access standard of the candidate member and the candidate member Member signal quality;
  • the identifiers of the candidate members in the response message are identifiers obtained after grouping according to the third information.
  • the response message also includes the area where the candidate member is located, the time period during which the candidate member is in the networking state, the algorithm type supported by the candidate member, the accuracy information of the candidate member, and the The wireless access standard of the candidate member, the signal quality of the candidate member, all The traffic information of the candidate members and the time period during which the candidate members can perform federated learning.
  • the device further includes: a processing module 23;
  • the processing module 23 is configured to determine target members participating in federated learning based on the identification of the candidate members.
  • the device of this embodiment can be used to execute the method of any of the foregoing second network element side method embodiments. Its specific implementation process and technical effects are similar to those in the second network element side method embodiment. For details, please refer to Chapter 1 The detailed introduction in the method embodiment on the second network element side will not be described again here.
  • the device for determining candidate members in the embodiment of the present application may be an electronic device, such as an electronic device with an operating system, or may be a component in the electronic device, such as an integrated circuit or chip.
  • the electronic device may be a terminal or other devices other than the terminal.
  • terminals may include but are not limited to the types of terminals 11 listed above, and other devices may be servers, network attached storage (Network Attached Storage, NAS), etc., which are not specifically limited in the embodiment of this application.
  • NAS Network Attached Storage
  • the device for determining candidate members provided by the embodiments of the present application can implement each process implemented by the method embodiments of Figures 2 to 4 and achieve the same technical effect. To avoid duplication, details will not be described here.
  • this embodiment of the present application also provides a communication device 700, which includes a processor 701 and a memory 702.
  • the memory 702 stores programs or instructions that can be run on the processor 701, for example.
  • the communication device 700 is a terminal, when the program or instruction is executed by the processor 701, each step of the above candidate member determination method embodiment is implemented, and the same technical effect can be achieved.
  • the communication device 700 is a network-side device, when the program or instruction is executed by the processor 701, each step of the above candidate member determination method embodiment is implemented, and the same technical effect can be achieved. To avoid duplication, the details are not repeated here.
  • An embodiment of the present application also provides a first network element, including a processor and a communication interface.
  • the processor is configured to determine one or more first devices as candidate members that can participate in federated learning based on the screening information.
  • the communication interface is configured to receive the third The request message sent by the second network element sends a response message to the second network element.
  • the response message includes the identification of the candidate member, 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 standard, signal quality requirements, traffic range, member type information, quantity information, and area information and federated learning type information.
  • This terminal embodiment corresponds to the above-mentioned terminal-side method embodiment. Each implementation process and implementation manner of the above-mentioned method embodiment can be applied to this terminal embodiment, and can achieve the same technical effect.
  • FIG. 8 is a schematic diagram of the hardware structure of a first network element that implements an embodiment of the present application.
  • the first network element 800 includes but is not limited to: radio frequency unit 801, network module 802, audio output unit 803, input unit 804, sensor 805, display unit 806, user input unit 807, interface unit 808, memory 809 and processor 810 At least some parts of etc.
  • the first network element 800 may also include a power supply (such as a battery) that supplies power to various components.
  • the power supply may be logically connected to the processor 810 through a power management system, thereby managing charging, discharging, and And functions such as power consumption management.
  • the structure of the first network element shown in Figure 8 does not constitute a limitation on the first network element.
  • the first network element may include more or less components than shown in the figure, or combine certain components, or arrange different components. , which will not be described in detail here.
  • the input unit 804 may include a graphics processing unit (Graphics Processing Unit, GPU) 8041 and a microphone 8042.
  • the graphics processor 8041 is responsible for the image capture device (GPU) in the video capture mode or the image capture mode. Process the image data of still pictures or videos obtained by cameras (such as cameras).
  • the display unit 806 may include a display panel 8061, which 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 a touch panel 8071 and at least one of other input devices 8072 . Touch panel 8 071, also known as 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 physical keyboards, function keys (such as volume control keys, switch keys, etc.), trackballs, mice, and joysticks, which will not be described again here.
  • the radio frequency unit 801 after receiving downlink data from the network side device, the radio frequency unit 801 can transmit it to the processor 810 for processing; in addition, the radio frequency unit 801 can send uplink data to the network side device.
  • the radio frequency unit 801 includes, but is not limited to, an antenna, amplifier, transceiver, coupler, low noise amplifier, duplexer, etc.
  • Memory 809 may be used to store software programs or instructions as well as various data.
  • the memory 809 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first The storage area can store an operating system, an application program or instructions required for at least one function (such as a sound playback function, an image playback function, etc.), etc.
  • memory 809 may include volatile memory or non-volatile memory, or memory 809 may include both volatile and non-volatile memory.
  • the non-volatile memory can be read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable ROM, PROM), erasable programmable read-only memory (Erasable PROM, EPROM), electrically removable memory. Erase programmable read-only memory (Electrically EPROM, EEPROM) or flash memory.
  • Volatile memory can be random access memory (Random Access Memory, RAM), static random access memory (Static RAM, SRAM), dynamic random access memory (Dynamic RAM, DRAM), synchronous dynamic random access memory (Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDRSDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, ESDRAM), synchronous link dynamic random access memory (Synch link DRAM) , SLDRAM) and direct memory bus random access memory (Direct Rambus RAM, DRRAM).
  • RAM Random Access Memory
  • SRAM static random access memory
  • DRAM dynamic random access memory
  • DRAM synchronous dynamic random access memory
  • SDRAM double data rate synchronous dynamic random access memory
  • Double Data Rate SDRAM Double Data Rate SDRAM
  • DDRSDRAM double data rate synchronous dynamic random access memory
  • Enhanced SDRAM, ESDRAM enhanced synchronous dynamic random access memory
  • Synch link DRAM synchronous link dynamic random access memory
  • SLDRAM direct memory bus
  • the processor 810 may include one or more processing units; optionally, the processor 810 integrates an application processor and a modem processor, where the application processor mainly handles operations related to the operating system, user interface, application programs, etc., Modem processors mainly process wireless communication signals, such as baseband processors. It can be understood that the above modem processor may not be integrated into the processor 810.
  • the radio frequency unit 801 is configured to receive a request message sent by the 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 the identity of the candidate member.
  • Processor 810 configured to determine one or more first devices as candidate members that can participate in federated learning based on screening information, where the screening information includes at least one of the following:
  • Algorithm type used to indicate the type of algorithm that needs to be supported for model training for federated learning
  • Accuracy threshold used to indicate the accuracy requirements of model training that need to be met to perform the federated learning
  • Wireless access standard used to indicate the wireless access standard to be selected for the federated learning
  • Signal quality requirements used to indicate wireless signal quality requirements when performing the federated learning
  • Traffic range used to indicate the traffic usage range requirements of the candidate member
  • Quantity information used to indicate the quantity requirements of the candidate members
  • Area of interest AOI used to indicate the area where the candidate member is located
  • the type information of federated learning is used to indicate whether the federated learning belongs to vertical federation or horizontal federation.
  • the first network element receives the request message sent by the second network element and determines one or more first devices as candidate members that can participate in federated learning based on the screening information included in the request message, and The identification information of the candidate members is sent to the second network element, where the screening information includes at least one of the following: time period, algorithm type, accuracy threshold, wireless access standard, signal quality requirements, traffic range, member type information, quantity information, interest Types of regional AOI and federated learning.
  • the first network element can select participating members suitable for federated learning, thereby improving the training efficiency of federated learning.
  • the first device includes at least one of the following:
  • the first device that is connected to the Internet during the said time period
  • the first device to support federated learning during the stated time period
  • the first device to support federated learning utilizing the described algorithm type
  • the first device in the wireless access standard is the first device in the wireless access standard
  • the first device that supports federated learning under the wireless access standard
  • the first device whose signal quality is greater than the signal quality requirement
  • the first device whose flow rate is within the flow range
  • the first device located within the AOI
  • the type of federated learning is the same as the type information of federated learning included in the screening information.
  • the processor 810 is also used to determine the data type corresponding to the filtering information
  • the candidate members are determined based on the attribute information and the screening information.
  • the data type includes at least one of the following:
  • Networking information and corresponding time period algorithm type, accuracy information, wireless access standard, signal quality and traffic.
  • the processor 810 is also configured to obtain first indication information from the third network element, where the first indication information is used to represent the willingness information of each second device to participate in federated learning; wherein, the willingness The information indicates whether each device is willing to participate in federated learning;
  • one or more first devices are determined as the candidate members, and the first devices are those of the second devices that are willing to participate in federated learning and meet the screening information. equipment.
  • the processor 810 is configured to obtain the willingness information of the first device to participate in the federated learning from the third network element; according to the willingness information and the screening information, determine whether the first device is willing to participate in the federated learning and The first device matching the screening information is determined as a candidate member that can participate in federated learning.
  • the processor 810 is configured to obtain federated learning capability information of the first device, where the capability information includes at least one of the following information: a wireless access standard participating in federated learning, a region participating in federated learning, The time to participate in federated learning, the algorithm information that can be supported by participating in federated learning, the accuracy information that can be achieved by participating in federated learning, and the type of participating in federated learning; according to the ability information and the screening information, from the first device Candidate members who can participate in federated learning are determined, and the ability information of the candidate members matches the screening information.
  • the capability information includes at least one of the following information: a wireless access standard participating in federated learning, a region participating in federated learning, The time to participate in federated learning, the algorithm information that can be supported by participating in federated learning, the accuracy information that can be achieved by participating in federated learning, and the type of participating in federated learning; according to the ability information and the screening information, from the first device Candidate members who can
  • the capability information of the candidate member matches the screening information, including at least one of the following:
  • the wireless access standard of the candidate member participating in federated learning is the same as the wireless access standard included in the screening information;
  • the area where the candidate member participates in federated learning is located within the AOI included in the screening information
  • the time when the candidate member participates in federated learning is within the time period included in the screening information
  • the algorithm types that can be supported by the candidate members participating in federated learning are included in the algorithm types included in the screening information;
  • the accuracy information that the candidate members can achieve by participating in federated learning is higher than the accuracy threshold included in the screening information
  • the type of federated learning that the candidate members participate in is the same as the type of federated learning included in the screening information.
  • 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
  • the location information of the first device is the location information of the first device
  • candidate members that can participate in federated learning are determined from the first device, and the network status information corresponding to the candidate members matches the screening information.
  • the network status information corresponding to the candidate member matches the screening information, including at least one of the following:
  • the wireless access standard of the candidate member is the same as the wireless access standard 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 position of the candidate member is located within the AOI included in the screening information
  • the type of the candidate member includes member type information included in the screening information.
  • the processor 810 is configured to obtain capability information of each second device to participate in federated learning, where the capability information includes at least one of the following information: a wireless access standard participating in federated learning, The area participating in federated learning, the time of participating in federated learning, the algorithm information that can be supported by participating in federated learning, the accuracy information that can be achieved by participating in federated learning, and the type of participating in federated learning; according to the screening information and the first instruction information and the capability information, one or more first devices are determined as the candidate members, and the first device is a device among the second devices that is willing to participate in federated learning and whose federated learning capability information satisfies the screening information.
  • the capability information includes at least one of the following information: a wireless access standard participating in federated learning, The area participating in federated learning, the time of participating in federated learning, the algorithm information that can be supported by participating in federated learning, the accuracy information that can be achieved by participating in federated learning, and the type of participating
  • the ability information to participate in federated learning that meets the screening information includes:
  • the wireless access standard of the candidate members participating in federated learning is the same as the wireless access standard included in the screening information;
  • the time when the candidate member participates in federated learning is within the time period included in the screening information
  • the algorithm types that can be supported by the candidate members participating in federated learning are included in the algorithm types included in the screening information;
  • the accuracy information that the candidate members can achieve by participating in federated learning is higher than the accuracy threshold included in the screening information
  • the type of federated learning that the candidate members participate in is the same as the type of federated learning included in the screening information.
  • the request message also includes second indication information, where the second indication information is used to indicate the service type corresponding to federated learning;
  • the processor 810 is also configured to determine 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 that can support services corresponding to the service type.
  • the request message also includes sorting indication information, the sorting indication information is used to instruct the candidate members to be sorted according to second information, and the second information includes at least one of the following: the candidate members The signal quality, the accuracy information of the candidate member and the traffic information of the candidate member;
  • the identification order of the candidate members in the response message is the order obtained by sorting according to the second information.
  • the request message also includes grouping indication information, and the grouping indication information is used to Instruct the candidate members to be grouped according to third information, where the third information includes at least one of the following information:
  • the identifiers of the candidate members in the response message are identifiers obtained after grouping according to the third information.
  • the response message also includes at least one of the following information: the area where the candidate member is located, the time period during which the candidate member is in the networking state, the algorithm type supported by the candidate member, the The accuracy information of the candidate member, the wireless access standard of the candidate member, the signal quality of the candidate member, the traffic information of the candidate member, and the time period during which the candidate member can perform federated learning.
  • the first network element receives the request message sent by the second network element and determines one or more first devices as candidate members that can participate in federated learning based on the screening information included in the request message, and Send the 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 standard, signal quality requirements, traffic range, member type information, quantity information, Types of area of interest AOI and federated learning.
  • the first network element can select participating members suitable for federated learning, thereby improving the training efficiency of federated learning.
  • the embodiment of the present application also provides a second network element.
  • 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 (CPRI).
  • CPRI common public radio interface
  • the network side device 900 in the embodiment of the present application also includes: instructions or programs stored in the memory 903 and executable on the processor 901.
  • the processor 901 calls the instructions or programs in the memory 903 to execute each of the steps shown in Figure 6.
  • the method of module execution and achieving the same technical effect will not be described in detail here to avoid duplication.
  • An embodiment of the present application also provides a readable storage medium, the readable storage medium stores a program
  • the program or instruction when executed by the processor, implements each process of the candidate member determination method embodiment and can achieve the same technical effect. To avoid duplication, it will not be described again here.
  • the processor is the processor in the terminal described in the above embodiment.
  • the readable storage medium includes computer readable storage media, such as computer read-only memory ROM, random access memory RAM, magnetic disk or optical disk, etc.
  • An embodiment of the present application further provides a chip.
  • the chip includes a processor and a communication interface.
  • the communication interface is coupled to the processor.
  • the processor is used to run programs or instructions to implement the above method for determining candidate members.
  • Each process in the example can achieve the same technical effect. To avoid repetition, we will not repeat it here.
  • chips mentioned in the embodiments of this application may also be called system-on-chip, system-on-a-chip, system-on-chip or system-on-chip, etc.
  • Embodiments of the present application further provide a computer program/program product.
  • the computer program/program product is stored in a storage medium.
  • the computer program/program product is executed by at least one processor to implement the above candidate member determination method.
  • Each process of the embodiment can achieve the same technical effect, so to avoid repetition, it will not be described again here.
  • An embodiment of the present application also provides a system for determining candidate members, including: a first network element and a second network element.
  • the first network element can be used to perform the steps of the method for determining candidate members as described above.
  • the second network element may be configured to perform the steps of the method for determining candidate members as described above.
  • the methods of the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is better. implementation.
  • the technical solution of the present application can be embodied in the form of a computer software product that is essentially or contributes to the existing technology.
  • the computer software product is stored in a storage medium (such as ROM/RAM, disk , CD), including several instructions to cause a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in various embodiments of this application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请公开了一种候选成员的确定方法、装置及设备，属于通信技术领域，本申请实施例的候选成员的确定方法包括：第一网元接收第二网元发送的请求消息，所述请求消息中包括筛选信息；所述第一网元根据所述筛选信息，将一个或多个第一设备确定为可参与联邦学习的候选成员；所述第一网元向所述第二网元发送响应消息，所述响应消息中包括所述候选成员的标识，其中，筛选信息包括如下至少一个：时间段、算法类型、精度阈值、无线接入制式、信号质量要求、流量范围、成员类型信息、数量信息、区域信息和联邦学习的类型信息。

