WO2023143082A1 - 一种用户设备选择方法、装置、芯片及模组设备 - Google Patents

一种用户设备选择方法、装置、芯片及模组设备 Download PDF

Info

Publication number
WO2023143082A1
WO2023143082A1 PCT/CN2023/071729 CN2023071729W WO2023143082A1 WO 2023143082 A1 WO2023143082 A1 WO 2023143082A1 CN 2023071729 W CN2023071729 W CN 2023071729W WO 2023143082 A1 WO2023143082 A1 WO 2023143082A1
Authority
WO
WIPO (PCT)
Prior art keywords
user equipment
message
federated learning
information
requirement information
Prior art date
Application number
PCT/CN2023/071729
Other languages
English (en)
French (fr)
Inventor
陈晓宇
韩立锋
Original Assignee
展讯通信（上海）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
Application filed by 展讯通信（上海）有限公司 filed Critical 展讯通信（上海）有限公司
Publication of WO2023143082A1 publication Critical patent/WO2023143082A1/zh

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/10—Scheduling measurement reports ; Arrangements for measurement reports
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
    • G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N20/00—Machine learning
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/08—Learning methods
    • G06N3/098—Distributed learning, e.g. federated learning
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L1/00—Arrangements for detecting or preventing errors in the information received
    • H04L1/12—Arrangements for detecting or preventing errors in the information received by using return channel
    • H04L1/16—Arrangements for detecting or preventing errors in the information received by using return channel in which the return channel carries supervisory signals, e.g. repetition request signals
    • H04L1/1607—Details of the supervisory signal
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L5/00—Arrangements affording multiple use of the transmission path
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L5/00—Arrangements affording multiple use of the transmission path
    • H04L5/003—Arrangements for allocating sub-channels of the transmission path
    • H04L5/0053—Allocation of signaling, i.e. of overhead other than pilot signals

Definitions

  • the present invention relates to the communication field, in particular to a user equipment selection method, device, chip and module equipment.
  • Federated learning can obtain an artificial intelligence (AI) model that meets the user's performance requirements through multi-user joint training on the basis of protecting user data privacy, so it has attracted attention in communication networks.
  • AI artificial intelligence
  • the present application provides a user equipment selection method, device, chip, and module equipment, which are beneficial to improving the stability of user equipment participating in federated learning and reducing the probability of interruption of federated learning.
  • the present application provides a user equipment selection method, the method comprising: sending a first message to a first candidate user equipment group; receiving a second message from a second candidate user equipment group, the first candidate user equipment group Including a second candidate user equipment group; wherein, the first message is used to indicate the requirement information of the federated learning, and the second message is used to indicate whether the corresponding user equipment meets the required information of the federated learning, and/or, the corresponding user equipment Measurement information; or, the first message is used to request the measurement information of the user equipment, and the second message includes the measurement information; the federated learning requirement information includes federated learning duration requirement information, and/or, federated learning requires user equipment Stability requirement information; based on the second message, determine a target user equipment group participating in federated learning, and each target user equipment in the target user equipment group meets the requirement information of federated learning.
  • the network device can select the user equipment participating in the federated learning according to the duration requirement of the federated learning or the stability requirement of the user equipment, thereby improving the stability of the user equipment participating in the federated learning and ensuring the User equipment can fully participate in federated learning, thereby reducing the probability of interruption of federated learning.
  • the federated learning requirement information on user equipment stability includes requirement information on user equipment mobility, and/or requirement information on user equipment operating status.
  • the federated learning requirement information on user equipment stability includes one or more of the following items: mobility level requirement information, user equipment location requirement information, computing power change rate requirement information or signal Mass change rate demand information.
  • the measurement information includes one or more of the following items: mobility level, location of the user equipment, rate of change of computing power, or rate of change of signal quality.
  • a first indication message is sent to the first candidate user equipment group, where the first indication message is used to indicate to update the measurement information.
  • the first indication message includes measurement requirement information.
  • a second indication message is sent to the target user equipment group, where the second indication message is used to indicate whether continuous monitoring meets the requirement information of federated learning; receiving a third indication message from the first target user equipment, The third indication message is used to indicate that the first target user equipment does not meet the requirement information of federated learning, and the first target user equipment is one of the target user equipment group; remove the first target user equipment from the target user equipment group ; Or, lowering the priority of the first target user equipment in the target user equipment group.
  • the second indication message further includes requirement information of federated learning.
  • a fourth indication message from another network device is received, where the fourth indication message is used to indicate measurement information of at least one user equipment, and where the fourth indication message is used to determine the first candidate user equipment Group.
  • a fifth indication message is sent to other network devices, where the fifth indication message is used to indicate measurement information of at least one user equipment.
  • the second target user equipment when handover is performed for the second target user equipment, the second target user equipment is released, and the second target user equipment is one of the target user equipment group; remove the second target user equipment.
  • a secondary node addition request message is sent to a neighboring base station, where the secondary node addition request message is used to establish a communication connection between a third target user equipment and a neighboring base station; the third target user equipment is a target user One of the device groups.
  • a sixth indication message is sent to a neighboring base station, where the sixth indication message is used to indicate to send the federated learning data of the third target user equipment; receive the federated learning data of the third target user equipment from the neighboring base station .
  • the present application provides a method for selecting a user equipment, the method comprising: receiving a first message from a network device; sending a second message to the network device; wherein the first message is used to indicate federated learning requirement information, The second message is used to indicate whether the corresponding user equipment meets the requirement information of federated learning, and/or, the measurement information of the corresponding user equipment; or, the first message is used to request the measurement information of the user equipment, and the second message includes the measurement information ;
  • the federated learning requirement information includes federated learning duration requirement information, and/or, federated learning requirement information for user equipment stability.
  • the federated learning requirement information on user equipment stability includes requirement information on user equipment mobility, and/or requirement information on user equipment operating status.
  • the federated learning requirement information for user equipment stability includes one or more of the following items: mobility level requirement information, user equipment location requirement information, computing power change rate requirement information, or signal quality Rate of change demand information.
  • the measurement information includes one or more of the following items: mobility level, location of the user equipment, rate of change of computing power, or rate of change of signal quality.
  • a first indication message from the network device is received, where the first indication message is used to indicate to update the measurement information; and the measurement information is updated based on the first indication message.
  • the first indication message includes measurement requirement information, and if the user equipment does not meet the measurement requirement information, the measurement information is updated based on the first indication message.
  • a second indication message from the network device is received, and the second indication message is used to indicate whether continuous monitoring meets the requirement information of federated learning; a third indication message is sent to the network device, and the third indication message Information used to indicate that the user equipment does not meet the requirements of the federated learning.
  • the second indication message further includes requirement information of federated learning.
  • the present application provides an apparatus for user equipment selection, which includes: a sending unit, configured to send a first message to a first candidate user equipment group; a receiving unit, configured to receive a message from a second candidate user equipment group The second message, the first candidate user equipment group includes the second candidate user equipment group; wherein, the first message is used to indicate the requirement information of federated learning, and the second message is used to indicate whether the corresponding user equipment meets the federated learning requirements
  • the requirement information and/or, corresponds to the measurement information of the user equipment; or, the first message is used to request the measurement information of the user equipment, and the second message includes the measurement information; the requirement information of the federated learning includes the duration requirement of the federated learning Information, and/or information about the stability requirements of federated learning for user equipment;
  • the determining unit is configured to determine a target user equipment group participating in federated learning based on the second message, and each target user equipment in the target user equipment group satisfies the federated Learning needs information.
  • the federated learning requirement information on user equipment stability includes requirement information on user equipment mobility, and/or requirement information on user equipment operating status.
  • the federated learning requirement information on user equipment stability includes one or more of the following items: mobility level requirement information, user equipment location requirement information, computing power change rate requirement information or signal Mass change rate demand information.
  • the measurement information includes one or more of the following items: mobility level, location of the user equipment, rate of change of computing power, or rate of change of signal quality.
  • the sending unit is further configured to send a first indication message to the first candidate user equipment group, where the first indication message is used to indicate to update the measurement information.
  • the first indication message includes measurement requirement information.
  • the sending unit is further configured to send a second indication message to the target user equipment group, where the second indication message is used to indicate whether continuous monitoring meets the requirement information of federated learning;
  • the receiving unit is also configured to receiving a third indication message from the first target user equipment, where the third indication message is used to indicate that the first target user equipment does not meet the requirement information of federated learning, and the first target user equipment is one of the target user equipment group;
  • the determining unit is also used to remove the first target user equipment from the target user equipment group; or, reduce the priority of the first target user equipment in the target user equipment group.
  • the second indication message further includes requirement information of federated learning.
  • the receiving unit is further configured to receive a fourth indication message from other network devices, where the fourth indication message is used to indicate measurement information of at least one user equipment, and where the fourth indication message is used to determine the The first candidate user equipment group.
  • the sending unit is further configured to send a fifth indication message to other network devices, where the fifth indication message is used to indicate measurement information of at least one user equipment.
  • the determining unit when handover is performed for the second target user equipment, is further configured to release the second target user equipment, where the second target user equipment is one of the target user equipment group; from the The second target user equipment is removed from the target user equipment group.
  • the sending unit is configured to send a secondary node addition request message to a neighboring base station, where the secondary node addition request message is used to establish a communication connection between a third target user equipment and a neighboring base station; the third target The user equipment is one of the target user equipment group.
  • the sending unit is configured to send a sixth indication message to a neighboring base station, where the sixth indication message is used to indicate sending federated learning data of a third target user equipment; and receive the third target user equipment from a neighboring base station The federated learning data of the device.
  • the present application provides an apparatus for selecting user equipment, which includes: a receiving unit, configured to receive a first message from a network device; a sending unit, configured to send a second message to the network device; wherein, the first A message is used to indicate the requirement information of federated learning, and the second message is used to indicate whether the corresponding user equipment meets the required information of federated learning, and/or, the measurement information of the corresponding user equipment; or, the first message is used to request the user equipment
  • the second message includes measurement information; the federated learning requirement information includes federated learning duration requirement information, and/or, federated learning requirement information for user equipment stability.
  • the federated learning requirement information on user equipment stability includes requirement information on user equipment mobility, and/or requirement information on user equipment operating status.
  • the federated learning requirement information for user equipment stability includes one or more of the following items: mobility level requirement information, user equipment location requirement information, computing power change rate requirement information, or signal quality Rate of change demand information.
  • the measurement information includes one or more of the following items: mobility level, location of the user equipment, rate of change of computing power, or rate of change of signal quality.
  • the receiving unit is further configured to receive a first indication message from the network device, where the first indication message is used to indicate to update the measurement information; and update the measurement information based on the first indication message.
  • the apparatus further includes an updating unit, the first indication message includes measurement requirement information, and the updating unit is configured to update the measurement information based on the first indication message if the user equipment does not meet the measurement requirement information .
  • the receiving unit is further configured to receive a second indication message from the network device, the second indication message is used to indicate whether the continuous monitoring meets the requirement information of federated learning; the sending unit is also configured to send the network device The device sends a third indication message, where the third indication message is used to indicate that the user equipment does not meet the federated learning requirement information.
  • the second indication message further includes requirement information of federated learning.
  • the present application provides a chip, including a processor and a communication interface, the processor is configured to perform the following operations: send a first message to a first candidate user equipment group; receive a message from a second candidate user equipment group The second message of the first candidate user equipment group includes the second candidate user equipment group; wherein, the first message is used to indicate the requirement information of federated learning, and the second message is used to indicate whether the corresponding user equipment satisfies the federated learning and/or, corresponding to the measurement information of the user equipment; or, the first message is used to request the measurement information of the user equipment, and the second message includes the measurement information; the federated learning requirement information includes the duration of the federated learning Requirement information, and/or, federated learning requirement information for user equipment stability; based on the second message, determine a target user equipment group participating in federated learning, and each target user equipment in the target user equipment group meets the federated learning requirement information .
  • the present application provides a chip, including a processor and a communication interface, where the processor is configured to perform the following operations: receive a first message from a network device; send a second message to the network device; wherein, The first message is used to indicate the requirement information of federated learning, and the second message is used to indicate whether the corresponding user equipment meets the required information of federated learning, and/or, the measurement information of the corresponding user equipment; or, the first message is used to request the user
  • the second message includes measurement information; the federated learning requirement information includes federated learning duration requirement information, and/or, federated learning requirement information for user equipment stability.
  • the present application provides a module device, which includes a communication module, a power module, a storage module, and a chip, wherein: the power module is used to provide power for the module device; the The storage module is used to store data and instructions; the communication module is used for internal communication of the module device, or for the module device to communicate with external devices; the chip is used for: sending the first candidate user equipment group A message; receiving a second message from a second candidate user equipment group, the first candidate user equipment group includes a second candidate user equipment group; wherein, the first message is used to indicate the requirement information of federated learning, and the second message uses For indicating whether the corresponding user equipment meets the requirement information of the federated learning, and/or, the measurement information of the corresponding user equipment; or, the first message is used to request the measurement information of the user equipment, and the second message includes the measurement information;
  • the federated learning requirement information includes federated learning duration requirement information, and/or, federated learning requirement information for user equipment stability; based on the second message
  • the present application provides a module device, which includes a communication module, a power module, a storage module, and a chip, wherein: the power module is used to provide power for the module device; the The storage module is used to store data and instructions; the communication module is used for internal communication of the module device, or for the module device to communicate with external devices; the chip is used for: receiving the first message from the network device; Sending a second message to the network device; wherein, the first message is used to indicate the requirement information of federated learning, and the second message is used to indicate whether the corresponding user equipment meets the required information of federated learning, and/or, the measurement of the corresponding user equipment information; or, the first message is used to request the measurement information of the user equipment, and the second message includes the measurement information; the federated learning requirement information includes federated learning duration requirement information, and/or, the federated learning's influence on the stability of the user equipment demand information.
  • the embodiment of the present invention discloses a network device, the network device includes a memory and a processor, the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to invoke the program instructions , executing the method in the foregoing first aspect and any possible implementation manner thereof.
  • the embodiment of the present invention discloses a user equipment, the user equipment includes a memory and a processor, the memory is used to store a computer program, the computer program includes program instructions, and the processor is configured to call the program instructions , executing the method in the above second aspect and any possible implementation manner thereof.
  • the present application provides a computer-readable storage medium, where computer-readable instructions are stored in the computer-readable instruction, and when the computer-readable instructions are run on a communication device, the communication device executes the above-mentioned first The method in the second aspect and any possible implementation thereof, or execute the method in the second aspect and any possible implementation thereof.
  • the present application provides a computer program or a computer program product, including codes or instructions.
  • the codes or instructions are run on a computer, the computer executes the computer program described in the first aspect and any possible implementation thereof. method, or execute the method in the above second aspect and any possible implementation manner thereof.
  • FIG. 1 is a schematic diagram of a communication system provided by an embodiment of the present application.
  • Fig. 2 is a schematic flow diagram of a federated learning provided by the embodiment of the present application.
  • FIG. 3 is a schematic flowchart of a method for selecting a user equipment provided in an embodiment of the present application
  • FIG. 4 is a schematic flowchart of a method for selecting a user equipment provided in an embodiment of the present application
  • FIG. 5 is a schematic flowchart of a method for selecting a user equipment provided in an embodiment of the present application
  • FIG. 6 is a schematic flowchart of an optimization solution selected by a user equipment provided in an embodiment of the present application.
  • Fig. 7a is a schematic flowchart of an optimization solution selected by a user equipment provided in an embodiment of the present application.
  • Fig. 7b is a schematic flowchart of an optimization solution selected by a user equipment provided in an embodiment of the present application.
  • FIG. 8 is a schematic flowchart of an optimization solution selected by a user equipment provided in an embodiment of the present application.
  • FIG. 9 is a schematic structural diagram of an apparatus for selecting user equipment provided in an embodiment of the present application.
  • FIG. 10 is a schematic structural diagram of another apparatus for selecting user equipment provided by an embodiment of the present application.
  • FIG. 11 is a schematic structural diagram of a network device provided by an embodiment of the present application.
  • FIG. 12 is a schematic structural diagram of a user equipment provided in an embodiment of the present application.
  • Fig. 13 is a schematic structural diagram of a module device provided by an embodiment of the present application.
  • At least one (item) means one or more
  • “multiple” means two or more
  • “at least two (items)” means two or three and three
  • “and/or” is used to describe the association relationship of associated objects, which means that there can be three kinds of relationships, for example, “A and/or B” can mean: only A exists, only B exists, and A and B exist at the same time A case where A and B can be singular or plural.
  • the character “/” generally indicates that the contextual objects are an “or” relationship.
  • “At least one of the following” or similar expressions refer to any combination of these items, including any combination of single or plural items.
  • At least one item (piece) of a, b or c can mean: a, b, c, "a and b", “a and c", “b and c", or "a and b and c ", where a, b, c can be single or multiple.
  • GSM global system of mobile communication
  • CDMA code division multiple access
  • WCDMA broadband code division multiple access
  • general packet radio service general packet radio service, GPRS
  • long term evolution long term evolution
  • LTE long term evolution
  • LTE frequency division duplex frequency division duplex
  • FDD frequency division duplex
  • UMTS Universal Mobile Telecommunications System
  • WiMAX Worldwide Interoperability for Microwave Access
  • 5G Fifth Generation
  • 5G new radio
  • NR new radio
  • Fig. 1 is a schematic diagram of a communication system provided by an embodiment of the present application.
  • the solution in the present application is applicable to the communication system, and the communication system may include network equipment and user equipment.
  • Fig. 1 is an example in which a communication system includes one network device and two user equipments, which should not be regarded as a specific limitation to the present application.
  • UE User equipment
  • User equipment includes devices that provide voice and/or data connectivity to users.
  • user equipment is a device with wireless transceiver capabilities that can be deployed on land, including indoor or outdoor, handheld, wearable, or vehicle-mounted; it can also be deployed in On the water (such as ships, etc.); can also be deployed in the air (such as aircraft, balloons and satellites, etc.).
  • the terminal can be a mobile phone, a vehicle, a roadside unit, a tablet computer (Pad), a computer with a wireless transceiver function, a virtual reality (Virtual Reality, VR) terminal device, an augmented reality (Augmented Reality, AR) terminal Equipment, wireless terminals in industrial control, vehicle terminal equipment, wireless terminals in self driving, wireless terminals in remote medical, wireless terminals in smart grid , wireless terminals in transportation safety, wireless terminals in smart city, wireless terminals in smart home, wearable terminal devices, etc.
  • the embodiments of the present application do not limit the application scenarios.
  • a terminal can sometimes be called a terminal device, an access terminal device, a vehicle-mounted terminal, an industrial control terminal, a UE unit, a UE station, a mobile station, a mobile station, a remote station, a remote terminal device, a mobile device, a UE terminal device, a terminal device, Wireless communication device, UE agent or UE device, etc.
  • Terminals can also be fixed or mobile.
  • the device used to realize the function of the terminal device may be a terminal device, or a device capable of supporting the terminal device to realize the function, such as a chip system or a combined device or component that can realize the function of the terminal device. Can be installed in terminal equipment.
  • a network device is a node or device that connects a terminal device to a wireless network.
  • the network equipment mentioned in this application may include but not limited to: base station (base station), evolved base station (evolved NodeB, eNodeB), transmission reception point (transmission reception point, TRP), fifth generation (5th generation, The next generation base station (next generation NodeB, gNB) in the 5G) mobile communication system, the next generation base station in the sixth generation (6th generation, 6G) mobile communication system, the base station in the future mobile communication system or the access in the WiFi system node etc.
  • the base station mentioned in this application may also be a module or unit that completes some functions of the base station, for example, it may be a centralized unit (central unit, CU) or a distributed unit (distributed unit, DU).
  • the CU here completes the functions of the radio resource control protocol and the packet data convergence protocol (PDCP) of the base station, and also completes the function of the service data adaptation protocol (SDAP); the DU completes the functions of the base station
  • the functions of the radio link control layer and the medium access control (medium access control, MAC) layer can also complete the functions of part or all of the physical layer.
  • 3GPP 3rd generation partnership project
  • the network equipment may be a macro base station, a micro base station or an indoor station, or a relay node or a donor node.
  • the network device mentioned in this application may also be one or more CN devices in a core network (core network, CN).
  • the network device mentioned in this application may be one or more of the following CN devices: network slice selection function (network slice selection function, NSSF), network exposure function (network exposure function, NEF), network storage function ( network function repository function (NRF), policy control function (policy control function, PCF), unified data management (unified data management, UDM), application function (application function, AF), network control function (network control function, NCF), Specific network slice specific authentication and authorization function (network slice specific authentication and authorization function, NSSAAF), authentication server function (authentication server function, AUSF), access and mobility management function (access and mobility management function, AMF), session Management function (session management function, SMF), user plane function (user plane function, UPF), service communication proxy (service communication proxy, SCP), network slice admission control (Network Slice Admission Control Function, NSACF).
  • network slice selection function network slice selection function, NSSF
  • network exposure function network exposure function
  • the device for implementing the function of the network device may be the network device itself, or a device capable of supporting the network device to realize the function, such as a chip system or a combined device or component that can realize the function of the access network device,
  • the device can be installed in network equipment.
  • the embodiment of the present application does not limit the specific technology and specific device form adopted by the network device.
  • Federated learning is mainly a solution to the problem of data islands.
  • the goal of federated learning is to achieve joint modeling by multiple participants on the basis of ensuring data privacy, security and legality, so as to improve the effect of AI models.
  • Federated learning can be understood as a distributed machine learning technology.
  • the process of federated learning can be divided into one round of federated learning and multiple rounds of federated learning.
  • the following is an example of one round of federated learning. Please refer to Figure 2, which is a schematic diagram of a federated learning process.
  • a server corresponding to federated learning sends resource requests to multiple candidate user equipments.
  • the server corresponding to the federated learning receives the resource information reported by each candidate user equipment.
  • the server corresponding to the federated learning determines the user equipments participating in the federated learning according to the resource information sent by each candidate user equipment.
  • the server corresponding to the federated learning delivers the global model to each user device participating in the federated learning.
  • Each user equipment participating in federated learning performs a model update on the global model to obtain updated model parameters (including model weights and gradient values).
  • the server corresponding to the federated learning receives the updated model parameters sent by each user equipment participating in the federated learning.
  • the server corresponding to the federated learning may also send an Acknowledge character (ACK) signal to each user equipment participating in the federated learning.
  • ACK Acknowledge character
  • the server corresponding to the federated learning performs model aggregation according to each updated model parameter.
  • the participating user equipment when selecting a participating user equipment (or called a participant) for federated learning, the participating user equipment will be determined based on computing power information and signal quality.
  • the stability of the user equipment determined in this way cannot be guaranteed in the process of participating in federated learning, that is, there may be situations where the participating user equipment cannot fully participate in federated learning, which will lead to interruption of federated learning and affect the training time of federated learning. extension and the accuracy of the trained model.
  • the present application provides a user equipment selection method, device, chip and module equipment.
  • the user equipment selection method, device, chip, and module equipment provided in the embodiments of the present application are further described in detail below.
  • Fig. 3 is a schematic flowchart of a method for selecting a user equipment provided in an embodiment of the present application.