Description

候选成员的确定方法、装置及设备
相关申请的交叉引用
本申请要求于2022年3月28日提交的申请号为202210314821.7，发明名称为“参与联邦学习的成员的确定方法、装置及设备”的中国专利申请的优先权，及要求于2022年4月29日提交的申请号为202210476433.9，发明名称为“候选成员的确定方法、装置及设备”的中国专利申请的优先权，其通过引用方式全部并入本申请。
技术领域
本申请属于通信技术领域，具体涉及一种候选成员的确定方法、装置及设备。
背景技术
联邦学习(federated learning)是指，通过联合不同的参与者(participant，或者party，也称为数据拥有者(data owner)、或者客户(client))进行机器学习建模的方法。在联邦学习中，参与者不需要向其它参与者和协调者(coordinator，也称为服务器(server)，参数服务器(parameter server)，或者聚合服务器(aggregation server))暴露自己所拥有的数据，因而联邦学习可以很好的保护用户隐私和保障数据安全，并可以解决数据孤岛问题。
在将联邦学习应用到通信领域后，如何选择一些合适的成员进行联邦学习来提高训练的效率，是目前亟待解决的问题。
发明内容
本申请实施例提供一种候选成员的确定方法、装置及设备，能够解决由于联邦学习的参与成员不符合要求，导致联邦学习的训练效率较低的问题。
第一方面，提供了一种候选成员的确定方法，包括：
第一网元接收第二网元发送的请求消息，所述请求消息中包括筛选信息；
所述第一网元根据所述筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员；
所述第一网元向所述第二网元发送响应消息，所述响应消息中包括所述候选成员的标识；
其中，所述筛选信息包括如下至少一个：
时间段，用于指示进行所述联邦学习的可选时间段；
算法类型，用于指示进行所述联邦学习需要支持的模型训练的算法类型；
精度阈值，用于指示进行所述联邦学习需满足的模型训练的精度要求；
无线接入制式，用于指示进行所述联邦学习需选择的无线接入制式；
信号质量要求，用于指示进行所述联邦学习时无线信号质量要求；
流量范围，用于指示所述候选成员的流量使用范围要求；
成员类型信息，用于指示参与联邦学习的候选成员的类型要求；
数量信息，用于指示所述候选成员的数量要求；
兴趣区域AOI，用于指示所述候选成员所处的区域；
联邦学习的类型信息，用于指示所述联邦学习的属于纵向联邦或横向联邦。
第二方面，提供了一种候选成员的确定方法，包括：
第二网元向第一网元发送请求消息，所述请求消息中包括筛选信息，所述请求消息用于指示所述第一网元根据所述筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员；
所述第二网元接收所述第一网元发送的响应消息，所述响应消息中包括所述候选成员的标识；
其中，所述筛选信息包括如下至少一个：
时间段，用于指示进行所述联邦学习的可选时间段；
算法类型，用于指示进行所述联邦学习需要支持的模型训练的算法类型；
精度阈值，用于指示进行所述联邦学习需满足的模型训练的精度要求；
无线接入制式，用于指示进行所述联邦学习需选择的无线接入制式；
信号质量要求，用于指示进行所述联邦学习时无线信号质量要求；
流量范围，用于指示所述候选成员的流量使用范围要求；
成员类型信息，用于指示参与联邦学习的候选成员的类型要求；
数量信息，用于指示所述候选成员的数量要求；
兴趣区域AOI，用于指示所述候选成员所处的区域；
联邦学习的类型信息，用于指示所述联邦学习的属于纵向联邦或横向联邦。
第三方面，提供了一种候选成员的确定装置，包括：
接收模块，用于接收第二网元发送的请求消息，所述请求消息中包括筛选信息；
处理模块，用于根据所述筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员；
发送模块，用于向所述第二网元发送响应消息，所述响应消息中包括所述候选成员的标识；
其中，所述筛选信息包括如下至少一个：
时间段，用于指示进行所述联邦学习的可选时间段；
算法类型，用于指示进行所述联邦学习需要支持的模型训练的算法类型；
精度阈值，用于指示进行所述联邦学习需满足的模型训练的精度要求；
无线接入制式，用于指示进行所述联邦学习需选择的无线接入制式；
信号质量要求，用于指示进行所述联邦学习时无线信号质量要求；
流量范围，用于指示所述候选成员的流量使用范围要求；
成员类型信息，用于指示参与联邦学习的候选成员的类型要求；
数量信息，用于指示所述候选成员的数量要求；
兴趣区域AOI，用于指示所述候选成员所处的区域；
联邦学习的类型信息，用于指示所述联邦学习的属于纵向联邦或横向联 邦。
第四方面，提供了一种候选成员的确定装置，包括：
发送模块，用于向第一网元发送请求消息，所述请求消息中包括筛选信息，所述请求消息用于指示所述第一网元根据所述筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员；
接收模块，用于接收所述第一网元发送的响应消息，所述响应消息中包括所述候选成员的标识；
其中，所述筛选信息包括如下至少一个：
时间段，用于指示进行所述联邦学习的可选时间段；
算法类型，用于指示进行所述联邦学习需要支持的模型训练的算法类型；
精度阈值，用于指示进行所述联邦学习需满足的模型训练的精度要求；
无线接入制式，用于指示进行所述联邦学习需选择的无线接入制式；
信号质量要求，用于指示进行所述联邦学习时无线信号质量要求；
流量范围，用于指示所述候选成员的流量使用范围要求；
成员类型信息，用于指示参与联邦学习的候选成员的类型要求；
数量信息，用于指示所述候选成员的数量要求；
兴趣区域AOI，用于指示所述候选成员所处的区域；
联邦学习的类型信息，用于指示所述联邦学习的属于纵向联邦或横向联邦。
第五方面，提供了一种第一网元，该第一网元包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如第一方面所述的方法的步骤。
第六方面，提供了一种第一网元，包括处理器及通信接口，其中，所述处理器用于根据筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员，所述通信接口用于接收第二网元发送的请求消息，向第二网元发送响应消息，所述响应消息中包括所述候选成员的标识，其中，所述请求消息中包括筛选信息，所述筛选信息包括如下至少一个：
时间段，用于指示进行所述联邦学习的可选时间段；
算法类型，用于指示进行所述联邦学习需要支持的模型训练的算法类型；
精度阈值，用于指示进行所述联邦学习需满足的模型训练的精度要求；
无线接入制式，用于指示进行所述联邦学习需选择的无线接入制式；
信号质量要求，用于指示进行所述联邦学习时无线信号质量要求；
流量范围，用于指示所述候选成员的流量使用范围要求；
成员类型信息，用于指示参与联邦学习的候选成员的类型要求；
数量信息，用于指示所述候选成员的数量要求；
兴趣区域AOI，用于指示所述候选成员所处的区域；
联邦学习的类型信息，用于指示所述联邦学习的属于纵向联邦或横向联邦。
第七方面，提供了一种第二网元，该网元包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如第二方面所述的方法的步骤。
第八方面，提供了一种第二网元，包括处理器及通信接口，其中，所述通信接口用于向第一网元发送请求消息，所述请求消息中包括筛选信息，所述请求消息用于指示所述第一网元根据所述筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员；接收所述第一网元发送的响应消息，所述响应消息中包括所述候选成员的标识；
其中，所述筛选信息包括如下至少一个：
时间段，用于指示进行所述联邦学习的可选时间段；
算法类型，用于指示进行所述联邦学习需要支持的模型训练的算法类型；
精度阈值，用于指示进行所述联邦学习需满足的模型训练的精度要求；
无线接入制式，用于指示进行所述联邦学习需选择的无线接入制式；
信号质量要求，用于指示进行所述联邦学习时无线信号质量要求；
流量范围，用于指示所述候选成员的流量使用范围要求；
成员类型信息，用于指示参与联邦学习的候选成员的类型要求；
数量信息，用于指示所述候选成员的数量要求；
兴趣区域AOI，用于指示所述候选成员所处的区域；
联邦学习的类型信息，用于指示所述联邦学习的属于纵向联邦或横向联邦。
第九方面，提供了一种候选成员的确定系统，包括：第一网元及第二网元，所述第一网元可用于执行如第一方面所述的候选成员的确定方法的步骤，所述第二网元可用于执行如第二方面所述的候选成员的确定方法的步骤。
第十方面，提供了一种可读存储介质，所述可读存储介质上存储程序或指令，所述程序或指令被处理器执行时实现如第一方面所述的方法的步骤，或者实现如第二方面所述的方法的步骤。
第十一方面，提供了一种芯片，所述芯片包括处理器和通信接口，所述通信接口和所述处理器耦合，所述处理器用于运行程序或指令，实现如第一方面所述的方法，或实现如第二方面所述的方法。
第十二方面，提供了一种计算机程序/程序产品，所述计算机程序/程序产品被存储在存储介质中，所述计算机程序/程序产品被至少一个处理器执行以实现如第一方面、第二方面所述的候选成员的确定方法的步骤。
在本申请实施例中，第一网元通过接收第二网元发送的请求消息，并根据该请求消息中包括的筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员，并将候选成员的标识信息发送给第二网元，其中，筛选信息包括如下至少一个：时间段、算法类型、精度阈值、无线接入制式、信号质量要求、流量范围、成员类型信息、数量信息、兴趣区域AOI和联邦学习的类型。第一网元通过筛选信息，可以筛选出适合进行联邦学习的参与成员，从而提高联邦学习的训练效率。
附图说明
图1示出本申请实施例可应用的一种无线通信系统的框图；
图2为本申请提供的一种候选成员的确定方法的流程示意图；
图3为本申请提供的另一种候选成员的确定方法的流程示意图；
图4为本申请提供的一种候选成员的确定方法的信令图；
图5是本申请提供的候选成员的确定装置的结构示意图之一；
图6是本申请提供的候选成员的确定装置的结构示意图之二；
图7是本申请实施例提供的通信设备的结构示意图；
图8是本申请实施例提供的第一网元的硬件结构示意图；
图9是本申请实施例提供的第二网元的硬件结构示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员所获得的所有其他实施例，都属于本申请保护的范围。
本申请的说明书和权利要求书中的术语“第一”、“第二”等是用于区别类似的对象，而不用于描述特定的顺序或先后次序。应该理解这样使用的术语在适当情况下可以互换，以便本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施，且“第一”、“第二”所区别的对象通常为一类，并不限定对象的个数，例如第一对象可以是一个，也可以是多个。此外，说明书以及权利要求中“和/或”表示所连接对象的至少其中之一，字符“/”一般表示前后关联对象是一种“或”的关系。
值得指出的是，本申请实施例所描述的技术不限于长期演进型(Long Term Evolution，LTE)/LTE的演进(LTE-Advanced，LTE-A)系统，还可用于其他无线通信系统，诸如码分多址(Code Division Multiple Access，CDMA)、时分多址(Time Division Multiple Access，TDMA)、频分多址(Frequency Division Multiple Access，FDMA)、正交频分多址(Orthogonal Frequency Division Multiple Access，OFDMA)、单载波频分多址(Single-carrier Frequency Division Multiple Access，SC-FDMA)和其他系统。本申请实施例中的术语“系 统”和“网络”常被可互换地使用，所描述的技术既可用于以上提及的系统和无线电技术，也可用于其他系统和无线电技术。以下描述出于示例目的描述了新空口(New Radio，NR)系统，并且在以下大部分描述中使用NR术语，但是这些技术也可应用于NR系统应用以外的应用，如第6代(6th Generation，6G)通信系统。
图1示出本申请实施例可应用的一种无线通信系统的框图。无线通信系统包括终端11和网络侧设备12。本申请中终端11也可以叫用户设备(User Equipment，UE)，其中，终端11可以是手机、平板电脑(Tablet Personal Computer)、膝上型电脑(Laptop Computer)或称为笔记本电脑、个人数字助理(Personal Digital Assistant，PDA)、掌上电脑、上网本、超级移动个人计算机(ultra-mobile personal computer，UMPC)、移动上网装置(Mobile Internet Device，MID)、增强现实(augmented reality，AR)/虚拟现实(virtual reality，VR)设备、机器人、可穿戴式设备(Wearable Device)、车载设备(VUE)、行人终端(PUE)、智能家居(具有无线通信功能的家居设备，如冰箱、电视、洗衣机或者家具等)、游戏机、个人计算机(personal computer，PC)、柜员机或者自助机等终端侧设备，可穿戴式设备包括：智能手表、智能手环、智能耳机、智能眼镜、智能首饰(智能手镯、智能手链、智能戒指、智能项链、智能脚镯、智能脚链等)、智能腕带、智能服装等。需要说明的是，在本申请实施例并不限定终端11的具体类型。网络侧设备12可以包括接入网设备或核心网设备，其中，接入网设备12也可以称为无线接入网设备、无线接入网(Radio Access Network,RAN)、无线接入网功能或无线接入网单元。接入网设备12可以包括基站、WLAN接入点或WiFi节点等，基站可被称为节点B、演进节点B(eNB)、接入点、基收发机站(Base Transceiver Station，BTS)、无线电基站、无线电收发机、基本服务集(Basic Service Set，BSS)、扩展服务集(Extended Service Set，ESS)、家用B节点、家用演进型B节点、发送接收点(Transmitting Receiving Point，TRP)或所述领域中其他某个合适的术语，只要达到相同的技术效果，所述基站不限于特定技术词汇，需要说明的 是，在本申请实施例中仅以NR系统中的基站为例进行介绍，并不限定基站的具体类型。核心网设备可以包含但不限于如下至少一项：核心网节点、核心网功能、移动管理实体(Mobility Management Entity，MME)、接入移动管理功能(Access and Mobility Management Function，AMF)、会话管理功能(Session Management Function，SMF)、用户平面功能(User Plane Function，UPF)、策略控制功能(Policy Control Function，PCF)、策略与计费规则功能单元(Policy and Charging Rules Function，PCRF)、边缘应用服务发现功能(Edge Application Server Discovery Function，EASDF)、统一数据管理(Unified Data Management，UDM)，统一数据仓储(Unified Data Repository，UDR)、归属用户服务器(Home Subscriber Server，HSS)、集中式网络配置(Centralized network configuration，CNC)、网络存储功能(Network Repository Function，NRF)，网络开放功能(Network Exposure Function，NEF)、本地NEF(Local NEF，或L-NEF)、绑定支持功能(Binding Support Function，BSF)、应用功能(Application Function，AF)等。需要说明的是，在本申请实施例中仅以NR系统中的核心网设备为例进行介绍，并不限定核心网设备的具体类型。
下面结合附图，通过一些实施例及其应用场景对本申请实施例提供的候选成员的确定方法进行详细地说明。
联邦学习包括横向联邦学习和纵向联邦学习，其中，横向联邦学习的本质是样本的联合，其适用于参与者间业态相同但触达客户不同，即特征重叠多，样本重叠少时的场景，比如通信网络内核心网(core network，CN)域和无线接入网(Radio Access Network，RAN)域服务不同用户(如每一个终端，即样本不同)的同一服务(如移动性管理(mobility management，MM)、会话管理(session management，SM)业务，某一业务)。通过联合参与方的不同样本的相同数据特征，横向联邦使训练样本的数量增多，从而得到一个更好的模型。
纵向联邦学习的本质是特征的联合，适用于样本重叠多，特征重叠少的 场景，比如通信网络内CN域和RAN域服务相同用户(如终端，即样本相同)的不同服务(如MM、SM业务，即特征不同)。通过联合参与方的共同样本的不同数据特征，纵向联邦使训练样本的特征维度增多，并得到一个更好的模型。
在联邦学习系统中，包括管理者(服务器)和多个参与方，其中管理者用于向各参与方发送模型以及根据各参与方的反馈结果做模型更新并将更新后的模型再次发送给各参与方用于下一轮模型训练。其中，每个参与方有各自的数据，为了不将本地数据发送给其他人，每个参与方均使用管理者发送的模型在本地进行训练后返回给管理者，用于模型更新。
示例性的，模型训练的过程可以包括如下步骤：
步骤1：参与方各自从管理者(服务器)下载最新模型；
步骤2：每个参与方利用本地数据训练模型，加密梯度上传给管理者(服务器)，管理者(服务器)聚合各参与方的梯度更新模型参数；
步骤3：管理者(服务器)返回更新后的模型给各参与方；
步骤4：各参与方更新各自模型。
上述步骤2到步骤4进行多次迭代/模型更新，在一定条件下(如完成一定次数的迭代数量或模型的损失函数计算值低于预设值后，完成模型训练。
在5GS中引入了网络数据分析功能网元(network data analytic function，NWDAF)。NWDAF可以从核心网各个网元、网管系统等处收集数据进行大数据统计、分析或者智能化的数据分析，得出网络侧的分析或者预测数据，从而辅助各个网元根据数据分析结果对终端接入进行更有效的控制。
本申请提供的候选成员的确定方法中，可以通过NWDAF收集其他网元的数据，并对这些数据进行分析，从而根据筛选信息和数据分析结果，筛选出适合进行联邦学习的参与成员，从而提高联邦学习的效率。
需要说明的，本申请中的“候选成员的确定”可以理解为“可参与联邦学习的候选成员的确定”。
图2为本申请提供的一种候选成员的确定方法的流程示意图，下面结合 图2描述本申请实施例提供的候选成员的确定方法。如图2所示，该方法包括：
步骤201：第一网元接收第二网元发送的请求消息。
其中，本申请实施例提供的候选成员的确定方法的执行主体为第一网元，该第一网元可以以各种形式来实施。例如，本申请实施例中描述的第一网元可以包括NWDAF，当然，也可以为其他能够从核心网各个网元、网管系统等处收集数据进行大数据统计、分析或者智能化的数据分析的网元。第二网元可以包括任务消费者网元，例如可以为应用功能网元(application function，AF)、基站、终端等网元或者设备，其还可以为第三方服务器。
其中，请求消息中包括筛选信息，该筛选信息包括如下至少一个：时间段、算法类型、精度阈值、无线接入制式、信号质量要求、流量范围、成员类型信息、数量信息、区域信息和联邦学习的类型信息。基于不同维度的筛选信息，可以筛选出更为合适的参与成员，从而提高联邦学习的效率。
其中，时间段用于指示进行所述联邦学习的可选时间段，时间段可以为过去的某个时间段，也可以为未来的时间段，可以理解的是，若为未来的时间段，第一网元将根据获取的历史数据，预测未来时间段内各个成员的联网状态。联网状态包括已连接网络的状态和未连接网络的状态。应理解，通常会按照实际进行联邦学习的时间段，灵活设置时间段。
算法类型，用于指示进行所述联邦学习需要支持的模型训练的算法类型，包括各个成员支持的关于机器学习等人工智能(artificial intelligence，AI)数据分析任务相关的算法类型，如“深度学习算法”，“线性回归算法”等。应理解，通常会根据模型的功能，选择支持与该功能匹配的算法类型的成员作为联邦学习的参与成员。
精度阈值，用于指示进行所述联邦学习需满足的模型训练的精度要求，包括各个成员经过训练后产生的模型所能达到的精度值，如模型的准确率，也可以理解为模型预测或判断正确的百分比。应理解，为了保证联邦学习的准确性，通常会选择一些精度比较高的成员作为联邦学习的参与成员。
无线接入制式，用于指示进行联邦学习需选择的无线接入制式，包括各个成员连接或者接入通信网络的方式，如连接non-3GPP的WLAN(类似连接WiFi)，连接3GPP的5g、4g网络等。应理解，为了保证联邦学习的稳定性，以及减少流量消耗，通常会选择一些连接WLAN或者WiFi的成员作为联邦学习的参与成员。
信号质量要求，用于指示进行所述联邦学习时无线信号质量要求。信号质量要求可以包括各个成员连接通信网络时的网络信号强度阈值和/或稳定性的阈值等。该信号质量要求包括WLAN网络信号质量要求，也可以包括5G NR网络信号质量要求，还可以包括4G长期演进(Long Term Evolution，LTE)网络信号质量要求。该信号质量要求也可以理解为保持信号强度在某个数值以上的时间比例最低要求，例如，有至少90％的时间中信号强度可以达到阈值要求。应理解，为了保证联邦学习的稳定性，通常会选择一些信号质量较强(如，网络信号强度较强并且稳定性较好)的成员作为联邦学习的参与成员。
其中，在连接WLAN的情况下，可以用接收的信号强度指示(Received Signal Strength Indication，RSSI)的平均值、方差等或信号的往返时延(round-trip time，RTT)表示信号质量。
流量范围，用于指示所述候选成员的流量使用范围要求，为各个成员在预设时间段内所消耗的流量值要求，如上传和下载的流量数量等信息，例如可以是UE的accumulated usage。应理解，为了减轻各个参与成员在进行联邦学习时的压力，通常会选择一些网络状态稳定，且流量消耗较少的成员作为联邦学习的参与方。
成员类型信息用于指示参与联邦学习的候选成员的类型要求，成员类型信息可以理解为此次参与联邦学习的候选成员的类型，例如可以为终端、核心网网元(如NWDAF)或基站等。
兴趣区域(area of interest，AOI)，用于指示所述候选成员所处的区域，AOI可为参与联邦学习的候选成员所处的区域，或者所关注的、或者所感兴 趣的区域范围，其可以是经纬度，或者是一个或多个小区/跟踪区(tracking area，TA)等。
数量信息，用于指示所述候选成员的数量要求，该数量信息可以理解为参与联邦学习的成员数量，也即第一网元所需要确定的候选成员的数量。第一网元可以确定该数量个候选网元。该数量信息可以是要求的候选成员的最小数量，也就是确定的候选成员的数量不能小于要求的候选成员的最小数量。该数量信息也可以是要求的候选成员的最大数量，也就是确定的候选成员的数量不能大于要求的候选成员的最大数量。
联邦学习的类型信息，用于指示所述联邦学习的属于纵向联邦或横向联邦。可选地，上述请求消息中还可以包括指示信息，该指示信息用于指示执行参与联邦学习的候选成员的确定任务，也即表示此次任务是用于联邦学习或者联邦学习的成员选择的，第一网元在接收到请求消息后，基于该请求消息或者基于该请求消息中的指示信息，即可获知需要执行参与联邦学习的候选成员的确定任务。
步骤202：第一网元根据筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员。
在本步骤中，第一网元在接收到请求消息后，将根据该请求消息中的筛选信息从多个第二设备中进行筛选，从而确定出满足筛选信息的一个或多个第一设备，并将这些第一设备确定为可参与联邦学习的候选成员。
可选地，上述第一设备包括以下至少一项：在时间段内处于联网状态的第一设备；支持在所述时间段进行联邦学习的第一设备，支持算法类型的第一设备；支持利用所述算法类型进行联邦学习的第一设备，模型训练精度信息大于精度阈值的第一设备；支持无线接入制式的第一设备；处于所述无线接入制式下的第一设备；支持在所述无线接入制式下进行联邦学习的第一设备；信号质量大于信号质量阈值的第一设备；信号质量大于信号质量阈值要求的第一设备；所消耗流量处于流量范围的第一设备；满足所述候选成员的类型要求的第一设备；位于所述AOI内的第一设备；联邦学习的类型与所述 筛选信息中包括的联邦学习的类型相同的第一设备。
其中，在筛选信息中包括时间段的情况下，第一设备为在时间段内处于联网状态的第一设备，或者，第一设备为支持在所述时间段进行联邦学习的第一设备；在筛选信息中包括算法类型的情况下，第一设备为支持算法类型的第一设备，或者所述第一设备为支持利用所述算法类型进行联邦学习的第一设备；在筛选信息中包括精度阈值的情况下，第一设备为模型训练精度信息大于精度阈值的第一设备；在筛选信息中包括无线接入制式的情况下，第一设备为支持无线接入制式的第一设备，或者，所述第一设备为处于所述无线接入制式下的第一设备，或者所述第一设备为支持在所述无线接入制式下进行联邦学习的第一设备；在筛选信息中包括信号质量阈值的情况下，第一设备为信号质量大于信号质量阈值的第一设备；在筛选信息包括信号质量要求的情况下，所述第一设备为信号质量大于信号质量要求的第一设备；在筛选信息中包括流量范围的情况下，第一设备为所消耗流量处于流量范围的第一设备；在筛选信息中包括成员类型信息的情况下，第一设备为满足所述候选成员的类型要求的第一设备，可以理解，第一设备的类型与所述筛选信息中指示的成员类型信息相同；在所述筛选信息中包括AOI的情况下，第一设备为位于所述AOI内的第一设备；在所述筛选信息包括联邦学习的类型信息的情况下，第一设备的联邦学习的类型与所述筛选信息中包括的联邦学习的类型信息相同。
另外，在筛选信息中包括设备类型的情况下，第一设备为设备类型的第一设备，在筛选信息中包括区域信息的情况下，第一设备为处于该区域信息的第一设备。
应理解，在筛选信息中包括至少两项的情况下，第一设备为满足相应的至少两项筛选信息的设备。例如，在筛选信息中包括时间段以及支持的算法类型的情况下，第一设备为在时间段内处于联网状态，且支持算法类型的第一设备。对于筛选信息中包括其他至少两项信息的情况，与包括时间段以及支持的算法类型的情况类似，此处不再赘述。
在上述实施例中，通过筛选信息可以从至少一个第二设备中筛选满足筛选信息的第一设备作为候选网元，从而可以选择出适合可参与联邦学习的成员，有助于提高联邦学习训练的效率。
可选地，第一网元在确定候选成员时，可以确定筛选信息所对应的数据类型，基于数据类型从至少一个第三网元中获取至少一个设备各自对应的属性信息，并根据属性信息和筛选信息，确定候选成员。
可选地，上述数据类型包括如下至少一项联网信息以及对应的时间段、算法类型、无线接入制式、信号质量和流量。
具体的，在筛选信息包括处于时间段的情况下，该筛选信息所对应的数据类型为联网信息以及处于联网状态对应的时间段，在筛选信息包括支持的算法类型的情况下，该筛选信息所对应的数据类型为算法类型，在筛选信息包括精度阈值的情况下，该筛选信息所对应的数据类型为精度信息，在筛选信息包括无线接入制式的情况下，该筛选信息所对应的数据类型为无线接入制式，在筛选信息包括信号质量阈值的情况下，该筛选信息所对应的数据类型为信号质量，在筛选信息包括流量范围的情况下，该筛选信息所对应的数据类型为流量，在筛选信息包括成员类型的情况下，该筛选信息所对应的数据类型为设备类型。在筛选信息包括区域信息的情况下，该筛选信息所对应的数据类型为位置信息。
可以理解的是，在筛选信息中包括成员类型的情况下，成员类型不同，请求消息中包括的筛选信息不同，也就是说，筛选信息和成员类型相关。例如，若成员类型包括终端，则筛选信息可以包括时间段、支持的算法类型、精度阈值、无线接入制式和信号质量阈值、流量范围、以及区域信息中的至少一个。若成员类型包括网络侧设备，如基站或核心网设备，则筛选信息可以包括支持的算法类型、精度阈值中的至少一个。
第一网元在确定出数据类型后，将基于该数据类型从至少一个第三网元中获取至少一个第二设备各自对应的属性信息。其中，第三网元包括不同的多个网元，应理解，需要获取的属性信息不同，第一网元会从不同的第三网 元中获取属性信息。例如：第一网元从第三网元会话管理功能(session management function，SMF)中获取终端连接WLAN的时间、是否连接WLAN等信息。第一网元从第三网元中获取连接WLAN的信号质量等信息，其中，第三网元可以是网络管理设备，例如可以是运行管理和维护(operation administration and maintenance，OAM)，从第三网元统一数据管理功能(unified data management，UDM)或者从第三网元数据收集应用功能(data collection-application function，DC-AF)获取各个候选网元所支持的算法和精度，从第三网元用户面功能(the user plane function，UPF)获取终端的流量信息等，或者，第一网元从SMF或计费功能(Charging Function，CHF)获取UE accumulated usage信息。
可以理解的是，在数据类型包括联网信息以及对应的时间段的情况下，属性信息为至少一个设备各自对应的当前的联网状态，以及处于联网状态的时间，如是否连接网络以及在哪一个时间段内处于联网状态等。在数据类型包括算法类型的情况下，属性信息为至少一个设备各自对应的当前所能支持的算法，如是否支持深度学习算法，是否支持线性回归算法等。在数据类型包括精度信息的情况下，属性信息为至少一个设备各自对应的当前所能支持的精度，如各个设备对模型进行训练后，模型所能达到的精度等。在数据类型包括无线接入制式的情况下，属性信息为至少一个设备当前所接入的网络制式类型，如当前所接入的是WLAN，还是5G网络等。在数据类型包括信号质量的情况下，属性信息为至少一个第二设备当前所接入网络的信号质量，或者在预设时间段内所接入网络的信号质量。在数据类型包括流量信息的情况下，属性信息为至少一个第二设备各自对应的在预设时间段内所消耗的流量。在数据类型包括设备类型的情况下，属性信息为至少一个第二设备各自对应的类型，如为终端、基站还是核心网设备。在数据类型包括位置信息的情况下，属性信息为至少一个第二设备各自所处的位置，如当前所在的地理位置，或者所处的小区或TA等。
在获取到该属性信息后，将至少一个第二设备各自对应的属性信息和筛 选信息进行匹配，以根据该筛选信息，确定候选成员，例如可以将满足筛选信息的第一设备确定为候选成员。
举例来说，假设筛选信息中包括设备类型为终端、区域为A小区、A时间段、无线接入制式为连接non-3GPP的WiFi。则第一网元将根据这些筛选信息获取对应的属性信息，并根据获取的属性信息，选择处于A小区、在A时间段内处于联网状态、且连接non-3GPP的WiFi的终端作为候选成员。
在本实施例中，第一网元通过获取筛选信息所对应的数据类型，并从至少一个第三网元中获取各个第二设备的属性信息，从而基于该属性信息和筛选信息，确定候选成员，由于通过预先设置的筛选信息可以从至少一个第二设备中确定适合参与联邦学习的候选成员，从而可以提高联邦学习的效率。
在一种可能的实现方式中，第一网元可以直接将满足筛选信息的设备确定为候选成员，并且默认这些满足筛选信息的候选网元均可参与联邦学习，第一网元可以将这些满足筛选信息的候选成员的标识均发送给第二网元。