  • the method for selecting a user equipment includes the following steps S301-S303.
  • the execution body of the method shown in FIG. 3 may be a user equipment or a chip in the user equipment.
  • the subject of executing the method shown in FIG. 3 may be a network device or a chip in the network device.
  • FIG. 3 is illustrated by taking the user equipment and the network equipment as execution subjects of the method as an example.
  • the network device sends a first message to a first candidate user equipment group.
  • the network device selects the first candidate user equipment group, and sends the first message to each candidate user equipment in the first candidate user equipment group.
  • the first message will be described below first, and then the process for the network device to select the first candidate user equipment group will be described.
  • the content indicated by the first message can be divided into the following two situations. For details, please refer to FIG. 4 and FIG. 5. Both of FIG. 4 and FIG. 5
  • the network device sends a first message to multiple candidate user equipments (that is, the first candidate user equipment group), and the first message is used to indicate the requirement information of federated learning.
  • the federated learning requirement information includes federated learning duration requirement information, and/or federated learning requirement information for user equipment stability. It should be understood that the time requirement information of the federated learning is used to indicate the expected time that the federated learning will occupy the user equipment. The federated learning requirement information on user equipment stability is used to indicate the equipment stability conditions of the user equipment participating in the federated learning.
  • the federated learning requirement information on user equipment stability includes requirement information on user equipment mobility, and/or requirement information on user equipment operating status.
  • the demand information on the mobility of the user equipment can be understood as the user equipment is of a low mobility level, the user equipment is not at the cell edge, or the signal quality change rate of the user equipment is not greater than the first threshold, etc.
  • the requirement information of the equipment operating status can be understood as the change rate of the computing power of the user equipment is not greater than the second threshold and so on.
  • the specific value of the first threshold and the specific value of the second threshold can be adjusted according to specific application scenarios, which is not specifically limited in this application.
  • the federated learning requires information about the stability of user equipment, including but not limited to one or more of the following information: mobility level requirement information, user equipment location requirement information, computing power changes rate demand information or signal quality change rate demand information.
  • the federated learning requirement information mentioned in this application in addition to the above-mentioned requirement information (ie the duration requirement information of the federated learning and the requirement information of the federated learning for the stability of the user equipment), can also be Including other demand information, such as federated learning demand information for computing power, demand information for signal quality, etc.
  • the network device sends a first message to multiple candidate user equipments (that is, the first candidate user equipment group), and the first message is used to request measurement information of the user equipments.
  • the user equipment that has received the first message may report its own measurement information.
  • the measurement information includes internal measurement information of the user equipment (such as computing power change rate, temperature change rate, etc.), and air interface measurement information (such as mobility level, signal quality change rate, etc.).
  • the measurement information includes but is not limited to one or more of the following information: mobility level, location of the user equipment, rate of change of computing power, or rate of change of signal quality.
  • mobility level is used to indicate the mobility of the user equipment, for example, the mobility level of the user equipment is low mobility (that is, it can be understood that the user equipment does not move, moves infrequently, or only moves within the authorized area );
  • location information of the user equipment is used to indicate the relative location information of the user equipment in the cell where it is located, and the user equipment can perceive the relative location information of the cell where it is located according to information such as signal quality;
  • the rate of change of computing power is used to indicate the relative location information of the user equipment The amount of change in computing power within a period of time (for example, within 1 hour), the smaller the rate of change in computing power, the more stable the user equipment;
  • the amount of change in quality the smaller the rate of change in signal quality, the more stable the user equipment.
  • the process for the network device to select the first candidate user equipment group can be understood as that the network device obtains a plurality of user equipment information recorded on the network side, and selects a plurality of user equipment (ie, candidate user equipment) that may participate in federated learning according to the plurality of user equipment information. equipment), these user equipments that may participate in federated learning constitute the first candidate user equipment group.
  • a plurality of user equipment ie, candidate user equipment
  • these user equipments that may participate in federated learning constitute the first candidate user equipment group.
  • the aforementioned network side may be the network device itself, or other devices (which may be understood as devices having a communication connection with the network device).
  • the user equipment information includes but is not limited to capability information or computing power information.
  • the capability information is used to describe the capability of the user equipment, including but not limited to whether it supports federated learning;
  • the computing power information is used to describe the computing performance of the user equipment, which can describe the user equipment every second (or within a period of time) The number of operations that can be performed.
  • a network device receives a fourth indication message from another network device, where the fourth indication message is used to indicate measurement information of at least one user equipment, and where the fourth indication message is used to determine the first Candidate user equipment group.
  • the network device may also send a fifth indication message to other network devices, where the fifth indication message is used to indicate the measurement information of at least one user equipment, and the measurement information of the at least one user equipment It is used for other network devices to determine candidate user equipments for other federated learning (or the next round of federated learning of the federated learning).
  • the information of multiple user equipments recorded by the network side acquired by the network equipment may refer to Table 1.
  • Hash power information UE1 Support for Federated Learning 4.1 ⁇ 109 UE2 Support for Federated Learning 6 ⁇ 109 UE3 Support for Federated Learning 5.3 ⁇ 109
  • the network device selects a number of users who are likely to participate in federated learning User equipments (ie, the first candidate user equipment group): UE1, UE2, UE3, UE4, and UE5. Further, the network device sends the federated learning requirement information to the multiple user equipments (ie, UE1, UE2, UE3, UE4, and UE5) that may participate in the federated learning. Wherein, the network device may send the federated learning requirement information in a one-to-one unicast manner or in a one-to-many multicast manner, which is not specifically limited in this application.
  • the network device receives a second message from a second candidate user equipment group, where the first candidate user equipment group includes the second candidate user equipment group.
  • the user equipments in the first candidate user equipment group may choose to send the second message to the network device, or may not send the second message to the network device.
  • the number of user equipments in the second candidate user equipment group is less than or equal to the number of user equipments in the first candidate user equipment group, and the second candidate user equipment group is a subset of the first candidate user equipment group. It should be noted that, if the network device does not receive the second message sent by a candidate user equipment within a certain period of time, it may be considered that the candidate user equipment refuses to participate in federated learning.
  • the second message will be described below in conjunction with the two situations of the first message in S301.
  • Case 1 The first message is used to indicate the requirement information of federated learning.
  • the second message is used to indicate whether the corresponding user equipment meets the federated learning requirement information, and/or, the measurement information of the corresponding user equipment.
  • each candidate user equipment evaluates whether Participate in federated learning; further, as shown in S403 in FIG. 4, when the measurement information of the candidate user equipment meets the requirement information of the federated learning, the candidate user equipment may send a second message to the network device, and the second message is used for Information indicating that the requirements of federated learning are met (that is, federated learning is accepted), and/or, measurement information of the candidate user equipment.
  • the candidate user equipment may not send the second message to the network device.
  • the candidate user equipment may also send a second message to the network device, where the second message is used to indicate that the federated learning requirement information is not met (that is, federated learning is rejected), and/or the measurement information of the candidate user equipment.
  • Case 2 the first message is used to request measurement information of the user equipment.
  • the candidate user equipment sends a second message to the network device, where the second message includes measurement information of the corresponding user equipment.
  • each candidate user equipment after the candidate user equipment receives the first message from the network device, please refer to S502 in FIG. 5 for details, each candidate user equipment sends a second message to the network device, and the second message is used for The measurement information of each candidate user equipment is indicated, so that after receiving the second message, the network device can evaluate whether the candidate user equipment can participate in federated learning according to the second message.
  • the measurement information of a candidate user equipment is the key to evaluating whether the candidate user equipment satisfies the requirement information of federated learning.
  • the manner in which the candidate user equipment acquires its own measurement information will be described below. It should be noted that the following manners for the candidate user equipment to obtain its own measurement information (that is, the following manner 1 or manner 2) are applicable to the foregoing case 1 and case 2.
  • the candidate user equipment After the candidate user equipment receives the first message, the candidate user equipment obtains the recorded and stored measurement information of the user equipment from the storage space.
  • the candidate user equipment A performs measurement, and the measurement information obtained is: low mobility, not at the edge of a cell, a rate of change of computing power of 0.1, and a rate of change of signal quality of 0.2. Further, the candidate user equipment A records and stores the measurement information at the first moment. In this case, if the candidate user equipment A receives the federated learning requirement information at the second moment (a moment after the first moment), the candidate user equipment A obtains the measurement information at the first moment from the storage space , can quickly evaluate whether the federated learning requirement information is met according to the measurement information, and improve the evaluation efficiency.
  • the measurement information of the user equipment recorded and stored in the storage space is time-sensitive, it can be understood that the mobility level of the user equipment, the location information of the user equipment, the rate of change of computing power, or the rate of change of signal quality may change over time. change. In other words. The shorter the time interval between the first moment and the second moment, the closer the measurement information at the first moment is to the real measurement information corresponding to the user equipment at the second moment. If the time interval between the first moment and the second moment is large, it may lead to evaluation errors.
  • the network device before the network device receives the second message, the network device sends a first indication message to the candidate user equipment, and the first indication message is used to Indicates that the measurement information is updated.
  • the candidate user equipment receives a first indication message from the network device, where the first indication message is used to indicate to update the measurement information, and further, the candidate user equipment updates the measurement information according to the first indication message.
  • the measurement information is updated through such a method to ensure the accuracy of the measurement information.
  • the network device before the network device receives the second message, the network device sends a first indication message to the candidate user equipment, where the first indication message is used to indicate to update the measurement information, and the first indication message further includes Include measurement requirements information.
  • the measurement requirement information includes but not limited to the time requirement for the measurement information (for example, the measurement information within 5 minutes from the moment when the user equipment receives the first message).
  • the candidate user equipment receives a first indication message from the network device, the first indication message is used to indicate to update measurement information, the first indication message also includes measurement requirement information, and if the candidate user equipment does not meet the measurement requirement information, then The measurement information is updated based on the first indication message.
  • the measurement requirement information is "measurement information within 5 minutes from the moment of receiving the first message", if the measurement information stored in the storage space of the candidate user equipment A is recorded at the first moment, the candidate user equipment A at the If the federated learning requirement information is received at the second moment (a moment after the first moment), and the interval between the first moment and the second moment is 10 minutes, it can be considered that the candidate user equipment A does not meet the measurement requirement information. Further, the candidate user equipment A updates the measurement information based on the first indication message.
  • updating the measurement information based on the first indication message can be understood as if the first indication message includes the measurement configuration parameter, the candidate user equipment updates the measurement information according to the measurement configuration parameter in the first indication message; If the indication message does not include the measurement configuration parameter, the candidate user equipment updates the measurement information according to the preconfigured measurement configuration parameter.
  • the measurement configuration parameters include but are not limited to: one or more of measurement target, data format, measurement duration or measurement frequency.
  • Method 2 After the candidate user equipment receives the first message, the candidate user equipment performs real-time measurement on the user equipment according to the federated learning demand information, and obtains the measurement information of the candidate user equipment.
  • the candidate user equipment can perform real-time measurement according to the specific content of the federated-learned demand information to obtain the measurement information of the candidate user equipment.
  • the requirement information of federated learning includes requirement information on user equipment mobility and requirement information on user equipment operating status.
  • the candidate user equipment measures the mobility-related information of the user equipment (such as mobility level, signal quality change rate or location of the user equipment), and measures the operation state-related information of the user equipment (such as calculation rate of force change) to measure and obtain measurement information.
  • the network device determines a target user equipment group participating in federated learning based on the second message, and each target user equipment in the target user equipment group satisfies the requirement information of the federated learning.
  • the network device After receiving the response message from each candidate user equipment in the second candidate user equipment group, the network device determines a plurality of target user equipments from the candidate user equipments whose measurement information satisfies the requirement information of federated learning, and the multiple target user equipments are used to participate in the federation Learning means that the multiple target user equipments are target user equipment groups participating in the federated learning. Further, a protocol data unit (Protocol Data Unit, PDU) session is established between the network device and each target user equipment, and the network device participates in federated learning configuration parameters for each of the multiple target user equipments, and each target user equipment Start federated learning according to configuration parameters.
  • PDU Protocol Data Unit
  • the network device receives response messages from UE1-UE5, wherein the response messages sent by UE1-UE4 are response messages for participating in federated learning, and the response message sent by UE5 is a response message for rejecting federated learning.
  • the network device may determine UE1-UE4 as target user equipments participating in federated learning; or, the network device may select some user equipments from UE1-UE4 as target user equipments participating in federated learning.
  • the network device determines the target user equipment group participating in the federated learning based on the second message. It can be understood that the network device determines the target user equipment group from the candidate user equipments that the second message indicates meet the requirements of the federated learning. In other words, the user equipment evaluates (or decides) whether to participate in federated learning.
  • the network device determines the target user equipment participating in federated learning based on the second message Group, which can be understood as the network device evaluates whether the measurement information of each candidate user equipment meets the requirement information of federated learning according to the measurement information of each candidate user equipment, and determines multiple targets from the candidate user equipment whose measurement information meets the requirement information of federated learning User equipment, in other words, evaluates (or decides) whether to participate in federated learning by network equipment.
  • this application also provides the following optimization solutions: For details, please refer to FIG. 6 , FIG. 7a , FIG. 7b or FIG. 8 .
  • FIG. 6 is a schematic flowchart of an optimization scheme for user equipment selection provided by the present application.
  • each target user equipment can be continuously monitored, so that the network equipment understands each target The operating status of the user device.
  • the network device sends a second indication message to each target user equipment in the target user equipment group, where the second indication message is used to indicate whether continuous monitoring meets the federated learning requirement information.
  • the network device after the network device selects multiple target user equipments to participate in federated learning, the network device sends an indication message to each target user equipment, instructing each target user equipment to continuously monitor whether its own measurement information meets the requirements of federated learning during the federated learning process. demand information.
  • the second indication message when the target user equipment is not aware of (or understood to be unaware of) the requirement information of federated learning, the second indication message further includes the requirement information of federated learning.
  • the sending timing of the second indication message is not limited to after the target user equipment group is determined.
  • the network device when sending the first message to each candidate user equipment in the first candidate user equipment group (or before sending the first message, or before receiving the second message after sending the first message), the network device Sending a second indication message to each candidate user equipment, where the second indication message is used to indicate that each candidate user equipment needs to continuously monitor whether its own measurement information meets the federated learning requirement information after being determined as the target user equipment.
  • the network device receives a third indication message from the first target user equipment, where the third indication message is used to indicate that the first target user equipment does not meet the federated learning requirement, and the first target user equipment is a target user equipment in the target user equipment group one.
  • the first target user equipment When a certain target user equipment (that is, the first target user equipment) among the multiple target user equipments detects that its own measurement information does not meet the requirements of the federated learning, the first target user equipment sends a third indication message to the network device , the third indication message is used to indicate that the first target user equipment does not meet the federated learning requirement.
  • the network device removes the first target user equipment from the target user equipment group; or reduces the priority of the first target user equipment in the target user equipment group.
  • the network device may remove the first target user equipment from the target user equipment group; or, the network device may reduce the first target user equipment in the target user equipment Priority within the group. It should be noted that removing the first target user equipment from the target user equipment group can be understood as the first target user equipment no longer participates in the current round of federated learning when the federated learning is a multi-round federated learning federated learning. Lowering the priority of the first target user equipment in the target user equipment group can be understood as if the federated learning is multiple rounds of federated learning, the first target user equipment may participate in federated learning after the current round of federated learning.
  • the target user equipment group is UE1-UE4.
  • the network device sends the second indication message to UE1-UE4, so that each UE in UE1-UE4 continuously monitors whether its own measurement information satisfies the requirement information of the federated learning. If UE3 (that is, the first target user equipment) detects that it does not meet the federated learning requirement information, UE3 sends a third indication message to the network device, and the third indication message is used to indicate that UE3 does not meet the federated learning requirement information . Further, the network device removes the UE3 from the target user equipment group, that is, it can be understood that the target user equipment group of the federated learning includes UE1, UE2 and UE4.
  • the network device lowers the priority of the UE3 in the target user equipment group. It can be understood that if the federated learning is multiple rounds of federated learning, if the next round of federated learning does not consider the number of target user equipments after the UE3 If enough, the network device does not consider UE3 to participate in the next round of federated learning; if the number of target user equipment is not enough after the next round of federated learning does not consider UE3, UE3 will still participate in the next round of federated learning.
  • FIG. 7a and FIG. 7b are schematic flowcharts of the optimization scheme selected by the user equipment provided in this application.
  • FIG. 7a and FIG. 7b an implementation solution is provided for the scenario where the second target user equipment performing federated learning performs handover, which can increase the robustness of the federated learning system.
  • the source base station configures the target base station for the second target user equipment or before the second target user equipment receives the connection parameters of the target base station and is in the process of handover; it can also be understood as after the second target user equipment successfully switches the connection from the source base station to the target base station.
  • the second target user equipment is one of the target user equipment group, and when switching is performed for the second target user equipment, the network device may keep the second target user equipment to continue participating in the federated learning.
  • the network device sends the global model to the target base station.
  • the target base station delivers the global model to the second target user equipment.
  • the second target user equipment performs model update on the global model to obtain updated model parameters (including model weights and gradient values).
  • the second target user equipment sends the updated model parameters (including model weights and gradient values) to the target base station.
  • the target base station sends the updated model parameters (including model weights and gradient values) to the network device.
  • the network device can send the global model to the second target user equipment through the target base station, and the second target user equipment can also pass the target base station Send the updated model parameters to the network device.
  • the network device may also send the measurement information of the user equipment to the target base station, and the measurement information of the user equipment is used by the target base station to determine other federated learning (or the next round of federated learning after this round of federated learning) ) of the first candidate user equipment group.
  • the second target user equipment is one of the target user equipment group, and when handover is performed for the second target user equipment, the network device releases the second target user equipment.
  • Figure 7b where:
  • the network device sends a notification message to the second target user equipment, where the notification message is used to notify the second target user equipment that it no longer participates in federated learning.
  • the network device may send an explicit notification message to the second target user to notify the second target user that the equipment no longer participates in the federated learning.
  • the network device may also send an implicit notification message to the second target user (for example, no longer send federated learning-related messages to the second target user), so that the second target user equipment no longer participates in federated learning.
  • the network device removes the second target user equipment from the target user equipment group.
  • the network device can release the second target user equipment, the second target user equipment no longer participates in federated learning, and selects from the target user equipment group The second target user equipment is removed.
  • FIG. 8 is a schematic flowchart of an optimization solution for user equipment selection provided by the present application.
  • a solution for triggering the establishment of dual connections is provided for the third target user equipment performing federated learning, so as to increase the stability of the third target user equipment in federated learning.
  • the network device sends a secondary node addition request message to a neighboring base station, where the secondary node addition request message is used to establish a communication connection between a third target user equipment and a neighboring base station, where the third target user equipment is a target user equipment group one.
  • the network device is the serving base station of the third target user equipment, and when the network device perceives that the third target user equipment is at the cell edge, the network device may establish dual connectivity for the third target user equipment, that is, to the network device
  • the neighboring base station sends a secondary node addition request message.
  • the neighboring base station sends a secondary node addition request confirmation message to the network device.
  • the secondary node addition request acknowledgment message includes configuration information of the secondary cell corresponding to the neighboring base station (including wireless connection parameters of the secondary cell, etc.).
  • the network device sends an RRC reconfiguration message to the third target user equipment.
  • the RRC reconfiguration message includes configuration information of the secondary cell (including radio connection parameters of the secondary cell, etc.).
  • the third target user equipment sends an RRC reconfiguration complete message to the network device.
  • the network device sends a secondary node reconfiguration complete message to the neighboring base station.
  • the third target user equipment randomly accesses the neighboring base station, that is, it can be understood that the third target user equipment establishes a communication connection with the neighboring base station.
  • both the network device and the adjacent base station may receive federated learning data (such as updated model parameters and gradient values).
  • federated learning data such as updated model parameters and gradient values.
  • the neighboring base station sends the federated learning data of the third target user equipment to the network device by default.
  • the network device sends a sixth indication message to the neighboring base station, where the sixth indication message is used to indicate to send the federated learning data of the third target user equipment; further, the neighboring base station, according to the sixth indication message, Send the federated learning data of the third target user equipment to the network device.
  • FIG. 9 is a schematic structural diagram of an apparatus for selecting user equipment provided by an embodiment of the present invention.
  • the apparatus for selecting user equipment may be a network device or a device (such as a chip) having a network device function.
  • the user equipment selection means 900 may include:
  • a sending unit 901 configured to send a first message to a first candidate user equipment group; a receiving unit 902, configured to receive a second message from a second candidate user equipment group, where the first candidate user equipment group includes the second candidate user equipment group; wherein, the first message is used to indicate the requirement information of federated learning, and the second message is used to indicate whether the corresponding user equipment meets the required information of federated learning, and/or, the measurement information of the corresponding user equipment; or, the second A message is used to request measurement information of the user equipment, and the second message includes the measurement information; the federated learning requirement information includes federated learning duration requirement information, and/or, federated learning requirement information for user equipment stability;