在另一种可能的实现方式中，为了筛选出更合适的参与者，第一网元在根据属性信息和筛选信息确定候选成员时，还可以基于各个第二设备的意愿信息进行筛选。示例性的，第一网元可以从第三网元中获取第一指示信息，并根据筛选信息和第一指示信息，从第二设备中，将一个或多个第一设备确定为候选成员，其中，上述第一指示信息用于表示各个设备参与联邦学习的意愿信息，其中，意愿信息表示各个设备是否愿意参与联邦学习。
其中，上述第一设备为满足筛选信息，且愿意参与联邦学习的设备。
示例性的，每个设备对应自己的第一指示信息，对于某个设备来说，其对应的第一指示信息为0的情况下，表示该设备愿意参与联邦学习，在第一指示信息为1的情况下，表示该设备不愿意参与联邦学习，或者，也可以是在第一指示信息为1的情况下，表示该设备愿意参与联邦学习，在第一指示信息为0的情况下，表示该设备不愿意参与联邦学习。当然，第一指示信息也可以采用其他数值表示各设备参与联邦学习的意愿信息。
第一网元在获取到第一指示信息后，将各个设备的属性信息和筛选信息 进行匹配，将满足筛选信息、且第一指示信息指示愿意参与联邦学习的设备确定为候选成员。
在本实施例中，第一网元根据筛选信息和第一指示信息筛选满足筛选信息，且愿意参与联邦学习的设备作为候选成员，从而有助于提高后续进行联邦学习的效率。
可选地，第一网元在根据筛选信息和第一指示信息，从第二设备中，将一个或多个第一设备确定候选成员时，还可以从第三网元中获取各第二设备参与联邦学习的能力信息，从而根据筛选信息、第一指示信息和能力信息，确定候选成员。
其中，能力信息包括如下信息中的至少一个：参与联邦学习的无线接入制式、参与联邦学习的区域、参与联邦学习的时间、参与联邦学习所能支持的算法信息、参与联邦学习所能达到的精度信息和参与联邦学习的类型。
其中，参与联邦学习的类型，可以包括愿意参与横向联邦学习，或者愿意参与纵向联邦学习。
一种实施方式中，所述第一网元根据所述筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员，包括：
所述第一网元从第三网元中获取所述第一设备参与所述联邦学习的意愿信息；
所述第一网元根据所述意愿信息和所述筛选信息，将愿意参与所述联邦学习且与所述筛选信息匹配的第一设备确定为可参与联邦学习的候选成员。
一种实施方式中，所述第一网元根据所述筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员，包括：
所述第一网元获取所述第一设备的联邦学习的能力信息，所述能力信息包括如下信息中的至少一个：参与联邦学习的无线接入制式、参与联邦学习的区域、参与联邦学习的时间、参与联邦学习所能支持的算法信息、参与联邦学习所能达到的精度信息和参与联邦学习的类型；
所述第一网元根据所述能力信息和所述筛选信息，从所述第一设备中确 定可参与联邦学习的候选成员，所述候选成员的能力信息与所述筛选信息匹配。
需要说明的是，第一网元可以从第四网元获取所述第一设备的联邦学习的能力信息，其中，第四网元可以为以下至少一项：NRF、UDM、数据采集协调功能(Data Collection Coordination Function，DCCF)、AMF。
所述候选成员的能力信息与所述筛选信息匹配，包括以下至少一项：
所述候选成员参与联邦学习的无线接入制式与所述筛选信息中包括的无线接入制式相同；
所述候选成员参与联邦学习的区域位于所述筛选信息中包括的AOI内；
所述候选成员参与联邦学习的时间在所述筛选信息中包括的时间段内；
所述候选成员参与联邦学习所能支持的算法类型包含于所述筛选信息中包括的算法类型；
所述候选成员参与联邦学习所能达到的精度信息高于所述筛选信息中包括的精度阈值；
所述候选成员参与联邦学习的类型与所述筛选信息中包括的联邦学习的类型相同。
可选的，第一网元还可以从第三网元中获取各设备参与联邦学习的能力信息，从而将各个设备的属性信息和筛选信息进行匹配，将第一指示信息指示愿意参与联邦学习的设备且参与联邦学习的能力信息满足所述筛选信息的设备，确定为候选成员，从而提高了后续进行联邦学习的效率。
其中，参与联邦学习的能力信息满足所述筛选信息包括：
候选成员参与联邦学习的无线制式与所述筛选信息中包括的无线接入制式相同；
所述候选成员参与联邦学习的时间在所述筛选信息中包括的时间段内；
所述候选成员参与联邦学习所能支持的算法类型包含于所述筛选信息中包括的算法类型；
所述候选成员参与联邦学习所能达到的精度信息高于所述筛选信息中包 括的精度阈值；
所述候选成员参与联邦学习的类型与所述筛选信息中包括的联邦学习的类型相同。
举例来说，若筛选信息中包括有支持的算法类型，且该支持的算法类型为深度学习算法，第一网元基于第一指示信息筛选出有80个原意参与联邦学习的设备，第一网元基于从第三网元中获取的这80个设备参与联邦学习所能支持的算法信息可以进一步进行筛选，如获取到其中50个设备在参与联邦学习时既能支持深度学习算法，又能支持线性回归算法，有30个设备在参与联邦学习时仅支持深度学习算法，则第一网元可以基于参与联邦学习时能够支持的算法类型，进一步筛选出既能支持深度学习算法，又能支持线性回归算法的第一设备作为候选网元。
应理解，对于能力信息包括其他信息时，实现方式与能力信息包括参与联邦学习所能支持的算法类型的实现方式类似，此处不再赘述。
其中，第一网元可以从UDM/DCAF/NRF/这些第三网元中获取各设备参与联邦学习的意愿信息和能力信息，其中，UDM/DCAF/NRF/为具有存储能力的网元。
可选地，第二网元发送的请求消息中还包括第二指示信息，该第二指示信息用于指示联邦学习对应的业务类型。第一网元在确定候选成员时，可以是根据筛选信息和第二指示信息，从多个第二设备中，将一个或多个第一设备确定为候选成员，其中，第一设备为可支持业务类型对应业务的设备。
具体的，第二指示信息可以为analytic ID，其用于指示联邦学习对应的业务类型。在数据分析任务(即NWDAF所负责的其他任务)中，该analytic ID可为：“UE mobility”其用于指示此次任务为和UE移动性相关的分析任务，“NF load”，用于指示此次任务为和网元负载相关的分析任务。
通常，联邦学习对应的业务类型不同的情况下，所选择的候选成员可能不同，因此，第一网元在选择候选成员时，还可以依据筛选信息以及analytic ID所指示的业务类型，从多个第二设备中筛选可支持业务类型对应的业务的 第一设备作为候选成员。如：第二指示信息用于指示联邦学习对应的业务类型为和终端移动性相关的分析任务，则选择出的第一设备为能够支持终端移动性相关的分析任务的设备。
在本实施例中，进一步根据联邦学习对应的业务类型筛选候选成员，从而可以提高选择出的候选成员的准确度，提高了联邦学习的训练效率以及训练后的模型的精度。
一种实施方式中，所述第一网元根据所述筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员，包括：
所述第一网元获取所述第一设备对应的网络状态信息，所述网络状态信息包括以下至少一项：
所述第一设备的类型信息；
所述第一设备的位置信息；
所述第一设备所处的无线接入制式信息；
所述第一设备的无线信号质量信息；
所述第一网元根据所述第一设备对应的网络状态信息和所述筛选信息，从所述第一设备中确定可参与联邦学习的候选成员，所述候选成员对应的网络状态信息与所述筛选信息匹配。
需要说明的是，第一网元可以从第三设备获取第一设备对应的网络状态信息，所述第三设备可以为AMF、UDM、NRF、PCF、RAN或操作管理和维护设备(OAM Operation Administration and Maintenance，OAM)。例如，第一网元可以从UDM或NRF中获取所述第一设备的类型信息，或者，第一网元可以从AMF或PCF获取第一设备所处的无线接入制式信息，或者，第一网元可以从RAN或OAM获取所述第一设备的无线信号质量信息，或者，第一网络可以从SMF或CHF获取UE accumulated usage信息等。
其中，所述候选成员对应的网络状态信息与所述筛选信息匹配，包括以下至少一项：
所述候选成员所处的无线接入制式与所述筛选信息中包括的无线接入制 式相同；
所述候选成员的无线信号质量信息大于所述筛选信息中包括的信号质量要求；
所述候选成员的位置位于所述筛选信息中包括的AOI内；
所述候选成员的类型包含于所述筛选信息中包括的成员类型信息。
可选地，上述请求消息中还包括排序指示信息，该排序指示信息用于指示对候选成员按照第二信息进行排序，其中，第二信息包括如下至少一项：候选成员的信号质量、候选成员的精度信息和候选成员的流量信息；在请求消息中包括排序指示信息的情况下，响应消息中的候选成员的标识顺序为按照第二信息进行排序后得到的顺序。
具体的，在该请求消息中包括有排序指示信息的情况下，第一网元在确定出候选成员之后，将按照排序指示信息中的第二信息，对确定出的候选成员进行升序或者降序排序。例如，可以按照候选成员的信号质量进行升序或者降序排序，或者可以按照候选成员的精度信息进行升序或者降序排序，或者也可以按照候选成员的流量信息进行升序或者降序排序。
第一网元在向第二网元发送响应消息时，响应消息中的候选成员的标识顺序为按照第二信息进行排序后得到的顺序。
在该实施例中，第一网元按照第二信息，对确定出的候选成员进行排序，这样，第二网元在接收到第一网元发送的候选成员的标识信息后，将根据该排序信息进一步筛选出更为符合自身要求的候选成员，来参与后续的联邦学习，从而可以提高联邦学习的效率。
可选地，上述请求消息中还包括分组指示信息，该分组指示信息用于指示对候选成员按照第三信息进行分组，其中，第三信息包括如下至少一项：候选成员所处的区域、候选成员处于联网状态的时间段、候选成员所支持的算法类型、候选成员的精度信息、候选成员的无线接入制式和候选成员的信号质量；在请求消息中包括分组指示信息的情况下，响应消息中的候选成员的标识为按照第三信息进行排序后得到的标识。
具体的，在该请求消息中包括有分组指示信息的情况下，第一网元在确定出候选成员之后，将按照分组指示信息中的第三信息，对确定出的候选成员进行分组。例如，可以按照候选成员所处的区域进行分组，如将处于相同区域的候选成员分为一组等。或者可以按照候选成员处于联网状态的时间段进行分组，如在相同时间段处于联网状态的候选成员分为一组，或者也可以按照候选成员所支持的算法类型进行分组，如将支持相同算法类型的网元分为一组，或者也可以按照候选成员的精度信息进行分组，如将精度信息相同的候选成员分为一组，或者也可以按照候选成员的无线接入制式进行分组，如将无线接入制式相同的候选成员分为一组，或者也可以按照候选成员的信号质量进行分组，或者也可以按照候选成员的流量信息进行分组，或者也可以按照候选成员可进行联邦学习的时间段，如将可进行联邦学习的时间段相同的候选成员分为一组，例如均为白天10点到12点等等。
第一网元在向第二网元发送响应消息时，响应消息中的候选成员的标识为按照第三信息进行分组后得到的标识。
在该实施例中，第一网元按照第三信息，对确定出的候选成员进行分组，这样，第二网元在接收到第一网元发送的候选成员的标识信息后，将根据该分组信息进一步筛选出更为符合自身要求的候选成员，来参与后续的联邦学习，从而可以提高联邦学习的效率。
步骤203：第一网元向第二网元发送响应消息，该响应消息中包括候选成员的标识。
在本步骤中，第一网元在确定出可参与联邦学习的候选成员之后，可以将这些候选成员的标识携带在响应消息中发送给第二网元。其中，候选成员的标识可以包括候选成员的用户永久标识符(Subscriber Permanent Identifier，SUPI)或者网络协议(Internet Protocol，IP)，所述候选成员的标识还可以为通用公共签约标识(Generic Public Subscription Identifier，GPSI)、国际移动用户识别码(International Mobile Subscriber Identity，IMSI)、AF特定UE标识(AF specific UE ID)或者为UE IP地址。
可选地，上述响应消息中还包括如下信息中的至少一种：候选成员所处的区域、候选成员处于联网状态的时间段、候选成员所支持的算法类型、候选成员的精度信息、候选成员的无线接入制式、候选成员的信号质量、候选成员的流量信息和候选成员可进行联邦学习的时间段。
其中，候选成员所处的区域可以包括候选成员所处的位置，如经纬度或小区等。
候选成员可进行联邦学习的时间段为第一网元根据各个设备的意愿信息以及能力信息分析出的预测结果。
候选成员的信号质量，在连接WLAN的情况下，可以用RSSI的平均值、方差或RTT表示。
候选成员的流量信息，可以包括在预设时间段内所消耗的流量值，如上传和下载的流量数量等信息。
可选地，上述响应消息中还可以包括第二指示信息，该第二指示信息可以为analytic ID，其用于指示联邦学习对应的业务类型。
可选地，上述响应消息中还可以包括处于联网状态的覆盖时间比例，例如一天有多长时间处于wifi连接状态下。
可选地，上述响应消息中还可以包括候选成员的地址信息，第二网元可以根据该地址信息找到该候选成员，以进行连接和联邦学习。
可选地，上述响应消息中还可以包括数量信息，该数量信息可以理解为参与联邦学习的成员数量，也即第一网元所需要确定的候选成员的数量。
进一步地，第二网元在接收到第一网元发送的响应消息后，将根据实际进行联邦学习训练的情况，选择真正参与联邦学习的目标成员。
例如，第二网元可以根据响应消息中候选成员的信号质量确定要进行联邦学习的成员，可以选择信号质量最好的前100名成员进行联邦学习。当然，根据响应消息中的其他信息确定目标成员的方式，与根据信号质量确定目标成员的方式类似，此处不再赘述。
可选地，在第二网元向第一网元发送的请求消息中包括数量信息的情况 下，若第一网元返回的响应消息中包括目标数量个候选成员，则第二网元可以直接将这目标数量个候选成员作为目标成员。例如，若请求消息中包括要求50名联邦学习的成员，第一网元返回的响应消息中带有50名的候选成员，则第二网元可以直接选择这50名成员作为目标成员。
进一步地，第二网元在确定出目标成员之后，将根据目标成员的标识信息与这些目标成员进行连接，以进行联邦学习。
在本实施例中，第二网元可以根据第一网元返回的候选成员，进一步筛选真正参与联邦学习的目标成员，由此可以使得筛选出的目标成员更适合参与联邦学习，提高了联邦学习的效率。
本申请实施例提供的候选成员的确定方法，通过接收第二网元发送的请求消息，并根据该请求消息中包括的筛选信息，将一个或多个第一设备确定为可参与联邦学习的候选成员，并将候选成员的标识信息发送给第二网元，其中，筛选信息包括如下至少一个：时间段、算法类型、精度阈值、无线接入制式、信号质量要求、流量范围、成员类型信息、数量信息、区域信息和联邦学习的类型信息。第一网元通过筛选信息，可以筛选出适合进行联邦学习的参与成员，从而提高联邦学习的效率。
图3为本申请提供的另一种候选成员的确定方法的流程示意图。该实施例中执行主体为第二网元。如图3所示，该方法包括：
步骤301：第二网元向第一网元发送请求消息。
其中，请求消息中包括筛选信息，请求消息用于指示第一网元根据筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员。
其中，筛选信息包括如下至少一个：
时间段，用于指示进行所述联邦学习的可选时间段；
算法类型，用于指示进行所述联邦学习需要支持的模型训练的算法类型；
精度阈值，用于指示进行所述联邦学习需满足的模型训练的精度要求；
无线接入制式，用于指示进行所述联邦学习需选择的无线接入制式；
信号质量要求，用于指示进行所述联邦学习时无线信号质量要求；
流量范围，用于指示所述候选成员的流量使用范围要求；
成员类型信息，用于指示参与联邦学习的候选成员的类型要求；
数量信息，用于指示所述候选成员的数量要求；
兴趣区域AOI，用于指示所述候选成员所处的区域；
联邦学习的类型信息，用于指示所述联邦学习的属于纵向联邦或横向联邦。
步骤302：第二网元接收第一网元发送的响应消息，该响应消息中包括候选成员的标识。
本申请实施例中的具体实现过程以及有益效果可以参照图2所示实施例的内容，此处不再赘述。
可选地，所述第一设备包括以下至少一项：