  • the determining unit 903 is configured to determine a target user equipment group participating in federated learning based on the second message, and each target user equipment in the target user equipment group meets the requirement information of federated learning.
  • the federated learning requirement information on user equipment stability includes requirement information on user equipment mobility, and/or requirement information on user equipment operating status.
  • the federated learning requirement information on user equipment stability includes one or more of the following items: mobility level requirement information, user equipment location requirement information, computing power change rate requirement information or signal Mass change rate demand information.
  • the measurement information includes one or more of the following items: mobility level, location of the user equipment, rate of change of computing power, or rate of change of signal quality.
  • the sending unit 901 is further configured to send a first indication message to the first candidate user equipment group, where the first indication message is used to indicate to update the measurement information.
  • the first indication message includes measurement requirement information.
  • the sending unit 901 is further configured to send a second indication message to the target user equipment group, where the second indication message is used to indicate whether continuous monitoring meets the requirement information of federated learning;
  • the receiving unit 902 is also configured to It is used to receive a third indication message of the first target user equipment, where the third indication message is used to indicate that the first target user equipment does not meet the requirement information of federated learning, and the first target user equipment is one of the target user equipment group;
  • the determining unit 903 is further configured to remove the first target user equipment from the target user equipment group; or reduce the priority of the first target user equipment in the target user equipment group.
  • the second indication message further includes requirement information of federated learning.
  • the receiving unit 902 is further configured to receive a fourth indication message from other network devices, where the fourth indication message is used to indicate measurement information of at least one user equipment, and where the fourth indication message is used to determine The first candidate user equipment group.
  • the sending unit 901 is further configured to send a fifth indication message to other network devices, where the fifth indication message is used to indicate measurement information of at least one user equipment.
  • the determining unit 903 is further configured to release the second target user equipment, where the second target user equipment is one of the target user equipment group; from The second target user equipment is removed from the target user equipment group.
  • the sending unit 901 is configured to send a secondary node addition request message to a neighboring base station, where the secondary node addition request message is used to establish a communication connection between the third target user equipment and the neighboring base station; the third The target user equipment is one of the target user equipment group.
  • the sending unit 901 is configured to send a sixth indication message to a neighboring base station, where the sixth indication message is used to indicate sending federated learning data of a third target user equipment; and receive the third target user equipment from a neighboring base station. Federated learning data of user devices.
  • FIG. 10 is a schematic structural diagram of an apparatus for selecting user equipment according to an embodiment of the present invention.
  • the apparatus for selecting user equipment may be a user equipment or a device (such as a chip) having user equipment functions.
  • the user equipment selection apparatus 1000 may include:
  • the receiving unit 1001 is configured to receive a first message from a network device; the sending unit 1002 is configured to send a second message to the network device; wherein, the first message is used to indicate the requirement information of federated learning, and the second message is used to Information indicating whether the corresponding user equipment meets the requirements of federated learning, and/or, the measurement information of the corresponding user equipment; or, the first message is used to request the measurement information of the user equipment, and the second message includes the measurement information; the federated learning requirements
  • the information includes information about the duration requirement of the federated learning, and/or the requirement information of the federated learning for the stability of the user equipment.
  • the federated learning requirement information on user equipment stability includes requirement information on user equipment mobility, and/or requirement information on user equipment operating status.
  • the federated learning requirement information for user equipment stability includes one or more of the following items: mobility level requirement information, user equipment location requirement information, computing power change rate requirement information, or signal quality Rate of change demand information.
  • the measurement information includes one or more of the following items: mobility level, location of the user equipment, rate of change of computing power, or rate of change of signal quality.
  • the receiving unit 1001 is further configured to receive a first indication message from the network device, where the first indication message is used to indicate to update the measurement information; and update the measurement information based on the first indication message.
  • the apparatus further includes an updating unit 1003, where the first indication message includes measurement requirement information, and the updating unit 1003 is configured to update the user equipment based on the first indication message if the user equipment does not meet the measurement requirement information. measurement information.
  • the receiving unit 1001 is further configured to receive a second indication message from the network device, where the second indication message is used to indicate whether continuous monitoring meets the requirement information of federated learning; the sending unit 1002 is also configured to Sending a third indication message to the network device, where the third indication message is used to indicate that the user equipment does not meet the federated learning requirement information.
  • the second indication message further includes requirement information of federated learning.
  • FIG. 11 is a schematic structural diagram of a network device provided by an embodiment of the present invention.
  • the network device 1100 may include a memory 1101 and a processor 1102 .
  • a communication interface 1103 is also included.
  • the memory 1101, the processor 1102 and the communication interface 1103 are connected by one or more communication buses.
  • the communication interface 1103 is controlled by the processor 1102 for sending and receiving information.
  • the memory 1101 may include read-only memory and random-access memory, and provides instructions and data to the processor 1102 . A portion of memory 1101 may also include non-volatile random access memory.
  • the communication interface 1103 is used to receive or send data.
  • the processor 1102 can be a central processing unit (Central Processing Unit, CPU), and the processor 1102 can also be other general-purpose processors, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC ), off-the-shelf programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • the general-purpose processor may be a microprocessor, and optionally, the processor 1102 may also be any conventional processor. in:
  • the memory 1101 is used for storing program instructions.
  • the processor 1102 is configured to call the program instructions stored in the memory 1101 .
  • the processor 1102 calls the program instructions stored in the memory 1101 to make the network device 1100 perform the following operations: send a first message to a first candidate user equipment group; receive a second message from a second candidate user equipment group, the first candidate The user equipment group includes a second candidate user equipment group; wherein, the first message is used to indicate the federated learning requirement information, and the second message is used to indicate whether the corresponding user equipment meets the federated learning requirement information, and/or, the corresponding The measurement information of the user equipment; or, the first message is used to request the measurement information of the user equipment, and the second message includes the measurement information; the federated learning requirement information includes federated learning duration requirement information, and/or, federated learning Requirement information on user equipment stability; based on the second message, determine a target user equipment group participating in federated learning, and each target user equipment in the target user equipment group meets the requirement information of federated learning.
  • the federated learning requirement information on user equipment stability includes requirement information on user equipment mobility, and/or requirement information on user equipment operating status.
  • the federated learning requirement information on user equipment stability includes one or more of the following items: mobility level requirement information, user equipment location requirement information, computing power change rate requirement information or signal Mass change rate demand information.
  • the measurement information includes one or more of the following items: mobility level, location of the user equipment, rate of change of computing power, or rate of change of signal quality.
  • the processor 1102 is further configured to send a first indication message to the first candidate user equipment group, where the first indication message is used to indicate to update the measurement information.
  • the first indication message includes measurement requirement information.
  • the processor 1102 is further configured to send a second indication message to the target user equipment group, where the second indication message is used to indicate whether continuous monitoring meets the requirement information of federated learning; receiving the first target user equipment A third indication message of the device, where the third indication message is used to indicate that the first target user equipment does not meet the requirement information of federated learning, and the first target user equipment is one of the target user equipment group; remove from the target user equipment group Remove the first target user equipment; or, reduce the priority of the first target user equipment in the target user equipment group.
  • the second indication message further includes requirement information of federated learning.
  • the processor 1102 is further configured to receive a fourth indication message from other network devices, where the fourth indication message is used to indicate measurement information of at least one user equipment, and where the fourth indication message is used to determine The first candidate user equipment group.
  • the processor 1102 is further configured to send a fifth indication message to other network devices, where the fifth indication message is used to indicate measurement information of at least one user equipment.
  • the processor 1102 when handover is performed for the second target user equipment, is further configured to release the second target user equipment, where the second target user equipment is one of the target user equipment group; The second target user equipment is removed from the target user equipment group.
  • the processor 1102 is configured to send a secondary node addition request message to a neighboring base station, where the secondary node addition request message is used to establish a communication connection between the third target user equipment and the neighboring base station; the third The target user equipment is one of the target user equipment group.
  • the processor 1102 is configured to send a sixth indication message to a neighboring base station, where the sixth indication message is used to indicate sending federated learning data of a third target user equipment; and receive the third target user equipment from a neighboring base station. Federated learning data of user devices.
  • FIG. 12 is a schematic structural diagram of a user equipment provided by an embodiment of the present invention.
  • the user equipment 1200 may include a memory 1201 and a processor 1202 .
  • a communication interface 1203 is also included.
  • the memory 1201, the processor 1202 and the communication interface 1203 are connected by one or more communication buses.
  • the communication interface 1203 is controlled by the processor 1202 for sending and receiving information.
  • the memory 1201 may include read-only memory and random-access memory, and provides instructions and data to the processor 1202 .
  • a portion of memory 1201 may also include non-volatile random access memory.
  • the communication interface 1203 is used to receive or send data.
  • the processor 1202 can be a central processing unit (Central Processing Unit, CPU), and the processor 1202 can also be other general-purpose processors, a digital signal processor (Digital Signal Processor, DSP), an application specific integrated circuit (Application Specific Integrated Circuit, ASIC) ), off-the-shelf programmable gate array (Field-Programmable Gate Array, FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc.
  • the general processor may be a microprocessor, and optionally, the processor 1202 may also be any conventional processor. in:
  • the memory 1201 is used for storing program instructions.
  • the processor 1202 is configured to call the program instructions stored in the memory 1201 .
  • the processor 1202 invokes the program instructions stored in the memory 1201 to make the user equipment 1200 perform the following operations: receive a first message from the network device; send a second message to the network device; wherein the first message is used to indicate the federated learning Requirement information, the second message is used to indicate whether the corresponding user equipment meets the requirement information of federated learning, and/or, the measurement information of the corresponding user equipment; or, the first message is used to request the measurement information of the user equipment, and the second message Including measurement information; the federated learning requirement information includes federated learning duration requirement information, and/or, federated learning requirement information for user equipment stability.
  • the federated learning requirement information on user equipment stability includes requirement information on user equipment mobility, and/or requirement information on user equipment operating status.
  • the federated learning requirement information for user equipment stability includes one or more of the following items: mobility level requirement information, user equipment location requirement information, computing power change rate requirement information, or signal quality Rate of change demand information.
  • the measurement information includes one or more of the following items: mobility level, location of the user equipment, rate of change of computing power, or rate of change of signal quality.
  • the processor 1202 is further configured to receive a first indication message from the network device, where the first indication message is used to indicate to update the measurement information; and update the measurement information based on the first indication message.
  • the first indication message includes measurement requirement information
  • the processor 1202 is configured to update the measurement information based on the first indication message if the user equipment does not meet the measurement requirement information.
  • the processor 1202 is further configured to receive a second indication message from the network device, where the second indication message is used to indicate whether continuous monitoring meets the requirement information of federated learning; and send a third indication to the network device message, the third indication message is used to indicate that the user equipment does not meet the requirement information of the federated learning.
  • the second indication message further includes requirement information of federated learning.
  • the embodiment of the present application also provides a chip, which can execute the relevant steps of the network device in the foregoing method embodiments.
  • the chip includes a processor and a communication interface, the processor is configured to perform the following operations: send a first message to a first candidate user equipment group; receive a second message from a second candidate user equipment group, the first candidate The user equipment group includes a second candidate user equipment group; wherein, the first message is used to indicate the federated learning requirement information, and the second message is used to indicate whether the corresponding user equipment meets the federated learning requirement information, and/or, the corresponding The measurement information of the user equipment; or, the first message is used to request the measurement information of the user equipment, and the second message includes the measurement information; the federated learning requirement information includes federated learning duration requirement information, and/or, federated learning Requirement information on user equipment stability; based on the second message, determine a target user equipment group participating in federated learning, and each target user equipment in the target user equipment group meets the requirement information of feder
  • the federated learning requirement information on user equipment stability includes requirement information on user equipment mobility, and/or requirement information on user equipment operating status.
  • the federated learning requirement information on user equipment stability includes one or more of the following items: mobility level requirement information, user equipment location requirement information, computing power change rate requirement information or signal Mass change rate demand information.
  • the measurement information includes one or more of the following items: mobility level, location of the user equipment, rate of change of computing power, or rate of change of signal quality.
  • the processor is further configured to send a first indication message to the first candidate user equipment group, where the first indication message is used to indicate to update the measurement information.
  • the first indication message includes measurement requirement information.
  • the processor is further configured to send a second indication message to the target user equipment group, where the second indication message is used to indicate whether continuous monitoring meets the requirement information of federated learning; receiving the first target user equipment The third indication message, the third indication message is used to indicate that the first target user equipment does not meet the requirement information of federated learning, the first target user equipment is one of the target user equipment group; remove from the target user equipment group The first target user equipment; or, reducing the priority of the first target user equipment in the target user equipment group.
  • the second indication message further includes requirement information of federated learning.
  • the processor is further configured to receive a fourth indication message from other network devices, where the fourth indication message is used to indicate measurement information of at least one user equipment, and where the fourth indication message is used to determine the The first candidate user equipment group.
  • the processor is further configured to send a fifth indication message to other network devices, where the fifth indication message is used to indicate measurement information of at least one user equipment.
  • the processor when performing handover for the second target user equipment, is further configured to release the second target user equipment, where the second target user equipment is one of the target user equipment group; from the The second target user equipment is removed from the target user equipment group.
  • the processor is configured to send a secondary node addition request message to a neighboring base station, where the secondary node addition request message is used to establish a communication connection between a third target user equipment and a neighboring base station; the third target The user equipment is one of the target user equipment group.
  • the processor is configured to send a sixth indication message to a neighboring base station, where the sixth indication message is used to indicate sending federated learning data of a third target user equipment; and receive the third target user equipment from a neighboring base station The federated learning data of the device.
  • the embodiment of the present application also provides a chip, which can execute the relevant steps of the user equipment in the foregoing method embodiments.
  • the chip includes a processor and a communication interface, and the processor is configured to perform the following operations: receive a first message from a network device; send a second message to the network device; where the first message is used to indicate federated learning Requirement information, the second message is used to indicate whether the corresponding user equipment meets the requirement information of federated learning, and/or, the measurement information of the corresponding user equipment; or, the first message is used to request the measurement information of the user equipment, and the second message Including measurement information; the federated learning requirement information includes federated learning duration requirement information, and/or, federated learning requirement information for user equipment stability.
  • the federated learning requirement information on user equipment stability includes requirement information on user equipment mobility, and/or requirement information on user equipment operating status.
  • the federated learning requirement information for user equipment stability includes one or more of the following items: mobility level requirement information, user equipment location requirement information, computing power change rate requirement information, or signal quality Rate of change demand information.
  • the measurement information includes one or more of the following items: mobility level, location of the user equipment, rate of change of computing power, or rate of change of signal quality.
  • the processor is further configured to receive a first indication message from the network device, where the first indication message is used to indicate to update the measurement information; and update the measurement information based on the first indication message.
  • the first indication message includes measurement requirement information
  • the processor is configured to update the measurement information based on the first indication message if the user equipment does not meet the measurement requirement information.
  • the processor is further configured to receive a second indication message from the network device, where the second indication message is used to indicate whether continuous monitoring meets the requirement information of federated learning; and send a third indication message to the network device , the third indication message is used to indicate that the user equipment does not meet the federated learning requirement information.
  • the second indication message further includes requirement information of federated learning.
  • FIG. 13 is a schematic structural diagram of a module device provided by an embodiment of the present application.
  • the module device 1300 can perform the relevant steps of the user device in the foregoing method embodiments, and the module device 1300 includes: a communication module 1301 , a power module 1302 , a storage module 1303 and a chip 1304 .
  • the power supply module 1302 is used to provide electric energy for the module device;
  • the storage module 1303 is used to store data and instructions;
  • the communication module 1301 is used for internal communication of the module device, or for The module device communicates with external devices;
  • the chip 1304 is used for:
  • the requirement information of the corresponding user equipment, the second message is used to indicate whether the corresponding user equipment meets the requirement information of the federated learning, and/or, the measurement information of the corresponding user equipment; or, the first message is used to request the measurement information of the user equipment, the
  • the second message includes the measurement information;
  • the federated learning requirement information includes the federated learning duration requirement information, and/or, the federated learning requirement information for user equipment stability; based on the second message, determine the target user participating in the federated learning A device group, each target user device in the target user device group satisfies the requirement information of federated learning.
  • the federated learning requirement information on user equipment stability includes requirement information on user equipment mobility, and/or requirement information on user equipment operating status.
  • the federated learning requirement information on user equipment stability includes one or more of the following items: mobility level requirement information, user equipment location requirement information, computing power change rate requirement information or signal Mass change rate demand information.
  • the measurement information includes one or more of the following items: mobility level, location of the user equipment, rate of change of computing power, or rate of change of signal quality.
  • the chip 1304 is further configured to send a first indication message to the first candidate user equipment group, where the first indication message is used to indicate to update the measurement information.
  • the first indication message includes measurement requirement information.
  • the chip 1304 is further configured to send a second indication message to the target user equipment group, where the second indication message is used to indicate whether the continuous monitoring meets the requirement information of federated learning; receiving the first target user equipment The third indication message, the third indication message is used to indicate that the first target user equipment does not meet the requirement information of federated learning, the first target user equipment is one of the target user equipment group; remove from the target user equipment group The first target user equipment; or, reducing the priority of the first target user equipment in the target user equipment group.
  • the second indication message further includes requirement information of federated learning.
  • the chip 1304 is further configured to receive a fourth indication message from other network devices, where the fourth indication message is used to indicate measurement information of at least one user equipment, and where the fourth indication message is used to determine the The first candidate user equipment group.
  • the chip 1304 is further configured to send a fifth indication message to other network devices, where the fifth indication message is used to indicate measurement information of at least one user equipment.
  • the chip 1304 when performing handover for the second target user equipment, is also used to release the second target user equipment, and the second target user equipment is one of the target user equipment group; The second target user equipment is removed from the target user equipment group.
  • the chip 1304 is configured to send a secondary node addition request message to a neighboring base station, where the secondary node addition request message is used to establish a communication connection between a third target user equipment and a neighboring base station; the third target The user equipment is one of the target user equipment group.
  • the chip 1304 is configured to send a sixth indication message to a neighboring base station, where the sixth indication message is used to instruct to send federated learning data of a third target user equipment; receive the third target user equipment from a neighboring base station The federated learning data of the device.
  • the chip 1304 in the module device shown in FIG. 13 can also be used to: receive a first message from a network device; send a second message to the network device; where the first message is used to indicate federation Learning requirement information, the second message is used to indicate whether the corresponding user equipment meets the federated learning requirement information, and/or, the measurement information of the corresponding user equipment; or, the first message is used to request the measurement information of the user equipment, the second
  • the second message includes measurement information;
  • the federated learning requirement information includes federated learning duration requirement information, and/or, federated learning requirement information for user equipment stability.
  • the federated learning requirement information on user equipment stability includes requirement information on user equipment mobility, and/or requirement information on user equipment operating status.
  • the federated learning requirement information for user equipment stability includes one or more of the following items: mobility level requirement information, user equipment location requirement information, computing power change rate requirement information, or signal quality Rate of change demand information.
  • the measurement information includes one or more of the following items: mobility level, location of the user equipment, rate of change of computing power, or rate of change of signal quality.
  • the chip 1304 is further configured to receive a first indication message from the network device, where the first indication message is used to indicate to update the measurement information; and update the measurement information based on the first indication message.
  • the first indication message includes measurement requirement information
  • the chip 1304 is configured to update the measurement information based on the first indication message if the user equipment does not meet the measurement requirement information.
  • the chip 1304 is also configured to receive a second indication message from the network device, where the second indication message is used to indicate whether continuous monitoring meets the requirement information of federated learning; and send a third indication message to the network device , the third indication message is used to indicate that the user equipment does not meet the federated learning requirement information.
  • the second indication message further includes requirement information of federated learning.
  • the embodiment of the present application also provides a computer-readable storage medium, where instructions are stored in the computer-readable storage medium, and when the instruction is run on a processor, the method flow of the above-mentioned method embodiment is realized.
  • the embodiment of the present application further provides a computer program product.
  • the computer program product is run on a processor, the method flow of the above method embodiment is realized.
  • each module/unit contained in the product may be a software module/unit, or a hardware module/unit, or may be partly a software module/unit and partly a hardware module/unit.
  • each module/unit contained in the product may be realized by hardware such as a circuit, or at least part of the modules/units may be realized by a software program, and the software program runs
  • the processor is integrated inside the chip, and the remaining (if any) modules/units can be realized by hardware such as circuits; Realized by means of hardware such as circuits, different modules/units can be located in the same part of the chip module (such as chips, circuit modules, etc.) or in different components, or at least part of the modules/units can be implemented in the form of software programs, the software program Running on the integrated processor inside the chip module, the remaining (if any) modules/units can be realized by hardware such as circuits; It is implemented by means of hardware such as circuits, and different modules