在所述时间段内处于联网状态的第一设备；
支持在所述时间段进行联邦学习的第一设备；
支持所述算法类型的第一设备；
支持利用所述算法类型进行联邦学习的第一设备；
模型训练精度信息大于所述精度阈值的第一设备；
处于所述无线接入制式下的第一设备；
支持在所述无线接入制式下进行联邦学习的第一设备；
信号质量大于信号质量要求的第一设备；
所消耗流量处于所述流量范围的第一设备；
满足所述候选成员的类型要求的第一设备；
位于所述AOI内的第一设备；
联邦学习的类型与所述筛选信息中包括的联邦学习的类型信息相同的第一设备。
可选地，所述请求消息中还包括：指示信息，所述指示信息用于指示执行参与联邦学习的候选成员的确定任务，或用于指示此次任务是用于联邦学习或者联邦学习的成员选择的。
可选地，请求消息中还包括排序指示信息，排序指示信息用于指示对候选成员按照第二信息进行排序，第二信息包括如下至少一项：候选成员的信号质量、候选成员的精度信息和候选成员的流量信息；
响应消息中的候选成员的标识为按照第二信息进行排序后得到的标识。
可选地，请求消息中还包括分组指示信息，分组指示信息用于指示对候选成员按照第三信息进行分组，第三信息包括如下信息中的至少一种：候选成员所处的区域、候选成员处于联网状态的时间段、候选成员所支持的算法类型、候选成员的精度信息、候选成员的无线接入制式和候选成员的信号质量；
响应消息中的候选成员的标识为按照第三信息进行分组后得到的标识。
可选地，响应消息中还包括候选成员所处的区域、候选成员处于联网状态的时间段、候选成员所支持的算法类型、候选成员的精度信息、候选成员的无线接入制式、候选成员的信号质量、候选成员的流量信息和候选成员可进行联邦学习的时间段。
可选地，该方法还包括：
第二网元根据候选成员的标识，确定参与联邦学习的目标成员。
本申请实施例提供的候选成员的确定方法，通过接收第二网元发送的请求消息，并根据该请求消息中包括的筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员，并将候选成员的标识信息发送给第二网元，其中，筛选信息包括如下至少一个：时间段、算法类型、精度阈值、无线接入制式、信号质量要求、流量范围、成员类型信息、数量信息、区域信息和联邦学习的类型信息。第一网元通过筛选信息，可以筛选出适合进行联邦学习的参与成员，从而提高联邦学习的效率。另外，第二网元还可以基于第一网元发送的候选成员，进一步筛选得到真正适合参与联邦学习的成员，提高了提高联邦学习的效率。
本实施例的方法，其具体实现过程与技术效果与第一网元侧方法实施例中类似，具体可以参见第一网元侧方法实施例中的详细介绍，此处不再赘述。
图4为本申请提供的一种候选成员的确定方法的信令图。如图4所示，该方法包括：
步骤401：第二网元向第一网元发送请求消息。
其中，请求消息中包括筛选信息，筛选信息包括如下至少一个：时间段、算法类型、精度阈值、无线接入制式、信号质量要求、流量范围、成员类型信息、数量信息、区域信息和联邦学习的类型信息。
步骤402：第一网元根据筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员。
步骤403：第一网元向第二网元发送响应消息，该响应消息中包括候选成员的标识。
本申请实施例提供的候选成员的确定方法，通过接收第二网元发送的请求消息，并根据该请求消息中包括的筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员，并将候选成员的标识信息发送给第二网元，其中，筛选信息包括如下至少一个：时间段、算法类型、精度阈值、无线接入制式、信号质量要求、流量范围、成员类型信息、数量信息、兴趣区域AOI和联邦学习的类型。第一网元通过筛选信息，可以筛选出适合进行联邦学习的参与成员，从而提高联邦学习的训练效率。
本实施例的方法，其具体实现过程与技术效果与第一网元侧方法实施例中类似，具体可以参见第一网元侧方法实施例中的详细介绍，此处不再赘述。
本申请实施例提供的候选成员的确定方法，执行主体可以为候选成员的确定装置。本申请实施例中以候选成员的确定装置执行候选成员的确定方法为例，说明本申请实施例提供的候选成员的确定装置。
图5是本申请提供的候选成员的确定装置的结构示意图之一。如图5所示，本实施例提供的候选成员的确定装置，包括：
接收模块11，用于接收第二网元发送的请求消息，所述请求消息中包括筛选信息；
处理模块12，用于根据所述筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员；
发送模块13，用于向所述第二网元发送响应消息，所述响应消息中包括所述候选成员的标识；
其中，所述筛选信息包括如下至少一个：
时间段，用于指示进行所述联邦学习的可选时间段；
算法类型，用于指示进行所述联邦学习需要支持的模型训练的算法类型；
精度阈值，用于指示进行所述联邦学习需满足的模型训练的精度要求；
无线接入制式，用于指示进行所述联邦学习需选择的无线接入制式；
信号质量要求，用于指示进行所述联邦学习时无线信号质量要求；
流量范围，用于指示所述候选成员的流量使用范围要求；
成员类型信息，用于指示参与联邦学习的候选成员的类型要求；
数量信息，用于指示所述候选成员的数量要求；
兴趣区域AOI，用于指示所述候选成员所处的区域；
联邦学习的类型信息，用于指示所述联邦学习的属于纵向联邦或横向联邦。
本实施例的装置，接收模块接收第二网元发送的请求消息，该请求消息中包括筛选信息；处理模块根据筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员；发送模块向所述第二网元发送响应消息，所述响应消息中包括所述候选成员的标识，其中，筛选信息包括如下至少一个：时间段、算法类型、精度阈值、无线接入制式、信号质量要求、流量范围、成员类型信息、数量信息、兴趣区域AOI和联邦学习的类型。候选成员的确定装置通过筛选信息，可以筛选出适合进行联邦学习的参与成员，从而提高联邦学习的训练效率。
可选地，所述第一设备包括以下至少一项：
在所述时间段内处于联网状态的第一设备；
支持在所述时间段进行联邦学习的第一设备；
支持所述算法类型的第一设备；
支持利用所述算法类型进行联邦学习的第一设备；
模型训练精度信息大于所述精度阈值的第一设备；
处于所述无线接入制式下的第一设备；
支持在所述无线接入制式下进行联邦学习的第一设备；
信号质量大于信号质量要求的第一设备；
所消耗流量处于所述流量范围的第一设备；
满足所述候选成员的类型要求的第一设备；
位于所述AOI内的第一设备；
联邦学习的类型与所述筛选信息中包括的联邦学习的类型信息相同的第一设备。可选地，所述处理模块12，具体用于：
确定所述筛选信息所对应的数据类型；
基于所述数据类型从至少一个第三网元中获取至少一个设备各自对应的属性信息；
根据所述属性信息和所述筛选信息，确定所述候选成员。
可选地，所述数据类型包括如下至少一项：
联网信息以及对应的时间段、算法类型、精度信息、无线接入制式、信号质量和流量。
可选地，所述处理模块12，具体用于：
从第三网元中获取第一指示信息，所述第一指示信息用于表示各第二设备参与联邦学习的意愿信息；其中，所述意愿信息表示所述各第二设备是否愿意参与联邦学习；
根据所述筛选信息和所述第一指示信息将一个或多个第一设备确定为所述候选成员，所述第一设备为第二设备中愿意参与联邦学习的且满足所述筛选信息的设备。
可选地，所述处理模块12，具体用于：
从第三网元中获取所述第一设备参与所述联邦学习的意愿信息；
根据所述意愿信息和所述筛选信息，将愿意参与所述联邦学习且与所述筛选信息匹配的第一设备确定为可参与联邦学习的候选成员。
可选地，所述处理模块12，具体用于：
获取所述第一设备的联邦学习的能力信息，所述能力信息包括如下信息中的至少一个：参与联邦学习的无线接入制式、参与联邦学习的区域、参与联邦学习的时间、参与联邦学习所能支持的算法信息、参与联邦学习所能达到的精度信息和参与联邦学习的类型；
根据所述能力信息和所述筛选信息，从所述第一设备中确定可参与联邦学习的候选成员，所述候选成员的能力信息与所述筛选信息匹配。
其中，所述候选成员的能力信息与所述筛选信息匹配，包括以下至少一项：
所述候选成员参与联邦学习的无线接入制式与所述筛选信息中包括的无线接入制式相同；
所述候选成员参与联邦学习的区域位于所述筛选信息中包括的AOI内；
所述候选成员参与联邦学习的时间在所述筛选信息中包括的时间段内；
所述候选成员参与联邦学习所能支持的算法类型包含于所述筛选信息中包括的算法类型；
所述候选成员参与联邦学习所能达到的精度信息高于所述筛选信息中包括的精度阈值；
所述候选成员参与联邦学习的类型与所述筛选信息中包括的联邦学习的类型相同。
可选地，所述处理模块12，具体用于：
获取所述第一设备对应的网络状态信息，所述网络状态信息包括以下至少一项：
所述第一设备的类型信息；
所述第一设备的位置信息；
所述第一设备所处的无线接入制式信息；
所述第一设备的无线信号质量信息；
根据所述第一设备对应的网络状态信息和所述筛选信息，从所述第一设备中确定可参与联邦学习的候选成员，所述候选成员对应的网络状态信息与所述筛选信息匹配。
其中，所述候选成员对应的网络状态信息与所述筛选信息匹配，包括以下至少一项：
所述候选成员所处的无线接入制式与所述筛选信息中包括的无线接入制式相同；
所述候选成员的无线信号质量信息大于所述筛选信息中包括的信号质量要求；
所述候选成员的位置位于所述筛选信息中包括的AOI内；
所述候选成员的类型包含于所述筛选信息中包括的成员类型信息。
可选地，所述处理模块，具体用于：
获取各第二设备参与联邦学习的能力信息，所述能力信息包括如下信息中的至少一个：参与联邦学习的无线接入制式、参与联邦学习的区域、参与联邦学习的时间、参与联邦学习所能支持的算法信息、参与联邦学习所能达到的精度信息和参与联邦学习的类型；根据所述筛选信息、所述第一指示信息和所述能力信息，将一个或多个第一设备确定为所述候选成员，所述第一设备为第二设备中愿意参与联邦学习的、联邦学习的能力信息满足所述筛选信息的设备。
其中，参与联邦学习的能力信息满足所述筛选信息包括：
候选成员参与联邦学习的无线制式与所述筛选信息中包括的无线接入制式相同；
所述候选成员参与联邦学习的时间在所述筛选信息中包括的时间段内；
所述候选成员参与联邦学习所能支持的算法类型包含于所述筛选信息中包括的算法类型；
所述候选成员参与联邦学习所能达到的精度信息高于所述筛选信息中包 括的精度阈值；
所述候选成员参与联邦学习的类型与所述筛选信息中包括的联邦学习的类型相同。可选地，所述请求消息中还包括第二指示信息，所述第二指示信息用于指示联邦学习对应的业务类型；
所述处理模块，具体用于：
根据所述筛选信息和所述第二指示信息，将一个或多个第一设备确定为所述候选成员；
其中，所述第一设备为可支持业务类型对应业务的设备。
可选地，所述请求消息中还包括排序指示信息，所述排序指示信息用于指示对所述候选成员按照第二信息进行排序，所述第二信息包括如下至少一项：所述候选成员的信号质量、所述候选成员的精度信息和所述候选成员的流量信息；
所述响应消息中的候选成员的标识顺序为按照所述第二信息进行排序后得到的顺序。
可选地，所述请求消息中还包括分组指示信息，所述分组指示信息用于指示对所述候选成员按照第三信息进行分组，所述第三信息包括如下信息中的至少一种：
所述候选成员所处的区域、所述候选成员处于联网状态的时间段、所述候选成员所支持的算法类型、所述候选成员的精度信息、所述候选成员的无线接入制式和所述候选成员的信号质量；
所述响应消息中的候选成员的标识为按照所述第三信息进行分组后得到的标识。
可选地，所述响应消息中还包括如下信息中的至少一种：所述候选成员所处的区域、所述候选成员处于联网状态的时间段、所述候选成员所支持的算法类型、所述候选成员的精度信息、所述候选成员的无线接入制式、所述候选成员的信号质量、所述候选成员的流量信息和所述候选成员可进行联邦学习的时间段。
本实施例的装置，可以用于执行前述第一网元侧方法实施例中任一实施例的方法，其具体实现过程与技术效果与第一网元侧方法实施例中类似，具体可以参见第一网元侧方法实施例中的详细介绍，此处不再赘述。
图6是本申请提供的候选成员的确定装置的结构示意图之二。如图6所示，本实施例提供的候选成员的确定装置，包括：
发送模块21，用于向第一网元发送请求消息，所述请求消息中包括筛选信息，所述请求消息用于指示所述第一网元根据所述筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员；
接收模块22，用于接收所述第一网元发送的响应消息，所述响应消息中包括所述候选成员的标识；
其中，所述筛选信息包括如下至少一个：
时间段，用于指示进行所述联邦学习的可选时间段；
算法类型，用于指示进行所述联邦学习需要支持的模型训练的算法类型；
精度阈值，用于指示进行所述联邦学习需满足的模型训练的精度要求；
无线接入制式，用于指示进行所述联邦学习需选择的无线接入制式；
信号质量要求，用于指示进行所述联邦学习时无线信号质量要求；
流量范围，用于指示所述候选成员的流量使用范围要求；
成员类型信息，用于指示参与联邦学习的候选成员的类型要求；
数量信息，用于指示所述候选成员的数量要求；
兴趣区域AOI，用于指示所述候选成员所处的区域；
联邦学习的类型信息，用于指示所述联邦学习的属于纵向联邦或横向联邦。
可选地，所述第一设备包括以下至少一项：
在所述时间段内处于联网状态的第一设备；
支持在所述时间段进行联邦学习的第一设备；
支持所述算法类型的第一设备；
支持利用所述算法类型进行联邦学习的第一设备；
模型训练精度信息大于所述精度阈值的第一设备；
处于所述无线接入制式下的第一设备；
支持在所述无线接入制式下进行联邦学习的第一设备；
信号质量大于信号质量要求的第一设备；
所消耗流量处于所述流量范围的第一设备；
满足所述候选成员的类型要求的第一设备；
位于所述AOI内的第一设备；
联邦学习的类型与所述筛选信息中包括的联邦学习的类型信息相同的第一设备。可选地，所述请求消息中还包括：指示信息，所述指示信息用于指示执行参与联邦学习的候选成员的确定任务，或用于指示此次任务是用于联邦学习或者联邦学习的成员选择的。
可选地，所述请求消息中还包括排序指示信息，所述排序指示信息用于指示对所述候选成员按照第二信息进行排序，所述第二信息包括如下至少一项：所述候选成员的信号质量、所述候选成员的精度信息和所述候选成员的流量信息；
所述响应消息中的候选成员的标识为按照所述第二信息进行排序后得到的标识。
可选地，所述请求消息中还包括分组指示信息，所述分组指示信息用于指示对所述候选成员按照第三信息进行分组，所述第三信息包括如下信息中的至少一种：所述候选成员所处的区域、所述候选成员处于联网状态的时间段、所述候选成员所支持的算法类型、所述候选成员的精度信息、所述候选成员的无线接入制式和所述候选成员的信号质量；
所述响应消息中的候选成员的标识为按照所述第三信息进行分组后得到的标识。
可选地，所述响应消息中还包括所述候选成员所处的区域、所述候选成员处于联网状态的时间段、所述候选成员所支持的算法类型、所述候选成员的精度信息、所述候选成员的无线接入制式、所述候选成员的信号质量、所 述候选成员的流量信息和所述候选成员可进行联邦学习的时间段。
可选地，所述装置还包括：处理模块23；
处理模块23，用于根据所述候选成员的标识，确定参与联邦学习的目标成员。
本实施例的装置，可以用于执行前述第二网元侧方法实施例中任一实施例的方法，其具体实现过程与技术效果与第二网元侧方法实施例中类似，具体可以参见第二网元侧方法实施例中的详细介绍，此处不再赘述。
本申请实施例中的候选成员的确定装置可以是电子设备，例如具有操作系统的电子设备，也可以是电子设备中的部件，例如集成电路或芯片。该电子设备可以是终端，也可以为除终端之外的其他设备。示例性的，终端可以包括但不限于上述所列举的终端11的类型，其他设备可以为服务器、网络附属存储器(Network Attached Storage，NAS)等，本申请实施例不作具体限定。
本申请实施例提供的候选成员的确定装置能够实现图2至图4的方法实施例实现的各个过程，并达到相同的技术效果，为避免重复，这里不再赘述。
可选的，如图7所示，本申请实施例还提供一种通信设备700，包括处理器701和存储器702，存储器702上存储有可在所述处理器701上运行的程序或指令，例如，该通信设备700为终端时，该程序或指令被处理器701执行时实现上述候选成员的确定方法实施例的各个步骤，且能达到相同的技术效果。该通信设备700为网络侧设备时，该程序或指令被处理器701执行时实现上述候选成员的确定方法实施例的各个步骤，且能达到相同的技术效果，为避免重复，这里不再赘述。
本申请实施例还提供一种第一网元，包括处理器和通信接口，处理器用于根据筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员，通信接口用于接收第二网元发送的请求消息，向第二网元发送响应消息，所述响应消息中包括所述候选成员的标识，其中，所述请求消息中包括筛选信息，所述筛选信息包括如下至少一个：时间段、算法类型、精度阈值、无线接入制式、信号质量要求、流量范围、成员类型信息、数量信息、区域信息 和联邦学习的类型信息。该终端实施例与上述终端侧方法实施例对应，上述方法实施例的各个实施过程和实现方式均可适用于该终端实施例中，且能达到相同的技术效果。具体地，图8为实现本申请实施例的一种第一网元的硬件结构示意图。
该第一网元800包括但不限于：射频单元801、网络模块802、音频输出单元803、输入单元804、传感器805、显示单元806、用户输入单元807、接口单元808、存储器809以及处理器810等中的至少部分部件。
本领域技术人员可以理解，第一网元800还可以包括给各个部件供电的电源(比如电池)，电源可以通过电源管理系统与处理器810逻辑相连，从而通过电源管理系统实现管理充电、放电、以及功耗管理等功能。图8中示出的第一网元结构并不构成对第一网元的限定，第一网元可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置，在此不再赘述。
应理解的是，本申请实施例中，输入单元804可以包括图形处理单元(Graphics Processing Unit，GPU)8041和麦克风8042，图形处理器8041对在视频捕获模式或图像捕获模式中由图像捕获装置(如摄像头)获得的静态图片或视频的图像数据进行处理。显示单元806可包括显示面板8061，可以采用液晶显示器、有机发光二极管等形式来配置显示面板8061。用户输入单元807包括触控面板8071以及其他输入设备8072中的至少一种。触控面板8 071，也称为触摸屏。触控面板8071可包括触摸检测装置和触摸控制器两个部分。其他输入设备8072可以包括但不限于物理键盘、功能键(比如音量控制按键、开关按键等)、轨迹球、鼠标、操作杆，在此不再赘述。
本申请实施例中，射频单元801接收来自网络侧设备的下行数据后，可以传输给处理器810进行处理；另外，射频单元801可以向网络侧设备发送上行数据。通常，射频单元801包括但不限于天线、放大器、收发信机、耦合器、低噪声放大器、双工器等。
存储器809可用于存储软件程序或指令以及各种数据。存储器809可主要包括存储程序或指令的第一存储区和存储数据的第二存储区，其中，第一 存储区可存储操作系统、至少一个功能所需的应用程序或指令(比如声音播放功能、图像播放功能等)等。此外，存储器809可以包括易失性存储器或非易失性存储器，或者，存储器809可以包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(Read-Only Memory，ROM)、可编程只读存储器(Programmable ROM，PROM)、可擦除可编程只读存储器(Erasable PROM，EPROM)、电可擦除可编程只读存储器(Electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(Random Access Memory，RAM)，静态随机存取存储器(Static RAM，SRAM)、动态随机存取存储器(Dynamic RAM，DRAM)、同步动态随机存取存储器(Synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(Double Data Rate SDRAM，DDRSDRAM)、增强型同步动态随机存取存储器(Enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(Synch link DRAM，SLDRAM)和直接内存总线随机存取存储器(Direct Rambus RAM，DRRAM)。本申请实施例中的存储器x09包括但不限于这些和任意其它适合类型的存储器。
处理器810可包括一个或多个处理单元；可选的，处理器810集成应用处理器和调制解调处理器，其中，应用处理器主要处理涉及操作系统、用户界面和应用程序等的操作，调制解调处理器主要处理无线通信信号，如基带处理器。可以理解的是，上述调制解调处理器也可以不集成到处理器810中。
其中，射频单元801，用于接收第二网元发送的请求消息，所述请求消息中包括筛选信息，并向第二网元发送响应消息，所述响应消息中包括所述候选成员的标识。
处理器810，用于根据筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员，所述筛选信息包括如下至少一个：
时间段，用于指示进行所述联邦学习的可选时间段；
算法类型，用于指示进行所述联邦学习需要支持的模型训练的算法类型；
精度阈值，用于指示进行所述联邦学习需满足的模型训练的精度要求；
无线接入制式，用于指示进行所述联邦学习需选择的无线接入制式；
信号质量要求，用于指示进行所述联邦学习时无线信号质量要求；
流量范围，用于指示所述候选成员的流量使用范围要求；
成员类型信息，用于指示参与联邦学习的候选成员的类型要求；
数量信息，用于指示所述候选成员的数量要求；
兴趣区域AOI，用于指示所述候选成员所处的区域；
联邦学习的类型信息，用于指示所述联邦学习的属于纵向联邦或横向联邦。
上述实施方案中，第一网元通过接收第二网元发送的请求消息，并根据该请求消息中包括的筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员，并将候选成员的标识信息发送给第二网元，其中，筛选信息包括如下至少一个：时间段、算法类型、精度阈值、无线接入制式、信号质量要求、流量范围、成员类型信息、数量信息、兴趣区域AOI和联邦学习的类型。第一网元通过筛选信息，可以筛选出适合进行联邦学习的参与成员，从而提高联邦学习的训练效率。
可选地，所述第一设备包括以下至少一项：
在所述时间段内处于联网状态的第一设备；
支持在所述时间段进行联邦学习的第一设备；
支持所述算法类型的第一设备；
支持利用所述算法类型进行联邦学习的第一设备；
模型训练精度信息大于所述精度阈值的第一设备；
处于所述无线接入制式下的第一设备；
支持在所述无线接入制式下进行联邦学习的第一设备；
信号质量大于信号质量要求的第一设备；
所消耗流量处于所述流量范围的第一设备；
满足所述候选成员的类型要求的第一设备；
位于所述AOI内的第一设备；
联邦学习的类型与所述筛选信息中包括的联邦学习的类型信息相同的第 一设备。
可选地，处理器810，还用于确定所述筛选信息所对应的数据类型；
基于所述数据类型从至少一个第三网元中获取至少一个第二设备各自对应的属性信息；
根据所述属性信息和所述筛选信息，确定所述候选成员。
可选地，所述数据类型包括如下至少一项：
联网信息以及对应的时间段、算法类型、精度信息、无线接入制式、信号质量和流量。
可选地，处理器810，还用于从所述第三网元中获取第一指示信息，所述第一指示信息用于表示各第二设备参与联邦学习的意愿信息；其中，所述意愿信息表示所述各设备是否愿意参与联邦学习；
根据所述筛选信息和所述第一指示信息，将一个或多个第一设备确定为所述候选成员，所述第一设备为第二设备中愿意参与联邦学习的且满足所述筛选信息的设备。
可选地，处理器810，用于从第三网元中获取所述第一设备参与所述联邦学习的意愿信息；根据所述意愿信息和所述筛选信息，将愿意参与所述联邦学习且与所述筛选信息匹配的第一设备确定为可参与联邦学习的候选成员。
可选地，处理器810，用于获取所述第一设备的联邦学习的能力信息，所述能力信息包括如下信息中的至少一个：参与联邦学习的无线接入制式、参与联邦学习的区域、参与联邦学习的时间、参与联邦学习所能支持的算法信息、参与联邦学习所能达到的精度信息和参与联邦学习的类型；根据所述能力信息和所述筛选信息，从所述第一设备中确定可参与联邦学习的候选成员，所述候选成员的能力信息与所述筛选信息匹配。
其中，所述候选成员的能力信息与所述筛选信息匹配，包括以下至少一项：
所述候选成员参与联邦学习的无线接入制式与所述筛选信息中包括的无线接入制式相同；
所述候选成员参与联邦学习的区域位于所述筛选信息中包括的AOI内；
所述候选成员参与联邦学习的时间在所述筛选信息中包括的时间段内；
所述候选成员参与联邦学习所能支持的算法类型包含于所述筛选信息中包括的算法类型；
所述候选成员参与联邦学习所能达到的精度信息高于所述筛选信息中包括的精度阈值；
所述候选成员参与联邦学习的类型与所述筛选信息中包括的联邦学习的类型相同。
可选地，处理器810，用于获取所述第一设备对应的网络状态信息，所述网络状态信息包括以下至少一项：
所述第一设备的类型信息；
所述第一设备的位置信息；
所述第一设备所处的无线接入制式信息；
所述第一设备的无线信号质量信息；
根据所述第一设备对应的网络状态信息和所述筛选信息，从所述第一设备中确定可参与联邦学习的候选成员，所述候选成员对应的网络状态信息与所述筛选信息匹配。
其中，所述候选成员对应的网络状态信息与所述筛选信息匹配，包括以下至少一项：
所述候选成员所处的无线接入制式与所述筛选信息中包括的无线接入制式相同；
所述候选成员的无线信号质量信息大于所述筛选信息中包括的信号质量要求；
所述候选成员的位置位于所述筛选信息中包括的AOI内；
所述候选成员的类型包含于所述筛选信息中包括的成员类型信息。
可选地，处理器810，用于获取各第二设备参与联邦学习的能力信息，所述能力信息包括如下信息中的至少一个：参与联邦学习的无线接入制式、 参与联邦学习的区域、参与联邦学习的时间、参与联邦学习所能支持的算法信息、参与联邦学习所能达到的精度信息和参与联邦学习的类型；根据所述筛选信息、所述第一指示信息和所述能力信息，将一个或多个第一设备确定为所述候选成员，所述第一设备为第二设备中愿意参与联邦学习的、联邦学习的能力信息满足所述筛选信息的设备。
其中，参与联邦学习的能力信息满足所述筛选信息包括：
候选成员参与联邦学习的无线制式与所述筛选信息中包括的无线接入制式相同；
所述候选成员参与联邦学习的时间在所述筛选信息中包括的时间段内；
所述候选成员参与联邦学习所能支持的算法类型包含于所述筛选信息中包括的算法类型；
所述候选成员参与联邦学习所能达到的精度信息高于所述筛选信息中包括的精度阈值；
所述候选成员参与联邦学习的类型与所述筛选信息中包括的联邦学习的类型相同。
可选地，所述请求消息中还包括第二指示信息，所述第二指示信息用于指示联邦学习对应的业务类型；
处理器810，还用于根据所述筛选信息和所述第二指示信息，将一个或多个第一设备确定为所述候选成员；
其中，所述第一设备为可支持业务类型对应业务的设备。
可选地，所述请求消息中还包括排序指示信息，所述排序指示信息用于指示对所述候选成员按照第二信息进行排序，所述第二信息包括如下至少一项：所述候选成员的信号质量、所述候选成员的精度信息和所述候选成员的流量信息；
所述响应消息中的候选成员的标识顺序为按照所述第二信息进行排序后得到的顺序。
可选地，所述请求消息中还包括分组指示信息，所述分组指示信息用于 指示对所述候选成员按照第三信息进行分组，所述第三信息包括如下信息中的至少一种：
所述候选成员所处的区域、所述候选成员处于联网状态的时间段、所述候选成员所支持的算法类型、所述候选成员的精度信息、所述候选成员的无线接入制式和所述候选成员的信号质量；
所述响应消息中的候选成员的标识为按照所述第三信息进行分组后得到的标识。
可选地，所述响应消息中还包括如下信息中的至少一种：所述候选成员所处的区域、所述候选成员处于联网状态的时间段、所述候选成员所支持的算法类型、所述候选成员的精度信息、所述候选成员的无线接入制式、所述候选成员的信号质量、所述候选成员的流量信息和所述候选成员可进行联邦学习的时间段。
上述实施方案中，第一网元通过接收第二网元发送的请求消息，并根据该请求消息中包括的筛选信息，将一个或多个第一设备确定为可参与联邦学习的候选成员，并将候选成员的标识信息发送给第二网元，其中，筛选信息包括如下至少一个：时间段、算法类型、精度阈值、无线接入制式、信号质量要求、流量范围、成员类型信息、数量信息、兴趣区域AOI和联邦学习的类型。第一网元通过筛选信息，可以筛选出适合进行联邦学习的参与成员，从而提高联邦学习的训练效率。
具体地，本申请实施例还提供了一种第二网元。如图9所示，该第二网元900包括：处理器901、网络接口902和存储器903。其中，网络接口902例如为通用公共无线接口(common public radio interface，CPRI)。
具体地，本申请实施例的网络侧设备900还包括：存储在存储器903上并可在处理器901上运行的指令或程序，处理器901调用存储器903中的指令或程序执行图6所示各模块执行的方法，并达到相同的技术效果，为避免重复，故不在此赘述。
本申请实施例还提供一种可读存储介质，所述可读存储介质上存储有程 序或指令，该程序或指令被处理器执行时实现上述候选成员的确定方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
其中，所述处理器为上述实施例中所述的终端中的处理器。所述可读存储介质，包括计算机可读存储介质，如计算机只读存储器ROM、随机存取存储器RAM、磁碟或者光盘等。
本申请实施例另提供了一种芯片，所述芯片包括处理器和通信接口，所述通信接口和所述处理器耦合，所述处理器用于运行程序或指令，实现上述候选成员的确定方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
应理解，本申请实施例提到的芯片还可以称为系统级芯片，系统芯片，芯片系统或片上系统芯片等。
本申请实施例另提供了一种计算机程序/程序产品，所述计算机程序/程序产品被存储在存储介质中，所述计算机程序/程序产品被至少一个处理器执行以实现上述候选成员的确定方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
本申请实施例还提供了一种候选成员的确定系统，包括：第一网元及第二网元，所述第一网元可用于执行如上所述的候选成员的确定方法的步骤，所述第二网元可用于执行如上所述的候选成员的确定方法的步骤。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。此外，需要指出的是，本申请实施方式中的方法和装置的范围不限按示出或讨论的顺序来执行功能，还可包括根据所涉及的功能按基本同时的方式或按相反的顺序来执行功能，例如，可以按不同于所描述的次序来执行所描述的方法，并且还可以添加、省 去、或组合各种步骤。另外，参照某些示例所描述的特征可在其他示例中被组合。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡献的部分可以以计算机软件产品的形式体现出来，该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端(可以是手机，计算机，服务器，空调器，或者网络设备等)执行本申请各个实施例所述的方法。
上面结合附图对本申请的实施例进行了描述，但是本申请并不局限于上述的具体实施方式，上述的具体实施方式仅仅是示意性的，而不是限制性的，本领域的普通技术人员在本申请的启示下，在不脱离本申请宗旨和权利要求所保护的范围情况下，还可做出很多形式，均属于本申请的保护之内。