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Theoretical Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Software Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • General Physics & Mathematics (AREA)
  • Evolutionary Computation (AREA)
  • Artificial Intelligence (AREA)
  • Mathematical Physics (AREA)
  • Computing Systems (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Health & Medical Sciences (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computational Linguistics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • Medical Informatics (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请公开了一种用户设备选择方法、装置、芯片及模组设备，该方法包括：向第一候选用户设备组发送第一消息；接收来自第二候选用户设备组的第二消息；其中，第一候选用户设备组包括第二候选用户设备组；第一消息指示联邦学习需求信息，第二消息指示用户设备是否满足需求信息，和/或，对应用户设备的测量信息；或者，第一消息用于请求测量信息，第二消息包括测量信息；联邦学习的需求信息包括联邦学习的时长需求信息，和/或，联邦学习对用户设备稳定性的需求信息；基于第二消息确定参与联邦学习的目标用户设备组，该目标用户设备组中每个目标用户设备满足该需求信息。通过该方法可以提升用户设备参与联邦学习的稳定性，降低联邦学习的中断概率。

Description

一种用户设备选择方法、装置、芯片及模组设备 技术领域
本发明涉及通信领域，尤其涉及一种用户设备选择方法、装置、芯片及模组设备。
背景技术
联邦学习可以在保护用户数据隐私的基础上，通过多用户联合训练，获得满足使用方性能需求的人工智能(Artificial Intelligence，AI)模型，因此在通信网络中受到关注。
在选择联邦学习的参与方时，通常仅考虑参与方的算力信息和信号质量。但也存在参与方不能完整参与联邦学习的情况，进而导致联邦学习中断，影响联邦学习的训练时延和训练模型的精度。如何确定合适的参与方以降低联邦学习的中断概率，是一个亟待解决的问题。
发明内容
本申请提供一种用户设备选择方法、装置、芯片及模组设备，有利于提升用户设备参与联邦学习的稳定性，降低联邦学习的中断概率。
第一方面，本申请提供一种用户设备选择方法，该方法包括：向第一候选用户设备组发送第一消息；接收来自第二候选用户设备组的第二消息，该第一候选用户设备组包括第二候选用户设备组；其中，第一消息用于指示联邦学习的需求信息，该第二消息用于指示对应用户设备是否满足所述联邦学习的需求信息，和/或，对应用户设备的测量信息；或者，第一消息用于请求用户设备的测量信息，该第二消息包括所述测量信息；该联邦学习的需求信息包括联邦学习的时长需求信息，和/或，联邦学习对用户设备稳定性的需求信息；基于第二消息，确定参与联邦学习的目标用户设备组，该目标用户设备组中每个目标用户设备满足联邦学习的需求信息。
基于第一方面所描述的方法，网络设备可以根据联邦学习的时长需求或对用户设备稳定性的需求，选择参与联邦学习的用户设备，从而提升参与联邦学习的用户设备的稳定性，尽量保证该用户设备能完整的参与联邦学习，从而降低联邦学习的中断概率。
在一个可能的实施方式中，该联邦学习对用户设备稳定性的需求信息包括对用户设备移动性的需求信息，和/或，对用户设备运行状态的需求信息。通过实施该可能的实施方式，可以提升参与联邦学习的用户设备的移动性方面或运行状态方面的稳定性。
在一个可能的实施方式中，该联邦学习对用户设备稳定性的需求信息包括以下项的一种或多种：移动性等级需求信息、用户设备的位置需求信息、算力变化率需求信息或信号质量变化率需求信息。
在一个可能的实施方式中，该测量信息包括以下项的一种或多种：移动性等级、用户设备的位置、算力变化率或信号质量变化率。
在一个可能的实施方式中，向第一候选用户设备组发送第一指示消息，该第一指示消息用于指示更新测量信息。
在一个可能的实施方式中，该第一指示消息包括测量需求信息。
在一个可能的实施方式中，向该目标用户设备组发送第二指示消息，该第二指示消息用于指示持续监测是否满足联邦学习的需求信息；接收第一目标用户设备的第三指示消息，该第三指示消息用于指示第一目标用户设备不满足联邦学习的需求信息，该第一目标用户设备为目标用户设备组中的一个；从目标用户设备组中移除该第一目标用户设备；或者，降低该第一目标用户设备在目标用户设备组中的优先级。
在一个可能的实施方式中，该第二指示消息还包括联邦学习的需求信息。
在一个可能的实施方式中，接收来自其他网络设备的第四指示消息，该第四指示消息用于指示至少一个用户设备的测量信息，该第四指示消息用于确定所述第一候选用户设备组。
在一个可能的实施方式中，向其他网络设备发送第五指示消息，该第五指示消息用于指示至少一个用户设备的测量信息。
在一个可能的实施方式中，在为第二目标用户设备执行切换时，释放该第二目标用户设备，该第二目标用户设备为目标用户设备组中的一个；从该目标用户设备组中移除该第二目标用户设备。
在一个可能的实施方式中，向邻基站发送辅节点添加请求消息，该辅节点添加请求消息用于建立第三目标用户设备与邻基站之间的通信连接；该第三目标用户设备为目标用户设备组中的一个。通过实施该可能的实施方式，参与联邦学习的用户设备处于小区边缘时，网络设备可以为其建立双连接，以保证该用户设备能完整的参与联邦学习，从而降低联邦学习的中断概率。
在一个可能的实施方式中，向邻基站发送第六指示消息，该第六指示消息用于指示发送第三目标用户设备的联邦学习数据；接收来自邻基站的第三目标用户设备的联邦学习数据。
第二方面，本申请提供一种用户设备选择方法，该方法包括：接收来自网络设备的第一消息；向该网络设备发送第二消息；其中，第一消息用于指示联邦学习的需求信息，该第二消息用于指示对应用户设备是否满足联邦学习的需求信息，和/或，对应用户设备的测量信息；或者，第一消息用于请求用户设备的测量信息，该第二消息包括测量信息；该联邦学习的需求信息包括联邦学习的时长需求信息，和/或，联邦学习对用户设备稳定性的需求信息。
基于第一方面所描述的方法，其有益效果可参见前述第一方面所描述的有益效果，重复之处在此不再进行过多赘述。
在一个可能的实施方式中，联邦学习对用户设备稳定性的需求信息包括对用户设备移动性的需求信息，和/或，对用户设备运行状态的需求信息。
在一个可能的实施方式中，联邦学习对用户设备稳定性的需求信息包括以下项的一种或多种：移动性等级需求信息、用户设备的位置需求信息、算力变化率需求信息或信号质量变化率需求信息。
在一个可能的实施方式中，测量信息包括以下项的一种或多种：移动性等级、用户设备的位置、算力变化率或信号质量变化率。
在一个可能的实施方式中，接收来自网络设备的第一指示消息，该第一指示消息用于 指示更新测量信息；基于该第一指示消息更新测量信息。
在一个可能的实施方式中，该第一指示消息包括测量需求信息，若用户设备不满足该测量需求信息，则基于第一指示消息更新测量信息。
在一个可能的实施方式中，接收来自网络设备的第二指示消息，该第二指示消息用于指示持续监测是否满足联邦学习的需求信息；向网络设备发送第三指示消息，该第三指示消息用于指示用户设备不满足所述联邦学习的需求信息。
在一个可能的实施方式中，该第二指示消息还包括联邦学习的需求信息。
第三方面，本申请提供了一种用户设备选择装置，该装置包括：发送单元，用于向第一候选用户设备组发送第一消息；接收单元，用于接收来自第二候选用户设备组的第二消息，该第一候选用户设备组包括第二候选用户设备组；其中，第一消息用于指示联邦学习的需求信息，该第二消息用于指示对应用户设备是否满足所述联邦学习的需求信息，和/或，对应用户设备的测量信息；或者，第一消息用于请求用户设备的测量信息，该第二消息包括所述测量信息；该联邦学习的需求信息包括联邦学习的时长需求信息，和/或，联邦学习对用户设备稳定性的需求信息；确定单元，用于基于第二消息，确定参与联邦学习的目标用户设备组，该目标用户设备组中每个目标用户设备满足联邦学习的需求信息。
在一个可能的实施方式中，该联邦学习对用户设备稳定性的需求信息包括对用户设备移动性的需求信息，和/或，对用户设备运行状态的需求信息。
在一个可能的实施方式中，该联邦学习对用户设备稳定性的需求信息包括以下项的一种或多种：移动性等级需求信息、用户设备的位置需求信息、算力变化率需求信息或信号质量变化率需求信息。
在一个可能的实施方式中，该测量信息包括以下项的一种或多种：移动性等级、用户设备的位置、算力变化率或信号质量变化率。
在一个可能的实施方式中，发送单元，还用于向第一候选用户设备组发送第一指示消息，该第一指示消息用于指示更新该测量信息。
在一个可能的实施方式中，该第一指示消息包括测量需求信息。
在一个可能的实施方式中，发送单元，还用于向该目标用户设备组发送第二指示消息，该第二指示消息用于指示持续监测是否满足联邦学习的需求信息；接收单元，还用于接收第一目标用户设备的第三指示消息，该第三指示消息用于指示第一目标用户设备不满足联邦学习的需求信息，该第一目标用户设备为目标用户设备组中的一个；确定单元，还用于从目标用户设备组中移除该第一目标用户设备；或者，降低该第一目标用户设备在目标用户设备组中的优先级。
在一个可能的实施方式中，该第二指示消息还包括联邦学习的需求信息。
在一个可能的实施方式中，接收单元，还用于接收来自其他网络设备的第四指示消息，该第四指示消息用于指示至少一个用户设备的测量信息，该第四指示消息用于确定所述第一候选用户设备组。
在一个可能的实施方式中，发送单元，还用于向其他网络设备发送第五指示消息，该第五指示消息用于指示至少一个用户设备的测量信息。
在一个可能的实施方式中，在为第二目标用户设备执行切换时，确定单元，还用于释 放该第二目标用户设备，该第二目标用户设备为目标用户设备组中的一个；从该目标用户设备组中移除该第二目标用户设备。
在一个可能的实施方式中，发送单元，用于向邻基站发送辅节点添加请求消息，该辅节点添加请求消息用于建立第三目标用户设备与邻基站之间的通信连接；该第三目标用户设备为目标用户设备组中的一个。
在一个可能的实施方式中，发送单元，用于向邻基站发送第六指示消息，该第六指示消息用于指示发送第三目标用户设备的联邦学习数据；接收来自邻基站的第三目标用户设备的联邦学习数据。
第四方面，本申请提供了一种用户设备选择装置，该装置包括：接收单元，用于接收来自网络设备的第一消息；发送单元，用于向该网络设备发送第二消息；其中，第一消息用于指示联邦学习的需求信息，该第二消息用于指示对应用户设备是否满足联邦学习的需求信息，和/或，对应用户设备的测量信息；或者，第一消息用于请求用户设备的测量信息，该第二消息包括测量信息；该联邦学习的需求信息包括联邦学习的时长需求信息，和/或，联邦学习对用户设备稳定性的需求信息。
基于第一方面所描述的方法，其有益效果可参见前述第一方面所描述的有益效果，重复之处在此不再进行过多赘述。
在一个可能的实施方式中，联邦学习对用户设备稳定性的需求信息包括对用户设备移动性的需求信息，和/或，对用户设备运行状态的需求信息。
在一个可能的实施方式中，联邦学习对用户设备稳定性的需求信息包括以下项的一种或多种：移动性等级需求信息、用户设备的位置需求信息、算力变化率需求信息或信号质量变化率需求信息。
在一个可能的实施方式中，测量信息包括以下项的一种或多种：移动性等级、用户设备的位置、算力变化率或信号质量变化率。
在一个可能的实施方式中，接收单元，还用于接收来自网络设备的第一指示消息，该第一指示消息用于指示更新测量信息；基于该第一指示消息更新测量信息。
在一个可能的实施方式中，该装置还包括更新单元，该第一指示消息包括测量需求信息，该更新单元，用于若用户设备不满足该测量需求信息，则基于第一指示消息更新测量信息。
在一个可能的实施方式中，接收单元，还用于接收来自网络设备的第二指示消息，该第二指示消息用于指示持续监测是否满足联邦学习的需求信息；发送单元，还用于向网络设备发送第三指示消息，该第三指示消息用于指示用户设备不满足所述联邦学习的需求信息。
在一个可能的实施方式中，该第二指示消息还包括联邦学习的需求信息。
第五方面，本申请提供了一种芯片，包括处理器和通信接口，该处理器被配置用于执行如下操作：向第一候选用户设备组发送第一消息；接收来自第二候选用户设备组的第二消息，该第一候选用户设备组包括第二候选用户设备组；其中，第一消息用于指示联邦学习的需求信息，该第二消息用于指示对应用户设备是否满足所述联邦学习的需求信息，和/或，对应用户设备的测量信息；或者，第一消息用于请求用户设备的测量信息，该第二消 息包括所述测量信息；该联邦学习的需求信息包括联邦学习的时长需求信息，和/或，联邦学习对用户设备稳定性的需求信息；基于第二消息，确定参与联邦学习的目标用户设备组，该目标用户设备组中每个目标用户设备满足联邦学习的需求信息。
第六方面，本申请提供了一种芯片，包括处理器和通信接口，该处理器被配置用于执行如下操作：接收来自网络设备的第一消息；向该网络设备发送第二消息；其中，第一消息用于指示联邦学习的需求信息，该第二消息用于指示对应用户设备是否满足联邦学习的需求信息，和/或，对应用户设备的测量信息；或者，第一消息用于请求用户设备的测量信息，该第二消息包括测量信息；该联邦学习的需求信息包括联邦学习的时长需求信息，和/或，联邦学习对用户设备稳定性的需求信息。
第七方面，本申请提供了一种模组设备，该模组设备包括通信模组、电源模组、存储模组以及芯片，其中：该电源模组用于为该模组设备提供电能；该存储模组用于存储数据和指令；该通信模组用于进行模组设备内部通信，或者用于该模组设备与外部设备进行通信；该芯片用于：向第一候选用户设备组发送第一消息；接收来自第二候选用户设备组的第二消息，该第一候选用户设备组包括第二候选用户设备组；其中，第一消息用于指示联邦学习的需求信息，该第二消息用于指示对应用户设备是否满足所述联邦学习的需求信息，和/或，对应用户设备的测量信息；或者，第一消息用于请求用户设备的测量信息，该第二消息包括所述测量信息；该联邦学习的需求信息包括联邦学习的时长需求信息，和/或，联邦学习对用户设备稳定性的需求信息；基于第二消息，确定参与联邦学习的目标用户设备组，该目标用户设备组中每个目标用户设备满足联邦学习的需求信息。
第八方面，本申请提供了一种模组设备，该模组设备包括通信模组、电源模组、存储模组以及芯片，其中：该电源模组用于为该模组设备提供电能；该存储模组用于存储数据和指令；该通信模组用于进行模组设备内部通信，或者用于该模组设备与外部设备进行通信；该芯片用于：接收来自网络设备的第一消息；向该网络设备发送第二消息；其中，第一消息用于指示联邦学习的需求信息，该第二消息用于指示对应用户设备是否满足联邦学习的需求信息，和/或，对应用户设备的测量信息；或者，第一消息用于请求用户设备的测量信息，该第二消息包括测量信息；该联邦学习的需求信息包括联邦学习的时长需求信息，和/或，联邦学习对用户设备稳定性的需求信息。
第九方面，本发明实施例公开了一种网络设备，该网络设备包括存储器和处理器，该存储器用于存储计算机程序，该计算机程序包括程序指令，该处理器被配置用于调用该程序指令，执行上述第一方面及其任一种可能的实现方式中的方法。
第十方面，本发明实施例公开了一种用户设备，该用户设备包括存储器和处理器，该存储器用于存储计算机程序，该计算机程序包括程序指令，该处理器被配置用于调用该程序指令，执行上述第二方面及其任一种可能的实现方式中的方法。
第十一方面，本申请提供了一种计算机可读存储介质，该计算机存储介质中存储有计算机可读指令，当该计算机可读指令在通信装置上运行时，使得该通信装置执行上述第一方面及其任一种可能的实现方式中的方法，或者，执行上述第二方面及其任一种可能的实现方式中的方法。
第十二方面，本申请提供一种计算机程序或计算机程序产品，包括代码或指令，当代 码或指令在计算机上运行时，使得计算机执行如第一方面及其任一种可能的实现方式中的方法，或者，执行上述第二方面及其任一种可能的实现方式中的方法。
附图说明
为了更清楚地说明本申请实施例中的技术方案，下面将对实施例描述中所需要使用的附图作简单地介绍，显而易见地，下面描述中的附图仅仅是本发明的一些实施例，对于本领域普通技术人员来讲，在不付出创造性劳动的前提下，还可以根据这些附图获得其他的附图。
图1是本申请实施例提供的一种通信系统的示意图；
图2是本申请实施例提供的一种联邦学习的流程示意图；
图3是本申请实施例提供的一种用户设备选择方法的流程示意图；
图4是本申请实施例提供的一种用户设备选择方法的流程示意图；
图5是本申请实施例提供的一种用户设备选择方法的流程示意图；
图6是本申请实施例提供的一种用户设备选择的优化方案的流程示意图；
图7a是本申请实施例提供的一种用户设备选择的优化方案的流程示意图；
图7b是本申请实施例提供的一种用户设备选择的优化方案的流程示意图；
图8是本申请实施例提供的一种用户设备选择的优化方案的流程示意图；
图9是本申请实施例提供的一种用户设备选择装置的结构示意图；
图10是本申请实施例提供的另一种用户设备选择装置的结构示意图；
图11是本申请实施例提供的一种网络设备的结构示意图；
图12是本申请实施例提供的一种用户设备的结构示意图；
图13是本申请实施例提供的一种模组设备的结构示意图。
具体实施方式
下面结合附图对本申请具体实施例作进一步的详细描述。
本申请的说明书、权利要求书及附图中的术语“第一”和“第二”等是用于区别不同对象，而不是用于描述特定顺序。此外，术语“包括”和“具有”以及它们任何变形，意图在于覆盖不排他的包含。例如包含了一系列步骤或单元的过程、方法、系统、产品或设备没有限定于已列出的步骤或单元，而是可选地还包括没有列出的步骤或单元，或可选地还包括对于这些过程、方法、产品或设备固有的其它步骤或单元。
在本文中提及“实施例”意味着，结合实施例描述的特定特征、结构或特性可以包含在本申请的至少一个实施例中。在说明书中的各个位置出现该短语并不一定均是指相同的实施例，也不是与其它实施例互斥的独立的或备选的实施例。本领域技术人员显式地和隐式地理解的是，本文所描述的实施例可以与其它实施例相结合。
在本申请中，“至少一个(项)”是指一个或者多个，“多个”是指两个或两个以上，“至少两个(项)”是指两个或三个及三个以上，“和/或”，用于描述关联对象的关联关系，表示可以存在三种关系，例如，“A和/或B”可以表示：只存在A，只存在B以及同时存在A和B三种情况，其中A，B可以是单数或者复数。字符“/”一般表示前后关联对象是一种 “或”的关系。“以下至少一项(个)”或其类似表达，是指这些项中的任意组合，包括单项(个)或复数项(个)的任意组合。例如，a，b或c中的至少一项(个)，可以表示：a，b，c，“a和b”，“a和c”，“b和c”，或“a和b和c”，其中a，b，c可以是单个，也可以是多个。
为了更好地理解本申请实施例，下面首先对本申请实施例涉及的通信系统进行介绍：
本申请实施例的技术方案可以应用于各种通信系统，例如：全球移动通讯(global system of mobile communication，GSM)系统、码分多址(code division multiple access，CDMA)系统、宽带码分多址(wideband code division multiple access，WCDMA)系统、通用分组无线业务(general packet radio service，GPRS)、长期演进(long term evolution，LTE)系统、LTE频分双工(frequency division duplex，FDD)系统、LTE时分双工(time division duplex，TDD)、通用移动通信系统(universal mobile telecommunication system，UMTS)、全球互联微波接入(worldwide interoperability for microwave access，WiMAX)通信系统、第五代(5th generation，5G)系统或新无线(new radio，NR)以及未来的通信系统等。
图1是本申请实施例提供的一种通信系统的示意图，本申请中的方案可适用于该通信系统，该通信系统可以包括网络设备和用户设备。需要说明的是，图1是以通信系统中包括1个网络设备和2个用户设备为例，不应该视为对本申请的一个具体限定。
一、用户设备(user equipment，UE)
用户设备包括向用户提供语音和/或数据连通性的设备，例如用户设备是一种具有无线收发功能的设备，可以部署在陆地上，包括室内或室外、手持、穿戴或车载；也可以部署在水面上(如轮船等)；还可以部署在空中(例如飞机、气球和卫星上等)。所述终端可以是手机(mobile phone)、车辆、路侧单元、平板电脑(Pad)、带无线收发功能的电脑、虚拟现实(Virtual Reality，VR)终端设备、增强现实(Augmented Reality，AR)终端设备、工业控制(industrial control)中的无线终端、车载终端设备、无人驾驶(self driving)中的无线终端、远程医疗(remote medical)中的无线终端、智能电网(smart grid)中的无线终端、运输安全(transportation safety)中的无线终端、智慧城市(smart city)中的无线终端、智慧家庭(smart home)中的无线终端、可穿戴终端设备等等。本申请的实施例对应用场景不做限定。终端有时也可以称为终端设备、接入终端设备、车载终端、工业控制终端、UE单元、UE站、移动站、移动台、远方站、远程终端设备、移动设备、UE终端设备、终端设备、无线通信设备、UE代理或UE装置等。终端也可以是固定的或者移动的。本申请实施例中，用于实现终端设备的功能的装置可以是终端设备，也可以是能够支持终端设备实现该功能的装置，例如芯片系统或可实现终端设备功能的组合器件、部件，该装置可以被安装在终端设备中。
二、网络设备
网络设备即为将终端设备接入到无线网络的节点或设备。本申请中所提及的网络设备可以是包括但不限于：基站(base station)、演进型基站(evolved NodeB，eNodeB)、发送接收点(transmission reception point，TRP)、第五代(5th generation，5G)移动通信系统中的下一代基站(next generation NodeB，gNB)、第六代(6th generation，6G)移动通信系统中的下一代基站、未来移动通信系统中的基站或WiFi系统中的接入节点等。本申请所提及 的基站也可以是完成基站部分功能的模块或单元，例如，可以是集中式单元(central unit，CU)，也可以是分布式单元(distributed unit，DU)。这里的CU完成基站的无线资源控制协议和分组数据汇聚层协议(packet data convergence protocol，PDCP)的功能，还可以完成业务数据适配协议(service data adaptation protocol，SDAP)的功能；DU完成基站的无线链路控制层和媒体接入控制(medium access control，MAC)层的功能，还可以完成部分物理层或全部物理层的功能。有关上述各个协议层的具体描述，可以参考第三代合作伙伴计划(3rd generation partnership project，3GPP)的相关技术规范。网络设备可以是宏基站，也可以是微基站或室内站，还可以是中继节点或施主节点等。
本申请中所提及的网络设备也可以是核心网(core network，CN)中的一个或多个CN设备。例如，本申请所提及的网络设备可以是以下CN设备中的一个或多个：网络切片选择功能(network slice selection function，NSSF)、网络开放功能(network exposure function，NEF)、网络存储功能(network function repository function，NRF)、策略控制功能(policy control function，PCF)、统一数据管理(unified data management，UDM)、应用功能(application function，AF)、网络控制功能(network control function，NCF)、特定网络切片身份验证和鉴权功能(network slice specific authentication and authorization function，NSSAAF)、认证服务器功能(authentication server function，AUSF)、接入与移动性管理功能(access and mobility management function，AMF)、会话管理功能(session management function，SMF)、用户面功能(user plane function，UPF)、服务通信代理(service communication proxy，SCP)、网络切片准入控制(Network Slice Admission Control Function，NSACF)。
本申请实施例中，用于实现网络设备功能的装置可以是网络设备本身，也可以是能够支持网络设备实现该功能的装置，例如芯片系统或可实现接入网设备功能的组合器件、部件，该装置可以被安装在网络设备中。本申请的实施例对网络设备所采用的具体技术和具体设备形态不做限定。
为了便于理解本申请技术方案，下面先对联邦学习进行讲解说明。
联邦学习主要是针对数据孤岛问题提出的一种解决方案。联邦学习的目标是在保证数据隐私安全及合法的基础上，实现多个参与方共同建模，以提升AI模型的效果。可以将联邦学习理解为一种分布式的机器学习技术。联邦学习的过程可以分为一轮联邦学习和多轮联邦学习，下面以一轮联邦学习过程进行示例性说明，请参见图2所示，图2为一种联邦学习的流程示意图。
S201、联邦学习对应的服务器向多个候选用户设备发送资源请求。
S202、联邦学习对应的服务器接收每个候选用户设备上报的资源信息。
S203、联邦学习对应的服务器根据每个候选用户设备发送的资源信息，确定参与联邦学习的用户设备。
S204、联邦学习对应的服务器向每个参与联邦学习的用户设备，下发全局模型。
S205、每个参与联邦学习的用户设备对该全局模型进行模型更新，得到更新后的模型参数(包括模型权重和梯度值)。
S206、联邦学习对应的服务器接收每个参与联邦学习的用户设备发送的更新后的模型 参数。
并且，联邦学习对应的服务器在接收每个参与联邦学习的用户设备发送的更新后的模型参数之后，还可以向各个参与联邦学习的用户设备发送携带确认字符(Acknowledge character，ACK)信号。
S207、联邦学习对应的服务器根据各个更新后的模型参数，进行模型聚合。
通常，在选择联邦学习的参与用户设备(或称为参与方)时，会基于算力信息和信号质量确定参与用户设备。但通过这样的方式确定的用户设备，其参与联邦学习的过程中的稳定性无法保障，即可能存在该参与用户设备不能完整参与联邦学习的情况，进而导致联邦学习中断，影响联邦学习的训练时延和训练模型的精度。
为提高参与联邦学习的用户设备的可靠性和稳定性，本申请提供了一种用户设备选择方法、装置、芯片及模组设备。下面进一步对本申请实施例提供的用户设备选择方法、装置、芯片及模组设备进行详细描述。
图3是本申请实施例提供的一种用户设备选择方法的流程示意图。如图3所示，该用户设备选择方法包括如下S301～S303。图3所示的方法执行主体可以为用户设备或用户设备中的芯片。或者，图3所示的方法执行主体可以为网络设备或网络设备中的芯片。图3以用户设备和网络设备为方法的执行主体为例进行说明。
S301、网络设备向第一候选用户设备组发送第一消息。
换言之，网络设备选择第一候选用户设备组，并向该第一候选用户设备组中的各个候选用户设备发送第一消息。
为了便于理解，下面先对第一消息进行说明，之后再对网络设备选择第一候选用户设备组的过程进行说明。
第一消息指示的内容可以分为以下两种情况，具体地请参见图4和图5所示，图4和图5均为本申请提供的用户设备选择方法的流程示意图，其中：
情况一：请参见图4中S401所示，网络设备向多个候选用户设备(即第一候选用户设备组)发送第一消息，该第一消息用于指示联邦学习的需求信息。
其中，该联邦学习的需求信息包括联邦学习的时长需求信息，和/或，联邦学习对用户设备稳定性的需求信息。需要理解的是，该联邦学习的时长需求信息用于指示该联邦学习预计占用用户设备的时长。联邦学习对用户设备稳定性的需求信息，用于指示参与该联邦学习的用户设备的设备稳定性条件。
在一个可能的实现中，该联邦学习对用户设备稳定性的需求信息包括对用户设备移动性的需求信息，和/或，对用户设备运行状态的需求信息。其中，对用户设备移动性的需求信息，可以理解为该用户设备为低移动性等级、该用户设备不处于小区边缘、或该用户设备的信号质量变化率不大于第一阈值等等；对用户设备运行状态的需求信息，可以理解为该用户设备算力变化率不大于第二阈值等等。该第一阈值的具体数值和第二阈值的具体数值可根据具体应用场景进行调整，本申请对此不进行具体限定。
在另一个可能的实现中，该联邦学习对用户设备稳定性的需求信息，包括但不限于以下信息中的一种或多种：移动性等级需求信息、用户设备的位置需求信息、算力变化率需 求信息或信号质量变化率需求信息。
需要声明的是，本申请所提及的联邦学习的需求信息，除了包括上述提及的需求信息(即联邦学习的时长需求信息和联邦学习对用户设备稳定性的需求信息)之外，还可以包括其他需求信息，例如联邦学习对算力的需求信息、对信号质量的需求信息等。
情况二：请参见图5中的S501所示，网络设备向多个候选用户设备(即第一候选用户设备组)发送第一消息，该第一消息用于请求用户设备的测量信息。
换言之，接收到该第一消息的用户设备，可以上报自身的测量信息。其中，测量信息包括该用户设备的内部测量信息(例如算力变化率、温度变化率等)，以及空口测量信息(例如移动性等级、信号质量变化率等)。
在一个可能的实现中，该测量信息包括但不限于以下信息中的一种或多种：移动性等级、用户设备的位置、算力变化率或信号质量变化率。需要理解的是，移动性等级用于指示用户设备的移动性，例如，用户设备的移动性等级为低移动性(即可以理解为该用户设备不移动、不频繁移动或只在授权区域内移动)；用户设备的位置信息用于指示用户设备在其所处小区的相对位置信息，用户设备可以根据信号质量等信息感知在其所处小区的相对位置信息；算力变化率用于指示用户设备在一段时间内(例如1小时内)算力的变化量，算力变化率越小，则用户设备越稳定；信号质量变化率用于指示用户设备在在一段时间内(例如1小时内)信号质量的变化量，信号质量变化率越小，则用户设备越稳定。
在对第一消息进行说明之后，下面再对网络设备选择第一候选用户设备组的过程进行说明。
网络设备选择第一候选用户设备组的过程可以理解为，网络设备获取网络侧记录的多个用户设备信息，并根据该多个用户设备信息选择多个可能参与联邦学习的用户设备(即候选用户设备)，这些可能参与联邦学习的用户设备即组成了第一候选用户设备组。
需要说明的是，上述网络侧可以是该网络设备本身，也可以是其他设备(可以理解为与该网络设备具有通信连接的设备)。该用户设备信息包括但不限于能力信息或算力信息。其中，能力信息用于描述用户设备的能力，包括但不限于是否支持联邦学习的能力；算力信息用于描述该用户设备的计算性能，可以描述该用户设备每秒钟(或一段时间内)可进行操作运算的次数。换言之，在一个可能的实施方式中，网络设备接收来自其他网络设备的第四指示消息，该第四指示消息用于指示至少一个用户设备的测量信息，该第四指示消息用于确定该第一候选用户设备组。同理，在另一个可能的实施方式中，网络设备还可以向其他网络设备发送第五指示消息，该第五指示消息用于指示至少一个用户设备的测量信息，该至少一个用户设备的测量信息用于其他网络设备确定其他联邦学习(或该联邦学习的下一轮联邦学习)的候选用户设备。
示例性地，网络设备获取到网络侧记录的多个用户设备的信息可以参见表1所示。
表1
用户设备标识 能力信息 算力信息
UE1 支持联邦学习 4.1×109
UE2 支持联邦学习 6×109
UE3 支持联邦学习 5.3×109
UE4 支持联邦学习 4.8×109
UE5 支持联邦学习 4.5×109
UE6 支持联邦学习 3.2×109
UE7 不支持联邦学习 6.2×109
网络设备根据表1所示的网络侧记录的多个用户设备的信息、以及联邦学习的需求信息(例如该联邦学习对算力的需求信息为3.5×109)，选择多个有可能参与联邦学习的用户设备(即第一候选用户设备组)：UE1、UE2、UE3、UE4和UE5。进一步地，网络设备向该多个有可能参与联邦学习的用户设备(即UE1、UE2、UE3、UE4和UE5)发送该联邦学习的需求信息。其中，网络设备发送该联邦学习的需求信息的方式可以是一对一单播发送的方式，也可以是一对多组播发送的方式，本申请对此不进行具体限定。
S302、网络设备接收来自第二候选用户设备组的第二消息，该第一候选用户设备组包括该第二候选用户设备组。
换言之，第一候选用户设备组中的用户设备在接收到第一消息之后，可以选择向该网络设备发送第二消息，也可以不向该网络设备发送第二消息。可以理解为第二候选用户设备组中用户设备的数量，小于等于该第一候选用户设备组中用户设备的数量，该第二候选用户设备组为第一候选用户设备组的一个子集。需要说明的是，网络设备在一定的时长内未接受到某个候选用户设备发送的第二消息，则可视为该候选用户设备拒绝参与联邦学习。
下面结合S301中第一消息的两种情况，对第二消息进行说明。
情况一：第一消息用于指示联邦学习的需求信息。在这种情况下，该第二消息用于指示对应用户设备是否满足该联邦学习的需求信息，和/或，该对应用户设备的测量信息。
换言之，此种情况中候选用户设备在接收到来自网络设备的第一消息之后，具体请参见图4中S402所示，各个候选用户设备基于自身的测量信息和该联邦学习的需求信息，评估是否参与联邦学习；进一步地，如图4中S403所示，当候选用户设备的测量信息满足该联邦学习的需求信息时，该候选用户设备可以向网络设备发送第二消息，该第二消息用于指示满足联邦学习的需求信息(即接受联邦学习)，和/或，该候选用户设备的测量信息。
在一个可能的实现方式中，当该候选用户设备的测量信息不满足该联邦学习的需求信息时，该候选用户设备可以不向网络设备发送第二消息。或者，该候选用户设备也可以向网络设备发送第二消息，该第二消息用于指示不满足联邦学习的需求信息(即拒绝联邦学习)，和/或，该候选用户设备的测量信息。
情况二：第一消息用于请求用户设备的测量信息。在这种情况下，候选用户设备向网络设备发送第二消息，该第二消息包括对应用户设备的测量信息。
换言之，此种情况中候选用户设备在接收到来自网络设备的第一消息之后，具体请参见图5中S502所示，各个该候选用户设备向网络设备发送第二消息，该第二消息用于指示各个候选用户设备自身的测量信息，以便网络设备接收该第二消息之后，可以根据该第二消息评估该候选用户设备是否可以参与联邦学习。
可见，候选用户设备的测量信息，是评估该候选用户设备是否满足联邦学习的需求信息的关键。下面再对候选用户设备获取自身的测量信息的方式进行说明。需要说明的是，下述候选用户设备获取自身的测量信息的方式(即下述方式一或方式二)，均适用于前述情 况一和情况二。
方式一：候选用户设备接收到第一消息之后，候选用户设备从存储空间中获取记录存储的用户设备的测量信息。
例如，在第一时刻候选用户设备A进行了测量，得到测量信息为：低移动性、不处于小区边缘、算力变化率为0.1、信号质量变化率为0.2。进一步地，候选用户设备A记录存储该第一时刻的测量信息。在这种情况下，若候选用户设备A在第二时刻(第一时刻之后的某一时刻)接收到联邦学习的需求信息，则该候选用户设备A从存储空间中获取第一时刻的测量信息，可以快速地根据该测量信息评估是否满足该联邦学习的需求信息，提升评估效率。
但由于存储空间中记录存储的用户设备的测量信息具有时效性，可以理解为用户设备的移动性等级、用户设备的位置信息、算力变化率或信号质量变化率均可能随着时间的推移而发生变化。换言之。第一时刻和第二时刻之间时间间隔越小，则第一时刻的测量信息与第二时刻用户设备对应的真实测量信息越接近。若第一时刻和第二时刻之间时间间隔较大，则有可能导致评估失误。
为了避免由于测量信息的时效性导致评估失误的情况，在一个可能的实施方式中，网络设备在接收第二消息之前，网络设备向候选用户设备发送第一指示消息，该第一指示消息用于指示更新该测量信息。
换言之，候选用户设备接收来自网络设备的第一指示消息，该第一指示消息用于指示更新测量信息，进一步地，该候选用户设备根据该第一指示消息更新测量信息。通过这样的方法更新测量信息，确保该测量信息的准确性。
在另一个可能的实施方式中，网络设备在接收第二消息之前，网络设备向候选用户设备发送第一指示消息，该第一指示消息用于指示更新该测量信息，该第一指示消息中还包括测量需求信息。其中，测量需求信息包括但不限于对测量信息的时间需求(例如在用户设备接收到第一消息的时刻起，5分钟内的测量信息)。
换言之，候选用户设备接收来自网络设备的第一指示消息，该第一指示消息用于指示更新测量信息，该第一指示消息还包括测量需求信息，若候选用户设备不满足该测量需求信息，则基于该第一指示消息更新测量信息。
示例性地，该测量需求信息为“接收第一消息的时刻起，5分钟内的测量信息”，若候选用户设备A存储空间存储的测量信息为第一时刻记录的，候选用户设备A在第二时刻(第一时刻之后的某一时刻)接收到联邦学习的需求信息，第一时刻与第二时刻之间的间隔为10分钟，则可认为候选用户设备A不满足该测量需求信息。进一步地，候选用户设备A基于第一指示消息更新测量信息。
需要说明的是，基于第一指示消息更新测量信息，可以理解为若该第一指示消息中包括测量配置参数，则候选用户设备根据第一指示消息中的测量配置参数更新测量信息；若该第一指示消息中不包括测量配置参数，则候选用户设备根据预配置的测量配置参数更新测量信息。其中，测量配置参数包括但不限于：测量目标、数据格式、测量时长或测量频率中的一种或多种。
方式二：候选用户设备接收到第一消息之后，候选用户设备根据该联邦学习的需求信 息对用户设备进行实时测量，得到该候选用户设备的测量信息。
在这种方式中，候选用户设备可以根据联邦学习的需求信息的具体内容，进行实时测量，得到该候选用户设备的测量信息。
例如，联邦学习的需求信息包括对用户设备移动性的需求信息、和对用户设备的运行状态的需求信息。候选用户设备根据该联邦学习的需求信息，对用户设备的移动性相关信息(例如移动性等级、信号质量变化率或用户设备的位置)进行测量，以及对用户设备的运行状态相关信息(例如算力变化率)进行测量，得到测量信息。
S303、网络设备基于该第二消息，确定参与联邦学习的目标用户设备组，该目标用户设备组中每个目标用户设备满足该联邦学习的需求信息。
网络设备接收第二候选用户设备组中各个候选用户设备的响应消息之后，从测量信息满足联邦学习的需求信息的候选用户设备中确定多个目标用户设备，该多个目标用户设备用于参与联邦学习，即该多个目标用户设备为参与该联邦学习的目标用户设备组。进一步地，网络设备与各个目标用户设备之间建立协议数据单元(Protocol Data Unit，PDU)会话，网络设备向该多个目标用户设备中各个目标用户设备参与联邦学习的配置参数，各个目标用户设备根据配置参数开始进行联邦学习。
例如，网络设备接收来自UE1～UE5的响应消息，其中UE1～UE4发送的响应消息为参与联邦学习的响应消息，UE5发送的响应消息为拒绝联邦学习的响应消息。在这种情况下，网络设备可以将UE1～UE4确定为参与联邦学习的目标用户设备；或者，网络设备可以从UE1～UE4中选择部分用户设备为参与联邦学习的目标用户设备。
在一个可能的实施方式中，当第二消息用于指示对应用户设备是否满足该联邦学习的需求信息，和/或，该对应用户设备的测量信息的情况下，请参见图4中S404所示，该S404中网络设备基于第二消息确定参与联邦学习的目标用户设备组，可以理解为网络设备从第二消息指示满足联邦学习的需求信息的候选用户设备中，确定目标用户设备组。换言之，由用户设备评估(或决策)是否参与联邦学习。
在另一个可能的实施方式中，当第二消息包括对应用户设备的测量信息的情况下，请参见图5中S503所示，该S503中网络设备基于第二消息确定参与联邦学习的目标用户设备组，可以理解为网络设备根据各个候选用户设备的测量信息，评估各个候选用户设备的测量信息是否满足联邦学习的需求信息，从测量信息满足联邦学习的需求信息的候选用户设备中确定多个目标用户设备，换言之，由网络设备评估(或决策)是否参与联邦学习。
可见，基于图3所提供的用户设备选择方法，可以提升参与联邦学习的用户设备的稳定性，尽量保证该用户设备能完整的参与联邦学习，从而降低联邦学习的中断概率。
为了增强参与用户设备的稳定性和联邦学习的鲁棒性，以图3～图5任一所提供方法为基础，在网络设备确定目标用户设备组之后，本申请还提供如下几种优化方案，具体请参见图6、图7a、图7b或图8所示。
请参见图6所示，图6为本申请提供的一种用户设备选择的优化方案的流程示意图，通过图6所示的方法可以持续地对各个目标用户设备进行监测，使得网络设备了解各个目 标用户设备的运行状态。
S601、网络设备向目标用户设备组中各个目标用户设备发送第二指示消息，该第二指示消息用于指示持续监测是否满足该联邦学习的需求信息。
换言之，网络设备选择出参与联邦学习多个目标用户设备之后，网络设备向该各个目标用户发送指示消息，指示各个目标用户设备在进行联邦学习的过程中持续监测自身的测量信息是否满足联邦学习的需求信息。在一个可能的实施方式中，在目标用户设备未感知(或理解为不知晓)到联邦学习的需求信息的情况下，该第二指示消息还包括联邦学习的需求信息。
需要知晓的是，第二指示消息的发送时机不限于目标用户设备组被确定之后。在一个可能的实现中，在向第一候选用户设备组中各个候选用户设备发送第一消息时(或在发送第一消息之前，或者在发送第一消息之后接收第二消息之前)，网络设备向各个候选用户设备发送第二指示消息，该第二指示消息用于指示各个候选用户设备在被确定为目标用户设备之后，需要持续监测自身的测量信息是否满足该联邦学习的需求信息。
S602、网络设备接收第一目标用户设备的第三指示消息，该第三指示消息用于指示第一目标用户设备不满足该联邦学习的需求，该第一目标用户设备为目标用户设备组中的一个。
在该多个目标用户设备中的某个目标用户设备(即第一目标用户设备)监测到自身的测量信息不满足该联邦学习的需求时，第一目标用户设备向网络设备发送第三指示消息，该第三指示消息用于指示第一目标用户设备不满足该联邦学习的需求。
S603、网络设备从目标用户设备组中移除该第一目标用户设备；或者，降低该第一目标用户设备在目标用户设备组中的优先级。
当第一目标用户设备不再满足联邦学习的需求信息之后，网络设备可以从目标用户设备组中移除该第一目标用户设备；或者，网络设备可以降低该第一目标用户设备在目标用户设备组中的优先级。需要说明的是，从目标用户设备组中移除该第一目标用户设备，可以理解为在该联邦学习为多轮联邦学习的情况下，该第一目标用户设备不再参与本轮联邦学习之后的联邦学习。降低第一目标用户设备在目标用户设备组中的优先级，可以理解为在该联邦学习为多轮联邦学习的情况下，该第一目标用户设备可能会参与本轮联邦学习之后的联邦学习。
示例性的，目标用户设备组为UE1～UE4。网络设备向UE1～UE4发送第二指示消息，使得UE1～UE4中的各个UE持续监测自身的测量信息是否满足该联邦学习的需求信息。若UE3(即第一目标用户设备)监测到自身不满足该联邦学习的需求信息，则UE3向网络设备发送第三指示消息，该第三指示消息用于指示UE3不满足该联邦学习的需求信息。进一步地，网络设备从目标用户设备组中移除该UE3，即可以理解为该联邦学习的目标用户设备组包括UE1、UE2和UE4。或者，网络设备在该目标用户设备组中降低该UE3的优先级，可以理解为在该联邦学习为多轮联邦学习的情况下，若下一轮联邦学习在不考虑该UE3之后目标用户设备数量足够，则网络设备不考虑UE3参与该下一轮联邦学习；若下一轮联邦学习在不考虑该UE3之后目标用户设备数量不够，则UE3依然会参与下一轮联邦学习。
请参见图7a和图7b所示，图7a和图7b为本申请提供的用户设备选择的优化方案的 流程示意图。通过图7a和图7b所示的方法，为进行联邦学习的第二目标用户设备在执行切换的场景中，提供一种执行方案，可以增加该联邦学习系统的鲁棒性。
需要说明的是，在为第二目标用户设备执行切换时，可以理解为第二目标用户设备的源基站接收第二目标用户设备关于测量报告之后，源基站向该第二目标用户设备配置目标基站的连接参数之前；或者，理解为第二目标用户设备接收到目标基站的连接参数，执行切换的过程中；还可以理解为，第二目标用户设备成功从源基站切换连接至目标基站之后。
在一个可能的实施例中，第二目标用户设备为目标用户设备组中的一个，在为第二目标用户设备执行切换时，网络设备可以保持该第二目标用户设备继续参与该联邦学习。请参见图7a所示，其中：
S701、网络设备向目标基站发送全局模型。
S702、目标基站向第二目标用户设备下发全局模型。
S703、第二目标用户设备对该全局模型进行模型更新，得到更新后的模型参数(包括模型权重和梯度值)。
S704、第二目标用户设备向目标基站发送该更新后的模型参数(包括模型权重和梯度值)。
S705、目标基站向该网络设备发送该更新后的模型参数(包括模型权重和梯度值)。
换言之，在图7a所示的方法中，第二目标用户设备切换连接至目标基站之后，网络设备可以通过目标基站向第二目标用户设备下发全局模型，第二目标用户设备也可以通过目标基站向网络设备发送更新后的模型参数。在一个可能的实施方式中，网络设备还可以向目标基站发送该用户设备的测量信息，该用户设备的测量信息用于目标基站确定其他联邦学习(或本轮联邦学习之后的下一轮联邦学习)的第一候选用户设备组。
在另一个可能的实施例中，第二目标用户设备为目标用户设备组中的一个，在为第二目标用户设备执行切换时，网络设备释放该第二目标用户设备。请参见图7b所示，其中：
S711(可选的)、网络设备向第二目标用户设备发送通知消息，该通知消息用于通知第二目标用户设备不再参与联邦学习。
需要知晓的是，网络设备可以向第二目标用户发送显式的通知消息，通知第二目标用户设备不再参与联邦学习。网络设备还可以向第二目标用户发送隐式的通知消息(例如不再向第二目标用户发送联邦学习相关的消息)，以使第二目标用户设备不再参与联邦学习。
S712、网络设备从目标用户设备组中移除该第二目标用户设备。
换言之，在图7b所示的方法中，第二目标用户设备在执行切换时，网络设备可以释放该第二目标用户设备，第二目标用户设备不再参与联邦学习，并从目标用户设备组中移除该第二目标用户设备。
请参见图8所示，图8为本申请提供的一种用户设备选择的优化方案的流程示意图。通过图8所示的方法，为进行联邦学习的第三目标用户设备提供一种触发建立双连接的方案，增加第三目标用户设备在在联邦学习中的稳定性。
S801、网络设备向邻基站发送辅节点添加请求消息，该辅节点添加请求消息用于建立 第三目标用户设备与邻基站之间的通信连接，该第三目标用户设备为目标用户设备组中的一个。
换言之，网络设备为第三目标用户设备的服务基站，当该网络设备感知到该第三目标用户设备处于小区边缘时，该网络设备可以为第三目标用户设备建立双连接，即向该网络设备的邻基站发送辅节点添加请求消息。
S802、邻基站向网络设备发送辅节点添加请求确认消息。
该辅节点添加请求确认消息中包括邻基站对应辅小区的配置信息(包括该辅小区的无线连接参数等)。
S803、网络设备向第三目标用户设备发送RRC重配置消息。
该RRC重配置消息中包括辅小区的配置信息(包括该辅小区的无线连接参数等)。
S804、第三目标用户设备向网络设备发送RRC重配置完成消息。
S805、网络设备向邻基站发送辅节点重配置完成消息。
S806、第三目标用户设备随机接入该邻基站，即可以理解为第三目标用户设备与邻基站建立通信连接。
进一步地，在邻基站根据该辅节点添加请求消息，与第三目标用户设备建立通信连接之后，网络设备和邻基站均有可能接收来自第三目标用户设备的联邦学习数据(例如更新后的模型参数和梯度值)。在一个可能的实现方式中，在邻基站接收来自第三目标用户设备的联邦学习数据之后，邻基站默认会将该第三目标用户设备的联邦学习数据发送至网络设备。在另一个可能的实现中，网络设备向邻基站发送第六指示消息，该第六指示消息用于指示发送第三目标用户设备的联邦学习数据；进一步地，邻基站根据该第六指示消息，向网络设备发送该第三目标用户设备的联邦学习数据。