Claims (26)

  1. 一种候选成员的确定方法，包括：
    第一网元接收第二网元发送的请求消息，所述请求消息中包括筛选信息；
    所述第一网元根据所述筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员；
    所述第一网元向所述第二网元发送响应消息，所述响应消息中包括所述候选成员的标识；
    其中，所述筛选信息包括如下至少一个：
    时间段，用于指示进行所述联邦学习的可选时间段；
    算法类型，用于指示进行所述联邦学习需要支持的模型训练的算法类型；
    精度阈值，用于指示进行所述联邦学习需满足的模型训练的精度要求；
    无线接入制式，用于指示进行所述联邦学习需选择的无线接入制式；
    信号质量要求，用于指示进行所述联邦学习时无线信号质量要求；
    流量范围，用于指示所述候选成员的流量使用范围要求；
    成员类型信息，用于指示参与联邦学习的候选成员的类型要求；
    数量信息，用于指示所述候选成员的数量要求；
    兴趣区域AOI，用于指示所述候选成员所处的区域；
    联邦学习的类型信息，用于指示所述联邦学习的属于纵向联邦或横向联邦。
  2. 根据权利要求1所述的方法，其中，所述请求消息中还包括：
    指示信息，所述指示信息用于指示执行参与联邦学习的候选成员的确定任务，或用于指示此次任务是用于联邦学习或者联邦学习的成员选择的。
  3. 根据权利要求1或2所述的方法，其中，所述第一设备包括以下至少一项：
    在所述时间段内处于联网状态的第一设备；
    支持在所述时间段进行联邦学习的第一设备；
    支持所述算法类型的第一设备；
    支持利用所述算法类型进行联邦学习的第一设备；
    模型训练精度信息大于所述精度阈值的第一设备；
    处于所述无线接入制式下的第一设备；
    支持在所述无线接入制式下进行联邦学习的第一设备；
    信号质量大于信号质量要求的第一设备；
    所消耗流量处于所述流量范围的第一设备；
    满足所述候选成员的类型要求的第一设备；
    位于所述AOI内的第一设备；
    联邦学习的类型与所述筛选信息中包括的联邦学习的类型信息相同的第一设备。
  4. 根据权利要求3所述的方法，其中，所述第一网元根据所述筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员，包括：
    第一网元从第三网元中获取第一指示信息，所述第一指示信息用于表示各第二设备参与联邦学习的意愿信息；其中，所述意愿信息表示所述各第二设备是否愿意参与联邦学习；
    根据所述筛选信息和所述第一指示信息，将一个或多个第一设备确定为所述候选成员，所述第一设备为第二设备中愿意参与联邦学习的且满足所述筛选信息的设备。
  5. 根据权利要求1至3中任一项所述的方法，其中，所述第一网元根据所述筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员，包括：
    所述第一网元从第三网元中获取所述第一设备参与所述联邦学习的意愿信息；
    所述第一网元根据所述意愿信息和所述筛选信息，将愿意参与所述联邦学习且与所述筛选信息匹配的第一设备确定为可参与联邦学习的候选成员。
  6. 根据权利要求1至5中任一项所述的方法，其中，所述第一网元根据 所述筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员，包括：
    所述第一网元获取所述第一设备的联邦学习的能力信息，所述能力信息包括如下信息中的至少一个：参与联邦学习的无线接入制式、参与联邦学习的区域、参与联邦学习的时间、参与联邦学习所能支持的算法信息、参与联邦学习所能达到的精度信息和参与联邦学习的类型；
    所述第一网元根据所述能力信息和所述筛选信息，从所述第一设备中确定可参与联邦学习的候选成员，所述候选成员的能力信息与所述筛选信息匹配。
  7. 根据权利要求6所述的方法，其中，所述候选成员的能力信息与所述筛选信息匹配，包括以下至少一项：
    所述候选成员参与联邦学习的无线接入制式与所述筛选信息中包括的无线接入制式相同；
    所述候选成员参与联邦学习的区域位于所述筛选信息中包括的AOI内；
    所述候选成员参与联邦学习的时间在所述筛选信息中包括的时间段内；
    所述候选成员参与联邦学习所能支持的算法类型包含于所述筛选信息中包括的算法类型；
    所述候选成员参与联邦学习所能达到的精度信息高于所述筛选信息中包括的精度阈值；
    所述候选成员参与联邦学习的类型与所述筛选信息中包括的联邦学习的类型相同。
  8. 根据权利要求1至7中任一项所述的方法，其中，所述第一网元根据所述筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员，包括：
    所述第一网元获取所述第一设备对应的网络状态信息，所述网络状态信息包括以下至少一项：
    所述第一设备的类型信息；
    所述第一设备的位置信息；
    所述第一设备所处的无线接入制式信息；
    所述第一设备的无线信号质量信息；
    所述第一网元根据所述第一设备对应的网络状态信息和所述筛选信息，从所述第一设备中确定可参与联邦学习的候选成员，所述候选成员对应的网络状态信息与所述筛选信息匹配。
  9. 根据权利要求8所述的方法，其中，所述候选成员对应的网络状态信息与所述筛选信息匹配，包括以下至少一项：
    所述候选成员所处的无线接入制式与所述筛选信息中包括的无线接入制式相同；
    所述候选成员的无线信号质量信息大于所述筛选信息中包括的信号质量要求；
    所述候选成员的位置位于所述筛选信息中包括的AOI内；
    所述候选成员的类型包含于所述筛选信息中包括的成员类型信息。
  10. 根据权利要求4所述的方法，其中，所述根据所述筛选信息和所述第一指示信息，将一个或多个第一设备确定为所述候选成员，包括：
    所述第一网元获取各第二设备参与联邦学习的能力信息，所述能力信息包括如下信息中的至少一个：参与联邦学习的无线接入制式、参与联邦学习的区域、参与联邦学习的时间、参与联邦学习所能支持的算法信息、参与联邦学习所能达到的精度信息和参与联邦学习的类型；
    根据所述筛选信息、所述第一指示信息和所述能力信息，将一个或多个第一设备确定为所述候选成员，所述第一设备为第二设备中愿意参与联邦学习的、联邦学习的能力信息满足所述筛选信息的设备。
  11. 根据权利要求1至10任一项所述的方法，其中，所述请求消息中还包括第二指示信息，所述第二指示信息用于指示联邦学习对应的业务类型；
    所述第一网元根据所述筛选信息将一个或多个第一设备，确定为可参与联邦学习的候选成员，包括：
    所述第一网元根据所述筛选信息和所述第二指示信息，将一个或多个第一设备确定为所述候选成员；
    其中，所述第一设备为可支持业务类型对应业务的设备。
  12. 根据权利要求1至10任一项所述的方法，其中，所述请求消息中还包括排序指示信息，所述排序指示信息用于指示对所述候选成员按照第二信息进行排序，所述第二信息包括如下至少一项：所述候选成员的信号质量、所述候选成员的精度信息和所述候选成员的流量信息；
    所述响应消息中的候选成员的标识顺序为按照所述第二信息进行排序后得到的顺序。
  13. 根据权利要求1至10任一项所述的方法，其中，所述请求消息中还包括分组指示信息，所述分组指示信息用于指示对所述候选成员按照第三信息进行分组，所述第三信息包括如下信息中的至少一种：
    所述候选成员所处的区域、所述候选成员处于联网状态的时间段、所述候选成员所支持的算法类型、所述候选成员的精度信息、所述候选成员的无线接入制式和所述候选成员的信号质量；
    所述响应消息中的候选成员的标识为按照所述第三信息进行分组后得到的标识。
  14. 根据权利要求1至10任一项所述的方法，其中，所述响应消息中还包括如下信息中的至少一种：所述候选成员所处的区域、所述候选成员处于联网状态的时间段、所述候选成员所支持的算法类型、所述候选成员的精度信息、所述候选成员的无线接入制式、所述候选成员的信号质量、所述候选成员的流量信息和所述候选成员可进行联邦学习的时间段。
  15. 一种候选成员的确定方法，包括：
    第二网元向第一网元发送请求消息，所述请求消息中包括筛选信息，所述请求消息用于指示所述第一网元根据所述筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员；
    所述第二网元接收所述第一网元发送的响应消息，所述响应消息中包括 所述候选成员的标识；
    其中，所述筛选信息包括如下至少一个：
    时间段，用于指示进行所述联邦学习的可选时间段；
    算法类型，用于指示进行所述联邦学习需要支持的模型训练的算法类型；
    精度阈值，用于指示进行所述联邦学习需满足的模型训练的精度要求；
    无线接入制式，用于指示进行所述联邦学习需选择的无线接入制式；
    信号质量要求，用于指示进行所述联邦学习时无线信号质量要求；
    流量范围，用于指示所述候选成员的流量使用范围要求；
    成员类型信息，用于指示参与联邦学习的候选成员的类型要求；
    数量信息，用于指示所述候选成员的数量要求；
    兴趣区域AOI，用于指示所述候选成员所处的区域；
    联邦学习的类型信息，用于指示所述联邦学习的属于纵向联邦或横向联邦。
  16. 根据权利要求15所述的方法，其中，所述请求消息中还包括：
    指示信息，所述指示信息用于指示执行参与联邦学习的候选成员的确定任务，或用于指示此次任务是用于联邦学习或者联邦学习的成员选择的。
  17. 根据权利要求15或16所述的方法，其中，所述第一设备包括以下至少一项：
    在所述时间段内处于联网状态的第一设备；
    支持在所述时间段进行联邦学习的第一设备；
    支持所述算法类型的第一设备；
    支持利用所述算法类型进行联邦学习的第一设备；
    模型训练精度信息大于所述精度阈值的第一设备；
    处于所述无线接入制式下的第一设备；
    支持在所述无线接入制式下进行联邦学习的第一设备；
    信号质量大于信号质量要求的第一设备；
    所消耗流量处于所述流量范围的第一设备；
    满足所述候选成员的类型要求的第一设备；
    位于所述AOI内的第一设备；
    联邦学习的类型与所述筛选信息中包括的联邦学习的类型信息相同的第一设备。
  18. 根据权利要求15至17中任一项所述的方法，其中，所述请求消息中还包括排序指示信息，所述排序指示信息用于指示对所述候选成员按照第二信息进行排序，所述第二信息包括如下至少一项：所述候选成员的信号质量、所述候选成员的精度信息和所述候选成员的流量信息；
    所述响应消息中的候选成员的标识为按照所述第二信息进行排序后得到的标识。
  19. 根据权利要求15至17中任一项所述的方法，其中，所述请求消息中还包括分组指示信息，所述分组指示信息用于指示对所述候选成员按照第三信息进行分组，所述第三信息包括如下信息中的至少一种：所述候选成员所处的区域、所述候选成员处于联网状态的时间段、所述候选成员所支持的算法类型、所述候选成员的精度信息、所述候选成员的无线接入制式和所述候选成员的信号质量；
    所述响应消息中的候选成员的标识为按照所述第三信息进行分组后得到的标识。
  20. 根据权利要求15至17中任一项所述的方法，其中，所述响应消息中还包括所述候选成员所处的区域、所述候选成员处于联网状态的时间段、所述候选成员所支持的算法类型、所述候选成员的精度信息、所述候选成员的无线接入制式、所述候选成员的信号质量、所述候选成员的流量信息和所述候选成员可进行联邦学习的时间段。
  21. 根据权利要求17或18所述的方法，其中，所述方法还包括：
    所述第二网元根据所述候选成员的标识，确定参与联邦学习的目标成员。
  22. 一种候选成员的确定装置，包括：
    接收模块，用于接收第二网元发送的请求消息，所述请求消息中包括筛 选信息；
    处理模块，用于根据所述筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员；
    发送模块，用于向所述第二网元发送响应消息，所述响应消息中包括所述候选成员的标识；
    其中，所述筛选信息包括如下至少一个：
    时间段，用于指示进行所述联邦学习的可选时间段；
    算法类型，用于指示进行所述联邦学习需要支持的模型训练的算法类型；
    精度阈值，用于指示进行所述联邦学习需满足的模型训练的精度要求；
    无线接入制式，用于指示进行所述联邦学习需选择的无线接入制式；
    信号质量要求，用于指示进行所述联邦学习时无线信号质量要求；
    流量范围，用于指示所述候选成员的流量使用范围要求；
    成员类型信息，用于指示参与联邦学习的候选成员的类型要求；
    数量信息，用于指示所述候选成员的数量要求；
    兴趣区域AOI，用于指示所述候选成员所处的区域；
    联邦学习的类型信息，用于指示所述联邦学习的属于纵向联邦或横向联邦。
  23. 一种候选成员的确定装置，包括：
    发送模块，用于向第一网元发送请求消息，所述请求消息中包括筛选信息，所述请求消息用于指示所述第一网元根据所述筛选信息将一个或多个第一设备确定为可参与联邦学习的候选成员；
    接收模块，用于接收所述第一网元发送的响应消息，所述响应消息中包括所述候选成员的标识；
    其中，所述筛选信息包括如下至少一个：
    时间段，用于指示进行所述联邦学习的可选时间段；
    算法类型，用于指示进行所述联邦学习需要支持的模型训练的算法类型；
    精度阈值，用于指示进行所述联邦学习需满足的模型训练的精度要求；
    无线接入制式，用于指示进行所述联邦学习需选择的无线接入制式；
    信号质量要求，用于指示进行所述联邦学习时无线信号质量要求；
    流量范围，用于指示所述候选成员的流量使用范围要求；
    成员类型信息，用于指示参与联邦学习的候选成员的类型要求；
    数量信息，用于指示所述候选成员的数量要求；
    兴趣区域AOI，用于指示所述候选成员所处的区域；
    联邦学习的类型信息，用于指示所述联邦学习的属于纵向联邦或横向联邦。
  24. 一种第一网元，包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如权利要求1至14任一项所述的候选成员的确定方法的步骤。
  25. 一种第二网元，包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如权利要求15至21任一项所述的候选成员的确定方法的步骤。
  26. 一种可读存储介质，所述可读存储介质上存储程序或指令，所述程序或指令被处理器执行时实现如权利要求1至14任一项所述的候选成员的确定方法，或者实现如权利要求15至21任一项所述的候选成员的确定方法的步骤。
PCT/CN2023/084244 2022-03-28 2023-03-28 候选成员的确定方法、装置及设备 WO2023185788A1 (zh)