请参见图9，图9是本发明实施例提供的一种用户设备选择装置的结构示意图，该用户设备选择装置可以为网络设备或具有网络设备功能的装置(例如芯片)。具体的，如图9所示，用户设备选择装置900，可以包括：
发送单元901，用于向第一候选用户设备组发送第一消息；接收单元902，用于接收来自第二候选用户设备组的第二消息，该第一候选用户设备组包括第二候选用户设备组；其中，第一消息用于指示联邦学习的需求信息，该第二消息用于指示对应用户设备是否满足所述联邦学习的需求信息，和/或，对应用户设备的测量信息；或者，第一消息用于请求用户设备的测量信息，该第二消息包括所述测量信息；该联邦学习的需求信息包括联邦学习的时长需求信息，和/或，联邦学习对用户设备稳定性的需求信息；确定单元903，用于基于第二消息，确定参与联邦学习的目标用户设备组，该目标用户设备组中每个目标用户设备满足联邦学习的需求信息。
在一个可能的实施方式中，该联邦学习对用户设备稳定性的需求信息包括对用户设备移动性的需求信息，和/或，对用户设备运行状态的需求信息。
在一个可能的实施方式中，该联邦学习对用户设备稳定性的需求信息包括以下项的一种或多种：移动性等级需求信息、用户设备的位置需求信息、算力变化率需求信息或信号质量变化率需求信息。
在一个可能的实施方式中，该测量信息包括以下项的一种或多种：移动性等级、用户设备的位置、算力变化率或信号质量变化率。
在一个可能的实施方式中，发送单元901，还用于向第一候选用户设备组发送第一指示消息，该第一指示消息用于指示更新该测量信息。
在一个可能的实施方式中，该第一指示消息包括测量需求信息。
在一个可能的实施方式中，发送单元901，还用于向该目标用户设备组发送第二指示消息，该第二指示消息用于指示持续监测是否满足联邦学习的需求信息；接收单元902，还用于接收第一目标用户设备的第三指示消息，该第三指示消息用于指示第一目标用户设备不满足联邦学习的需求信息，该第一目标用户设备为目标用户设备组中的一个；确定单元903，还用于从目标用户设备组中移除该第一目标用户设备；或者，降低该第一目标用户设备在目标用户设备组中的优先级。
在一个可能的实施方式中，该第二指示消息还包括联邦学习的需求信息。
在一个可能的实施方式中，接收单元902，还用于接收来自其他网络设备的第四指示消息，该第四指示消息用于指示至少一个用户设备的测量信息，该第四指示消息用于确定所述第一候选用户设备组。
在一个可能的实施方式中，发送单元901，还用于向其他网络设备发送第五指示消息，该第五指示消息用于指示至少一个用户设备的测量信息。
在一个可能的实施方式中，在为第二目标用户设备执行切换时，确定单元903，还用于释放该第二目标用户设备，该第二目标用户设备为目标用户设备组中的一个；从该目标用户设备组中移除该第二目标用户设备。
在一个可能的实施方式中，发送单元901，用于向邻基站发送辅节点添加请求消息，该辅节点添加请求消息用于建立第三目标用户设备与邻基站之间的通信连接；该第三目标用户设备为目标用户设备组中的一个。
在一个可能的实施方式中，发送单元901，用于向邻基站发送第六指示消息，该第六指示消息用于指示发送第三目标用户设备的联邦学习数据；接收来自邻基站的第三目标用户设备的联邦学习数据。
请参见图10，图10是本发明实施例提供的一种用户设备选择装置的结构示意图，该用户设备选择装置可以为用户设备或具有用户设备功能的装置(例如芯片)。具体的，如图10所示，用户设备选择装置1000，可以包括：
接收单元1001，用于接收来自网络设备的第一消息；发送单元1002，用于向该网络设备发送第二消息；其中，第一消息用于指示联邦学习的需求信息，该第二消息用于指示对应用户设备是否满足联邦学习的需求信息，和/或，对应用户设备的测量信息；或者，第一消息用于请求用户设备的测量信息，该第二消息包括测量信息；该联邦学习的需求信息包括联邦学习的时长需求信息，和/或，联邦学习对用户设备稳定性的需求信息。
在一个可能的实施方式中，联邦学习对用户设备稳定性的需求信息包括对用户设备移动性的需求信息，和/或，对用户设备运行状态的需求信息。
在一个可能的实施方式中，联邦学习对用户设备稳定性的需求信息包括以下项的一种 或多种：移动性等级需求信息、用户设备的位置需求信息、算力变化率需求信息或信号质量变化率需求信息。
在一个可能的实施方式中，测量信息包括以下项的一种或多种：移动性等级、用户设备的位置、算力变化率或信号质量变化率。
在一个可能的实施方式中，接收单元1001，还用于接收来自网络设备的第一指示消息，该第一指示消息用于指示更新测量信息；基于该第一指示消息更新测量信息。
在一个可能的实施方式中，该装置还包括更新单元1003，该第一指示消息包括测量需求信息，该更新单元1003，用于若用户设备不满足该测量需求信息，则基于第一指示消息更新测量信息。
在一个可能的实施方式中，接收单元1001，还用于接收来自网络设备的第二指示消息，该第二指示消息用于指示持续监测是否满足联邦学习的需求信息；发送单元1002，还用于向网络设备发送第三指示消息，该第三指示消息用于指示用户设备不满足所述联邦学习的需求信息。
在一个可能的实施方式中，该第二指示消息还包括联邦学习的需求信息。
请参阅图11，图11是本发明实施例提供的一种网络设备的结构示意图。该网络设备1100可以包括存储器1101、处理器1102。可选的，还包括通信接口1103。存储器1101、处理器1102和通信接口1103通过一条或多条通信总线连接。其中，通信接口1103受处理器1102的控制用于收发信息。
存储器1101可以包括只读存储器和随机存取存储器，并向处理器1102提供指令和数据。存储器1101的一部分还可以包括非易失性随机存取存储器。
通信接口1103用于接收或发送数据。
处理器1102可以是中央处理单元(Central Processing Unit，CPU)，该处理器1102还可以是其他通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器，可选的，该处理器1102也可以是任何常规的处理器等。其中：
存储器1101，用于存储程序指令。
处理器1102，用于调用存储器1101中存储的程序指令。
处理器1102调用存储器1101中存储的程序指令，使该网络设备1100执行以下操作：向第一候选用户设备组发送第一消息；接收来自第二候选用户设备组的第二消息，该第一候选用户设备组包括第二候选用户设备组；其中，第一消息用于指示联邦学习的需求信息，该第二消息用于指示对应用户设备是否满足所述联邦学习的需求信息，和/或，对应用户设备的测量信息；或者，第一消息用于请求用户设备的测量信息，该第二消息包括所述测量信息；该联邦学习的需求信息包括联邦学习的时长需求信息，和/或，联邦学习对用户设备稳定性的需求信息；基于第二消息，确定参与联邦学习的目标用户设备组，该目标用户设备组中每个目标用户设备满足联邦学习的需求信息。
在一个可能的实施方式中，该联邦学习对用户设备稳定性的需求信息包括对用户设备移动性的需求信息，和/或，对用户设备运行状态的需求信息。
在一个可能的实施方式中，该联邦学习对用户设备稳定性的需求信息包括以下项的一种或多种：移动性等级需求信息、用户设备的位置需求信息、算力变化率需求信息或信号质量变化率需求信息。
在一个可能的实施方式中，该测量信息包括以下项的一种或多种：移动性等级、用户设备的位置、算力变化率或信号质量变化率。
在一个可能的实施方式中，处理器1102，还用于向第一候选用户设备组发送第一指示消息，该第一指示消息用于指示更新该测量信息。
在一个可能的实施方式中，该第一指示消息包括测量需求信息。
在一个可能的实施方式中，处理器1102，还用于向该目标用户设备组发送第二指示消息，该第二指示消息用于指示持续监测是否满足联邦学习的需求信息；接收第一目标用户设备的第三指示消息，该第三指示消息用于指示第一目标用户设备不满足联邦学习的需求信息，该第一目标用户设备为目标用户设备组中的一个；从目标用户设备组中移除该第一目标用户设备；或者，降低该第一目标用户设备在目标用户设备组中的优先级。
在一个可能的实施方式中，该第二指示消息还包括联邦学习的需求信息。
在一个可能的实施方式中，处理器1102，还用于接收来自其他网络设备的第四指示消息，该第四指示消息用于指示至少一个用户设备的测量信息，该第四指示消息用于确定所述第一候选用户设备组。
在一个可能的实施方式中，处理器1102，还用于向其他网络设备发送第五指示消息，该第五指示消息用于指示至少一个用户设备的测量信息。
在一个可能的实施方式中，在为第二目标用户设备执行切换时，处理器1102，还用于释放该第二目标用户设备，该第二目标用户设备为目标用户设备组中的一个；从该目标用户设备组中移除该第二目标用户设备。
在一个可能的实施方式中，处理器1102，用于向邻基站发送辅节点添加请求消息，该辅节点添加请求消息用于建立第三目标用户设备与邻基站之间的通信连接；该第三目标用户设备为目标用户设备组中的一个。
在一个可能的实施方式中，处理器1102，用于向邻基站发送第六指示消息，该第六指示消息用于指示发送第三目标用户设备的联邦学习数据；接收来自邻基站的第三目标用户设备的联邦学习数据。
请参阅图12，图12是本发明实施例提供的一种用户设备的结构示意图。该用户设备1200可以包括存储器1201、处理器1202。可选的，还包括通信接口1203。存储器1201、处理器1202和通信接口1203通过一条或多条通信总线连接。其中，通信接口1203受处理器1202的控制用于收发信息。
存储器1201可以包括只读存储器和随机存取存储器，并向处理器1202提供指令和数据。存储器1201的一部分还可以包括非易失性随机存取存储器。
通信接口1203用于接收或发送数据。
处理器1202可以是中央处理单元(Central Processing Unit，CPU)，该处理器1202还可以是其他通用处理器、数字信号处理器(Digital Signal Processor，DSP)、专用集成电路(Application Specific Integrated Circuit，ASIC)、现成可编程门阵列(Field-Programmable Gate Array，FPGA)或者其他可编程逻辑器件、分立门或者晶体管逻辑器件、分立硬件组件等。通用处理器可以是微处理器，可选的，该处理器1202也可以是任何常规的处理器等。其中：
存储器1201，用于存储程序指令。
处理器1202，用于调用存储器1201中存储的程序指令。
处理器1202调用存储器1201中存储的程序指令，使该用户设备1200执行以下操作：接收来自网络设备的第一消息；向该网络设备发送第二消息；其中，第一消息用于指示联邦学习的需求信息，该第二消息用于指示对应用户设备是否满足联邦学习的需求信息，和/或，对应用户设备的测量信息；或者，第一消息用于请求用户设备的测量信息，该第二消息包括测量信息；该联邦学习的需求信息包括联邦学习的时长需求信息，和/或，联邦学习对用户设备稳定性的需求信息。
在一个可能的实施方式中，联邦学习对用户设备稳定性的需求信息包括对用户设备移动性的需求信息，和/或，对用户设备运行状态的需求信息。
在一个可能的实施方式中，联邦学习对用户设备稳定性的需求信息包括以下项的一种或多种：移动性等级需求信息、用户设备的位置需求信息、算力变化率需求信息或信号质量变化率需求信息。
在一个可能的实施方式中，测量信息包括以下项的一种或多种：移动性等级、用户设备的位置、算力变化率或信号质量变化率。
在一个可能的实施方式中，处理器1202，还用于接收来自网络设备的第一指示消息，该第一指示消息用于指示更新测量信息；基于该第一指示消息更新测量信息。
在一个可能的实施方式中，该第一指示消息包括测量需求信息，该处理器1202，用于若用户设备不满足该测量需求信息，则基于第一指示消息更新测量信息。
在一个可能的实施方式中，处理器1202，还用于接收来自网络设备的第二指示消息，该第二指示消息用于指示持续监测是否满足联邦学习的需求信息；向网络设备发送第三指示消息，该第三指示消息用于指示用户设备不满足所述联邦学习的需求信息。
在一个可能的实施方式中，该第二指示消息还包括联邦学习的需求信息。
需要说明的是，图9～图12对应的实施例中未提及的内容以及各个步骤的具体实现方式可参见图3～图8所示实施例以及前述内容，这里不再赘述。
本申请实施例还提供了一种芯片，该芯片可以执行前述方法实施例中网络设备的相关步骤。该芯片，包括处理器和通信接口，该处理器被配置用于执行如下操作：向第一候选用户设备组发送第一消息；接收来自第二候选用户设备组的第二消息，该第一候选用户设备组包括第二候选用户设备组；其中，第一消息用于指示联邦学习的需求信息，该第二消息用于指示对应用户设备是否满足所述联邦学习的需求信息，和/或，对应用户设备的测量信息；或者，第一消息用于请求用户设备的测量信息，该第二消息包括所述测量信息；该 联邦学习的需求信息包括联邦学习的时长需求信息，和/或，联邦学习对用户设备稳定性的需求信息；基于第二消息，确定参与联邦学习的目标用户设备组，该目标用户设备组中每个目标用户设备满足联邦学习的需求信息。
在一个可能的实施方式中，该联邦学习对用户设备稳定性的需求信息包括对用户设备移动性的需求信息，和/或，对用户设备运行状态的需求信息。
在一个可能的实施方式中，该联邦学习对用户设备稳定性的需求信息包括以下项的一种或多种：移动性等级需求信息、用户设备的位置需求信息、算力变化率需求信息或信号质量变化率需求信息。
在一个可能的实施方式中，该测量信息包括以下项的一种或多种：移动性等级、用户设备的位置、算力变化率或信号质量变化率。
在一个可能的实施方式中，处理器，还用于向第一候选用户设备组发送第一指示消息，该第一指示消息用于指示更新该测量信息。
在一个可能的实施方式中，该第一指示消息包括测量需求信息。
在一个可能的实施方式中，处理器，还用于向该目标用户设备组发送第二指示消息，该第二指示消息用于指示持续监测是否满足联邦学习的需求信息；接收第一目标用户设备的第三指示消息，该第三指示消息用于指示第一目标用户设备不满足联邦学习的需求信息，该第一目标用户设备为目标用户设备组中的一个；从目标用户设备组中移除该第一目标用户设备；或者，降低该第一目标用户设备在目标用户设备组中的优先级。
在一个可能的实施方式中，该第二指示消息还包括联邦学习的需求信息。
在一个可能的实施方式中，处理器，还用于接收来自其他网络设备的第四指示消息，该第四指示消息用于指示至少一个用户设备的测量信息，该第四指示消息用于确定所述第一候选用户设备组。
在一个可能的实施方式中，处理器，还用于向其他网络设备发送第五指示消息，该第五指示消息用于指示至少一个用户设备的测量信息。
在一个可能的实施方式中，在为第二目标用户设备执行切换时，处理器，还用于释放该第二目标用户设备，该第二目标用户设备为目标用户设备组中的一个；从该目标用户设备组中移除该第二目标用户设备。
在一个可能的实施方式中，处理器，用于向邻基站发送辅节点添加请求消息，该辅节点添加请求消息用于建立第三目标用户设备与邻基站之间的通信连接；该第三目标用户设备为目标用户设备组中的一个。
在一个可能的实施方式中，处理器，用于向邻基站发送第六指示消息，该第六指示消息用于指示发送第三目标用户设备的联邦学习数据；接收来自邻基站的第三目标用户设备的联邦学习数据。
本申请实施例还提供了一种芯片，该芯片可以执行前述方法实施例中用户设备的相关步骤。该芯片，包括处理器和通信接口，该处理器被配置用于执行如下操作：接收来自网络设备的第一消息；向该网络设备发送第二消息；其中，第一消息用于指示联邦学习的需求信息，该第二消息用于指示对应用户设备是否满足联邦学习的需求信息，和/或，对应用 户设备的测量信息；或者，第一消息用于请求用户设备的测量信息，该第二消息包括测量信息；该联邦学习的需求信息包括联邦学习的时长需求信息，和/或，联邦学习对用户设备稳定性的需求信息。
在一个可能的实施方式中，联邦学习对用户设备稳定性的需求信息包括对用户设备移动性的需求信息，和/或，对用户设备运行状态的需求信息。
在一个可能的实施方式中，联邦学习对用户设备稳定性的需求信息包括以下项的一种或多种：移动性等级需求信息、用户设备的位置需求信息、算力变化率需求信息或信号质量变化率需求信息。
在一个可能的实施方式中，测量信息包括以下项的一种或多种：移动性等级、用户设备的位置、算力变化率或信号质量变化率。
在一个可能的实施方式中，处理器，还用于接收来自网络设备的第一指示消息，该第一指示消息用于指示更新测量信息；基于该第一指示消息更新测量信息。
在一个可能的实施方式中，该第一指示消息包括测量需求信息，该处理器，用于若用户设备不满足该测量需求信息，则基于第一指示消息更新测量信息。
在一个可能的实施方式中，处理器，还用于接收来自网络设备的第二指示消息，该第二指示消息用于指示持续监测是否满足联邦学习的需求信息；向网络设备发送第三指示消息，该第三指示消息用于指示用户设备不满足所述联邦学习的需求信息。
在一个可能的实施方式中，该第二指示消息还包括联邦学习的需求信息。
如图13所示，图13是本申请实施例提供的一种模组设备的结构示意图。该模组设备1300可以执行前述方法实施例中用户设备的相关步骤，该模组设备1300包括：通信模组1301、电源模组1302、存储模组1303以及芯片1304。
其中，所述电源模组1302用于为所述模组设备提供电能；所述存储模组1303用于存储数据和指令；所述通信模组1301用于进行模组设备内部通信，或者用于所述模组设备与外部设备进行通信；所述芯片1304用于：
向第一候选用户设备组发送第一消息；接收来自第二候选用户设备组的第二消息，该第一候选用户设备组包括第二候选用户设备组；其中，第一消息用于指示联邦学习的需求信息，该第二消息用于指示对应用户设备是否满足所述联邦学习的需求信息，和/或，对应用户设备的测量信息；或者，第一消息用于请求用户设备的测量信息，该第二消息包括所述测量信息；该联邦学习的需求信息包括联邦学习的时长需求信息，和/或，联邦学习对用户设备稳定性的需求信息；基于第二消息，确定参与联邦学习的目标用户设备组，该目标用户设备组中每个目标用户设备满足联邦学习的需求信息。
在一个可能的实施方式中，该联邦学习对用户设备稳定性的需求信息包括对用户设备移动性的需求信息，和/或，对用户设备运行状态的需求信息。
在一个可能的实施方式中，该联邦学习对用户设备稳定性的需求信息包括以下项的一种或多种：移动性等级需求信息、用户设备的位置需求信息、算力变化率需求信息或信号质量变化率需求信息。
在一个可能的实施方式中，该测量信息包括以下项的一种或多种：移动性等级、用户 设备的位置、算力变化率或信号质量变化率。
在一个可能的实施方式中，芯片1304，还用于向第一候选用户设备组发送第一指示消息，该第一指示消息用于指示更新该测量信息。
在一个可能的实施方式中，该第一指示消息包括测量需求信息。
在一个可能的实施方式中，芯片1304，还用于向该目标用户设备组发送第二指示消息，该第二指示消息用于指示持续监测是否满足联邦学习的需求信息；接收第一目标用户设备的第三指示消息，该第三指示消息用于指示第一目标用户设备不满足联邦学习的需求信息，该第一目标用户设备为目标用户设备组中的一个；从目标用户设备组中移除该第一目标用户设备；或者，降低该第一目标用户设备在目标用户设备组中的优先级。
在一个可能的实施方式中，该第二指示消息还包括联邦学习的需求信息。
在一个可能的实施方式中，芯片1304，还用于接收来自其他网络设备的第四指示消息，该第四指示消息用于指示至少一个用户设备的测量信息，该第四指示消息用于确定所述第一候选用户设备组。
在一个可能的实施方式中，芯片1304，还用于向其他网络设备发送第五指示消息，该第五指示消息用于指示至少一个用户设备的测量信息。
在一个可能的实施方式中，在为第二目标用户设备执行切换时，芯片1304，还用于释放该第二目标用户设备，该第二目标用户设备为目标用户设备组中的一个；从该目标用户设备组中移除该第二目标用户设备。
在一个可能的实施方式中，芯片1304，用于向邻基站发送辅节点添加请求消息，该辅节点添加请求消息用于建立第三目标用户设备与邻基站之间的通信连接；该第三目标用户设备为目标用户设备组中的一个。
在一个可能的实施方式中，芯片1304，用于向邻基站发送第六指示消息，该第六指示消息用于指示发送第三目标用户设备的联邦学习数据；接收来自邻基站的第三目标用户设备的联邦学习数据。
在一个实施例中，图13所示的模组设备中的芯片1304还可以用于：接收来自网络设备的第一消息；向该网络设备发送第二消息；其中，第一消息用于指示联邦学习的需求信息，该第二消息用于指示对应用户设备是否满足联邦学习的需求信息，和/或，对应用户设备的测量信息；或者，第一消息用于请求用户设备的测量信息，该第二消息包括测量信息；该联邦学习的需求信息包括联邦学习的时长需求信息，和/或，联邦学习对用户设备稳定性的需求信息。
在一个可能的实施方式中，联邦学习对用户设备稳定性的需求信息包括对用户设备移动性的需求信息，和/或，对用户设备运行状态的需求信息。
在一个可能的实施方式中，联邦学习对用户设备稳定性的需求信息包括以下项的一种或多种：移动性等级需求信息、用户设备的位置需求信息、算力变化率需求信息或信号质量变化率需求信息。
在一个可能的实施方式中，测量信息包括以下项的一种或多种：移动性等级、用户设备的位置、算力变化率或信号质量变化率。
在一个可能的实施方式中，芯片1304，还用于接收来自网络设备的第一指示消息，该第一指示消息用于指示更新测量信息；基于该第一指示消息更新测量信息。
在一个可能的实施方式中，该第一指示消息包括测量需求信息，该芯片1304，用于若用户设备不满足该测量需求信息，则基于第一指示消息更新测量信息。
在一个可能的实施方式中，芯片1304，还用于接收来自网络设备的第二指示消息，该第二指示消息用于指示持续监测是否满足联邦学习的需求信息；向网络设备发送第三指示消息，该第三指示消息用于指示用户设备不满足所述联邦学习的需求信息。
在一个可能的实施方式中，该第二指示消息还包括联邦学习的需求信息。
本申请实施例还提供一种计算机可读存储介质，该计算机可读存储介质中存储有指令，当其在处理器上运行时，上述方法实施例的方法流程得以实现。
本申请实施例还提供一种计算机程序产品，当所述计算机程序产品在处理器上运行时，上述方法实施例的方法流程得以实现。
关于上述实施例中描述的各个装置、产品包含的各个模块/单元，其可以是软件模块/单元，也可以是硬件模块/单元，或者也可以部分是软件模块/单元，部分是硬件模块/单元。例如，对于应用于或集成于芯片的各个装置、产品其包含的各个模块/单元可以都采用电路等硬件的方式实现，或者，至少部分模块/单元可以采用软件程序的方式实现，该软件程序运行于芯片内部集成处理器，剩余的(如果有)部分模块/单元可以采用电路等硬件方式实现；对于应用于或集成于芯片模组的各个装置、产品，其包含的各个模块/单元可以都采用电路等硬件的方式实现，不同模块/单元可以位于芯片模组的同一件(例如芯片、电路模块等)或者不同组件中，或者，至少部分模块/单元可以采用软件程序的方式实现，该软件程序运行于芯片模组内部集成的处理器，剩余的(如果有)部分模块/单元可以采用电路等硬件方式实现；对于应用于或集成于终端的各个装置、产品，其包含的模块/单元可以都采用电路等硬件的方式实现，不同的模块/单元可以位于终端内同一组件(例如，芯片、电路模块等)或者不同组件中，或者，至少部分模块/单元可以采用软件程序的方式实现，该软件程序运行于终端内部集成的处理器，剩余的(如果有)部分模块/单元可以采用电路等硬件方式实现。
需要说明的是，对于前述的各方法实施例，为了简单描述，故将其都表述为一系列的动作组合，但是本领域技术人员应该知悉，本申请并不受所描述的动作顺序的限制，因为依据本申请，某些操作可以采用其他顺序或者同时进行。其次，本领域技术人员也应该知悉，说明书中所描述的实施例均属于优选实施例，所涉及的动作和模块并不一定是本申请所必须的。
本申请提供的各实施例的描述可以相互参照，对各个实施例的描述都各有侧重，某个实施例中没有详述的部分，可以参见其他实施例的相关描述。为描述的方便和简洁，例如关于本申请实施例提供的各装置、设备的功能以及执行的操作可以参照本申请方法实施例的相关描述，各方法实施例之间、各装置实施例之间也可以互相参考、结合或引用。
最后应说明的是：以上各实施例仅用以说明本申请的技术方案，而非对其限制；尽管参照前述各实施例对本申请进行了详细的说明，本领域的普通技术人员应当理解：其依然 可以对前述各实施例所记载的技术方案进行修改，或者对其中部分或者全部技术特征进行等同替换；而这些修改或者替换，并不使相应技术方案的本质脱离本申请各实施例技术方案的范围。