Applications Claiming Priority (4)

Application Number Priority Date Filing Date Title
CN202210314821 2022-03-28
CN202210314821.7 2022-03-28
CN202210476433.9 2022-04-29
CN202210476433.9A CN116866882A (zh) 2022-03-28 2022-04-29 候选成员的确定方法、装置及设备

Publications (1)

Publication Number Publication Date
WO2023185788A1 true WO2023185788A1 (zh) 2023-10-05

Family

ID=88199132

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/084244 WO2023185788A1 (zh) 2022-03-28 2023-03-28 候选成员的确定方法、装置及设备

Country Status (1)

Country Link
WO (1) WO2023185788A1 (zh)

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112308720A (zh) * 2019-08-02 2021-02-02 深圳前海微众银行股份有限公司 一种确定联邦学习联盟中参与方的方法及装置
CN113191484A (zh) * 2021-04-25 2021-07-30 清华大学 基于深度强化学习的联邦学习客户端智能选取方法及系统
CN113642239A (zh) * 2021-07-16 2021-11-12 北京融数联智科技有限公司 一种联邦学习建模方法及系统
CN114079902A (zh) * 2020-08-13 2022-02-22 Oppo广东移动通信有限公司 联邦学习的方法和装置

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112308720A (zh) * 2019-08-02 2021-02-02 深圳前海微众银行股份有限公司 一种确定联邦学习联盟中参与方的方法及装置
CN114079902A (zh) * 2020-08-13 2022-02-22 Oppo广东移动通信有限公司 联邦学习的方法和装置
CN113191484A (zh) * 2021-04-25 2021-07-30 清华大学 基于深度强化学习的联邦学习客户端智能选取方法及系统
CN113642239A (zh) * 2021-07-16 2021-11-12 北京融数联智科技有限公司 一种联邦学习建模方法及系统

Similar Documents

Publication Publication Date Title
US12015936B2 (en) Proactively adjusting network infrastructure in response to reporting of real-time network performance
CN108011937A (zh) 消息推送方法、服务器、智能终端及计算机可读存储介质
WO2023169523A1 (zh) 数据采集方法、装置、终端及网络侧设备
WO2023185788A1 (zh) 候选成员的确定方法、装置及设备
WO2022160670A1 (zh) Wi-Fi网络的识别方法和Wi-Fi网络的识别装置
CN116846771A (zh) 业务操作方法、装置、终端及可读存储介质
CN116866882A (zh) 候选成员的确定方法、装置及设备
WO2023185818A1 (zh) 设备确定方法、装置及通信设备
WO2023207980A1 (zh) 模型信息获取方法、发送方法、装置、节点和储存介质
WO2023179558A1 (zh) 数据采集方法及通信设备
WO2023207870A1 (zh) 数据处理方法、终端及网络侧设备
WO2023213288A1 (zh) 模型获取方法及通信设备
WO2024078400A1 (zh) 模型请求方法、装置、通信设备及可读存储介质
WO2023151585A1 (zh) 终端目标面能力上报、获取方法、终端及网络设备
WO2024088119A1 (zh) 数据处理方法、装置、终端及网络侧设备
WO2023185725A1 (zh) 数据获取方法、装置及网络侧设备
WO2023207984A1 (zh) 行为处理方法、装置、终端、网络侧设备及介质
WO2023213246A1 (zh) 模型选择方法、装置及网络侧设备
WO2023185850A1 (zh) Ursp规则优化方法、装置、终端、网络侧设备及介质
WO2023169469A1 (zh) 数据采集方法、装置、终端及网络侧设备
WO2023185822A1 (zh) 网元注册方法、模型确定方法、装置、网元、通信系统及存储介质
US20240056362A1 (en) Apparatus, methods, and computer programs
WO2023179709A1 (zh) 信息处理方法、装置、通信设备及可读存储介质
WO2023169425A1 (zh) 通信网络中的数据处理方法及网络侧设备
WO2024051565A1 (zh) 选网处理方法、装置、终端及可读存储介质

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23778153

Country of ref document: EP

Kind code of ref document: A1