Claims (30)

  1. 一种用户设备选择方法，其特征在于，所述方法包括：
    向第一候选用户设备组发送第一消息；
    接收来自第二候选用户设备组的第二消息，所述第一候选用户设备组包括所述第二候选用户设备组；
    其中，所述第一消息用于指示联邦学习的需求信息，所述第二消息用于指示对应用户设备是否满足所述联邦学习的需求信息，和/或，所述对应用户设备的测量信息；
    或者，所述第一消息用于请求用户设备的测量信息，所述第二消息包括所述测量信息；所述联邦学习的需求信息包括所述联邦学习的时长需求信息，和/或，所述联邦学习对用户设备稳定性的需求信息；
    基于所述第二消息，确定参与所述联邦学习的目标用户设备组，所述目标用户设备组中每个目标用户设备满足所述联邦学习的需求信息。
  2. 根据权利要求1所述方法，其特征在于，所述联邦学习对用户设备稳定性的需求信息包括对所述用户设备移动性的需求信息，和/或，对所述用户设备运行状态的需求信息。
  3. 根据权利要求1或2所述方法，其特征在于，所述联邦学习对用户设备稳定性的需求信息包括以下项的一种或多种：移动性等级需求信息、用户设备的位置需求信息、算力变化率需求信息或信号质量变化率需求信息。
  4. 根据权利要求1-3中任一项所述方法，其特征在于，所述测量信息包括以下项的一种或多种：移动性等级、用户设备的位置、算力变化率或信号质量变化率。
  5. 根据权利要求1-4中任一项所述方法，其特征在于，所述接收来自第二候选用户设备组的第二消息之前，所述方法还包括：
    向所述第一候选用户设备组发送第一指示消息，所述第一指示消息用于指示更新所述测量信息。
  6. 根据权利要求5所述方法，其特征在于，所述第一指示消息包括测量需求信息。
  7. 根据权利要求1-6中任一项所述方法，其特征在于，所述方法还包括：
    向所述目标用户设备组发送第二指示消息，所述第二指示消息用于指示持续监测是否满足所述联邦学习的需求信息；
    接收第一目标用户设备的第三指示消息，所述第三指示消息用于指示所述第一目标用户设备不满足所述联邦学习的需求信息；所述第一目标用户设备为所述目标用户设备组中的一个；
    从所述目标用户设备组中移除所述第一目标用户设备；或者，降低所述第一目标用户设备在所述目标用户设备组中的优先级。
  8. 根据权利要求7所述方法，其特征在于，所述第二指示消息还包括所述联邦学习的需求信息。
  9. 根据权利要求1所述方法，其特征在于，所述方法还包括：
    接收来自其他网络设备的第四指示消息，所述第四指示消息用于指示至少一个用户设备的所述测量信息，所述第四指示消息用于确定所述第一候选用户设备组。
  10. 根据权利要求1所述方法，其特征在于，所述方法还包括：
    向其他网络设备发送第五指示消息，所述第五指示消息用于指示至少一个用户设备的所述测量信息。
  11. 根据权利要求1-10中任一项所述方法，其特征在于，所述方法还包括：
    在为第二目标用户设备执行切换时，释放所述第二目标用户设备；所述第二目标用户设备为所述目标用户设备组中的一个；
    从所述目标用户设备组中移除所述第二目标用户设备。
  12. 根据权利要求1-7中任一项所述方法，其特征在于，所述方法还包括：
    向邻基站发送辅节点添加请求消息，所述辅节点添加请求消息用于建立第三目标用户设备与所述邻基站之间的通信连接；所述第三目标用户设备为所述目标用户设备组中的一个。
  13. 根据权利要求12所述方法，其特征在于，所述方法还包括：
    向所述邻基站发送第六指示消息，所述第六指示消息用于指示发送所述第三目标用户设备的联邦学习数据；
    接收来自所述邻基站的所述第三目标用户设备的联邦学习数据。
  14. 一种用户设备选择方法，其特征在于，所述方法包括：
    接收来自网络设备的第一消息；
    向所述网络设备发送第二消息；
    其中，所述第一消息用于指示联邦学习的需求信息，所述第二消息用于指示对应用户设备是否满足所述联邦学习的需求信息，和/或，所述对应用户设备的测量信息；或者，所述第一消息用于请求用户设备的测量信息，所述第二消息包括所述测量信息；所述联邦学习的需求信息包括所述联邦学习的时长需求信息，和/或，所述联邦学习对用户设备稳定性的需求信息。
  15. 根据权利要求14所述方法，其特征在于，所述联邦学习对用户设备稳定性的需求信息包括对所述用户设备移动性的需求信息，和/或，对所述用户设备运行状态的需求信息。
  16. 根据权利要求14或15所述方法，其特征在于，所述联邦学习对用户设备稳定性的需求信息包括以下项的一种或多种：移动性等级需求信息、用户设备的位置需求信息、算力变化率需求信息或信号质量变化率需求信息。
  17. 根据权利要求14-16中任一项所述方法，其特征在于，所述测量信息包括以下项的一种或多种：移动性等级、用户设备的位置、算力变化率或信号质量变化率。
  18. 根据权利要求14-17中任一项所述方法，其特征在于，所述向所述网络设备发送第二消息之前，所述方法还包括：
    接收来自所述网络设备的第一指示消息，所述第一指示消息用于指示更新所述测量信息；
    基于所述第一指示消息更新测量信息。
  19. 根据权利要求18所述方法，其特征在于，所述第一指示消息包括测量需求信息，所述基于所述第一指示消息更新测量信息，包括：
    若所述用户设备不满足所述测量需求信息，则基于所述第一指示消息更新测量信息。
  20. 根据权利要求14-17中任一项所述方法，其特征在于，所述方法还包括：
    接收来自所述网络设备的第二指示消息，所述第二指示消息用于指示持续监测是否满足所述联邦学习的需求信息；
    向所述网络设备发送第三指示消息，所述第三指示消息用于指示所述用户设备不满足所述联邦学习的需求信息。
  21. 根据权利要求20所述方法，其特征在于，所述第二指示消息还包括所述联邦学习的需求信息。
  22. 一种用户设备选择装置，其特征在于，所述装置包括：
    发送单元，用于向第一候选用户设备组发送第一消息；
    接收单元，用于接收来自第二候选用户设备组的第二消息，所述第一候选用户设备组包括所述第二候选用户设备组；
    其中，所述第一消息用于指示联邦学习的需求信息，所述第二消息用于指示对应用户设备是否满足所述联邦学习的需求信息，和/或，所述对应用户设备的测量信息；或者，所述第一消息用于请求用户设备的测量信息，所述第二消息包括所述测量信息；所述联邦学习的需求信息包括所述联邦学习的时长需求信息，和/或，所述联邦学习对用户设备稳定性的需求信息；
    确定单元，用于基于所述第二消息，确定参与所述联邦学习的目标用户设备组，所述目标用户设备组中每个目标用户设备满足所述联邦学习的需求信息。
  23. 一种用户设备选择装置，其特征在于，所述装置包括：
    接收单元，用于接收来自网络设备的第一消息；
    发送单元，用于向所述网络设备发送第二消息；
    其中，所述第一消息用于指示联邦学习的需求信息，所述第二消息用于指示对应用户设备是否满足所述联邦学习的需求信息，和/或，所述对应用户设备的测量信息；或者，所述第一消息用于请求用户设备的测量信息，所述第二消息包括所述测量信息；所述联邦学习的需求信息包括所述联邦学习的时长需求信息，和/或，所述联邦学习对用户设备稳定性的需求信息。
  24. 一种芯片，其特征在于，包括处理器和通信接口，所述处理器被配置用于执行如下操作：
    向第一候选用户设备组发送第一消息；
    接收来自第二候选用户设备组的第二消息，所述第一候选用户设备组包括所述第二候选用户设备组；
    其中，所述第一消息用于指示联邦学习的需求信息，所述第二消息用于指示对应用户设备是否满足所述联邦学习的需求信息，和/或，所述对应用户设备的测量信息；或者，所述第一消息用于请求用户设备的测量信息，所述第二消息包括所述测量信息；所述联邦学习的需求信息包括所述联邦学习的时长需求信息，和/或，所述联邦学习对用户设备稳定性的需求信息；
    基于所述第二消息，确定参与所述联邦学习的目标用户设备组，所述目标用户设备组中每个目标用户设备满足所述联邦学习的需求信息。
  25. 一种芯片，其特征在于，包括处理器和通信接口，所述处理器被配置用于执行如下操作：
    接收来自网络设备的第一消息；
    向所述网络设备发送第二消息；
    其中，所述第一消息用于指示联邦学习的需求信息，所述第二消息用于指示对应用户设备是否满足所述联邦学习的需求信息，和/或，所述对应用户设备的测量信息；或者，所述第一消息用于请求用户设备的测量信息，所述第二消息包括所述测量信息；所述联邦学习的需求信息包括所述联邦学习的时长需求信息，和/或，所述联邦学习对用户设备稳定性的需求信息。
  26. 一种模组设备，其特征在于，所述模组设备包括通信模组、电源模组、存储模组以及芯片，其中：
    所述电源模组用于为所述模组设备提供电能；
    所述存储模组用于存储数据和指令；
    所述通信模组用于进行模组设备内部通信，或者用于所述模组设备与外部设备进行通 信；
    所述芯片用于：向第一候选用户设备组发送第一消息；
    接收来自第二候选用户设备组的第二消息，所述第一候选用户设备组包括所述第二候选用户设备组；
    其中，所述第一消息用于指示联邦学习的需求信息，所述第二消息用于指示对应用户设备是否满足所述联邦学习的需求信息，和/或，所述对应用户设备的测量信息；或者，所述第一消息用于请求用户设备的测量信息，所述第二消息包括所述测量信息；所述联邦学习的需求信息包括所述联邦学习的时长需求信息，和/或，所述联邦学习对用户设备稳定性的需求信息；
    基于所述第二消息，确定参与所述联邦学习的目标用户设备组，所述目标用户设备组中每个目标用户设备满足所述联邦学习的需求信息。
  27. 一种模组设备，其特征在于，所述模组设备包括通信模组、电源模组、存储模组以及芯片，其中：
    所述电源模组用于为所述模组设备提供电能；
    所述存储模组用于存储数据和指令；
    所述通信模组用于进行模组设备内部通信，或者用于所述模组设备与外部设备进行通信；
    所述芯片用于：接收来自网络设备的第一消息；
    向所述网络设备发送第二消息；
    其中，所述第一消息用于指示联邦学习的需求信息，所述第二消息用于指示对应用户设备是否满足所述联邦学习的需求信息，和/或，所述对应用户设备的测量信息；或者，所述第一消息用于请求用户设备的测量信息，所述第二消息包括所述测量信息；所述联邦学习的需求信息包括所述联邦学习的时长需求信息，和/或，所述联邦学习对用户设备稳定性的需求信息。
  28. 一种网络设备，其特征在于，包括存储器和处理器，所述存储器用于存储计算机程序，所述计算机程序包括程序指令，所述处理器被配置用于调用所述程序指令，执行如权利要求1～13中任一项所述的方法。
  29. 一种用户设备，其特征在于，包括存储器和处理器，所述存储器用于存储计算机程序，所述计算机程序包括程序指令，所述处理器被配置用于调用所述程序指令，执行如权利要求14～21中任一项所述的方法。
  30. 一种计算机可读存储介质，其特征在于，所述计算机存储介质中存储有计算机可读指令，当所述计算机可读指令在通信装置上运行时，使得所述通信装置执行权利要求1～13中任一项所述的方法或执行如权利要求14～21中任一项所述的方法。
PCT/CN2023/071729 2022-01-26 2023-01-10 一种用户设备选择方法、装置、芯片及模组设备 WO2023143082A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210094316.6A CN116567702A (zh) 2022-01-26 2022-01-26 一种用户设备选择方法、装置、芯片及模组设备
CN202210094316.6 2022-01-26

Publications (1)

Publication Number Publication Date
WO2023143082A1 true WO2023143082A1 (zh) 2023-08-03

Family

ID=87470470

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/071729 WO2023143082A1 (zh) 2022-01-26 2023-01-10 一种用户设备选择方法、装置、芯片及模组设备

Country Status (2)

Country Link
CN (1) CN116567702A (zh)
WO (1) WO2023143082A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110598870A (zh) * 2019-09-02 2019-12-20 深圳前海微众银行股份有限公司 一种联邦学习方法及装置
CN111428885A (zh) * 2020-03-31 2020-07-17 深圳前海微众银行股份有限公司 一种联邦学习中用户的索引方法及联邦学习装置
CN111538598A (zh) * 2020-04-29 2020-08-14 深圳前海微众银行股份有限公司 联邦学习建模方法、装置、设备及可读存储介质
CN113255924A (zh) * 2020-11-25 2021-08-13 中兴通讯股份有限公司 联邦学习参与者选择方法、装置、设备和存储介质
US20210398026A1 (en) * 2020-10-16 2021-12-23 Beijing Baidu Netcom Science And Technology Co., Ltd. Federated learning for improving matching efficiency

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110598870A (zh) * 2019-09-02 2019-12-20 深圳前海微众银行股份有限公司 一种联邦学习方法及装置
CN111428885A (zh) * 2020-03-31 2020-07-17 深圳前海微众银行股份有限公司 一种联邦学习中用户的索引方法及联邦学习装置
CN111538598A (zh) * 2020-04-29 2020-08-14 深圳前海微众银行股份有限公司 联邦学习建模方法、装置、设备及可读存储介质
US20210398026A1 (en) * 2020-10-16 2021-12-23 Beijing Baidu Netcom Science And Technology Co., Ltd. Federated learning for improving matching efficiency
CN113255924A (zh) * 2020-11-25 2021-08-13 中兴通讯股份有限公司 联邦学习参与者选择方法、装置、设备和存储介质

Also Published As

Publication number Publication date
CN116567702A (zh) 2023-08-08

Similar Documents

Publication Publication Date Title
US11632788B2 (en) Data scheduling method, base station, and system
US8874112B2 (en) Service dependent inactivity timer
TWI602458B (zh) 處理節能的裝置及方法
US10172052B2 (en) Method and device for dynamically constructing virtual cell
KR20190073504A (ko) 데이터 전송/수신 장치 및 방법, 및 통신 시스템
WO2020221223A1 (zh) 通信方法、装置和系统
EP3043611A1 (en) Method, device and system for transmitting information
TWI753965B (zh) 切換通訊模式的方法、終端設備和網路設備
CN113973322A (zh) 一种通信方法及装置
US20220150784A1 (en) Handover method and apparatus
US20240056926A1 (en) Communication method and apparatus for obtaining load information
JP7111812B2 (ja) 通信装置のランダム・アクセス方法、装置、及び記憶媒体
JP2022502958A (ja) 早期データ伝送及び無線アクセスネットワーク通知エリア更新
WO2021189235A1 (zh) 一种数据传输方法及装置、通信设备
EP2991426B1 (en) Method and apparatus for sending information
CN105025593B (zh) 一种释放实体的方法及设备
WO2021190498A1 (zh) 传输模式转换指示方法、接入网设备及终端设备
JP2018507595A (ja) 無線装置、ネットワークノード、および、それらにおける方法
WO2021031035A1 (zh) 一种通信方法及装置
WO2023143082A1 (zh) 一种用户设备选择方法、装置、芯片及模组设备
CN107682874B (zh) 一种进行业务传输的方法和设备
WO2024169459A1 (zh) 通信方法与通信装置
WO2024007878A1 (zh) 通信方法及装置
WO2023103958A1 (zh) 一种通信方法及装置
WO2022228456A1 (zh) 载波配置方法及装置

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: 23745965

Country of ref document: EP

Kind code of ref document: A1

NENP Non-entry into the national phase

Ref country code: DE