WO2024120285A1 - 信息传输方法、装置、终端及网络侧设备 - Google Patents

信息传输方法、装置、终端及网络侧设备 Download PDF

Info

Publication number
WO2024120285A1
WO2024120285A1 PCT/CN2023/135261 CN2023135261W WO2024120285A1 WO 2024120285 A1 WO2024120285 A1 WO 2024120285A1 CN 2023135261 W CN2023135261 W CN 2023135261W WO 2024120285 A1 WO2024120285 A1 WO 2024120285A1
Authority
WO
WIPO (PCT)
Prior art keywords
model
information
network element
terminal
target network
Prior art date
Legal status (The legal status 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 status listed.)
Ceased
Application number
PCT/CN2023/135261
Other languages
English (en)
French (fr)
Inventor
崇卫微
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Vivo Mobile Communication Co Ltd
Original Assignee
Vivo Mobile Communication Co Ltd
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 Vivo Mobile Communication Co Ltd filed Critical Vivo Mobile Communication Co Ltd
Priority to EP23899864.5A priority Critical patent/EP4633213A4/en
Publication of WO2024120285A1 publication Critical patent/WO2024120285A1/zh
Priority to US19/230,149 priority patent/US20250300906A1/en
Anticipated expiration legal-status Critical
Ceased legal-status Critical Current

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W8/00—Network data management
    • H04W8/22—Processing or transfer of terminal data, e.g. status or physical capabilities
    • H04W8/24—Transfer of terminal data
    • H04W8/245—Transfer of terminal data from a network towards a terminal
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L61/00—Network arrangements, protocols or services for addressing or naming
    • H04L61/45—Network directories; Name-to-address mapping
    • H04L61/4505—Network directories; Name-to-address mapping using standardised directories; using standardised directory access protocols
    • H04L61/4511—Network directories; Name-to-address mapping using standardised directories; using standardised directory access protocols using domain name system [DNS]
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/08—Configuration management of networks or network elements
    • H04L41/0803—Configuration setting
    • H04L41/0823—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability
    • H04L41/083—Configuration setting characterised by the purposes of a change of settings, e.g. optimising configuration for enhancing reliability for increasing network speed
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/16—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
    • H04L41/34—Signalling channels for network management communication
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W24/00—Supervisory, monitoring or testing arrangements
    • H04W24/02—Arrangements for optimising operational condition
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W48/00—Access restriction; Network selection; Access point selection
    • H04W48/18—Selecting a network or a communication service
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W76/00—Connection management
    • H04W76/10—Connection setup
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W76/00—Connection management
    • H04W76/10—Connection setup
    • H04W76/12—Setup of transport tunnels
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W8/00—Network data management
    • H04W8/22—Processing or transfer of terminal data, e.g. status or physical capabilities
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W8/00—Network data management
    • H04W8/22—Processing or transfer of terminal data, e.g. status or physical capabilities
    • H04W8/24—Transfer of terminal data

Definitions

  • the present application belongs to the field of communication technology, and specifically relates to an information transmission method, device, terminal and network side equipment.
  • the communication technology field is studying the communication performance enhancement based on the machine learning (ML) model, so as to further improve the communication performance of the terminal.
  • ML machine learning
  • the terminal downloads or uploads the information of the ML model.
  • the terminal is currently unable to transmit the information of the ML model, such as being unable to obtain the information of the ML model from other devices or to send the information of the ML model to other devices, which leads to the poor performance of the transmission model of the terminal.
  • the embodiments of the present application provide an information transmission method, apparatus, terminal and network-side equipment, which can solve the problem of poor performance of the terminal's transmission model.
  • an information transmission method comprising:
  • the terminal obtains information of a target network element corresponding to the machine learning ML model, where the target network element supports a function of a user plane transmission model;
  • the terminal performs model transmission with the target network element by using a user plane session or a tunnel, wherein the tunnel is a tunnel between the terminal and the target network element established based on the user plane session, and the model transmission includes at least one of the following:
  • an information transmission method comprising:
  • the target network element uses a user plane session or a tunnel to perform model transmission with the terminal, wherein the target network element supports the function of the user plane transmission model, the tunnel is a tunnel between the target network element and the terminal established based on the user plane session, and the model transmission includes at least one of the following:
  • an information transmission method comprising:
  • the access mobility management function sends the information of the target network element to the terminal.
  • an information transmission method comprising:
  • the model training logic network element sends information of a target network element corresponding to the ML model to the access mobility management function or the terminal, and the target network element supports the function of the user plane transmission model.
  • an information transmission device comprising:
  • An acquisition module used to acquire information of a target network element corresponding to a machine learning ML model, wherein the target network element supports a function of a user plane transmission model
  • a transmission module configured to perform model transmission with the target network element by using a user plane session or a tunnel, wherein the tunnel is a tunnel between the terminal and the target network element established based on the user plane session, and the model transmission includes at least one of the following:
  • an information transmission device comprising:
  • a transmission module configured to perform model transmission between the terminal and the target network element corresponding to the device by using a user plane session or a tunnel, wherein the target network element supports the function of the user plane transmission model, the tunnel is a tunnel between the target network element and the terminal established based on the user plane session, and the model transmission includes at least one of the following:
  • an information transmission device comprising:
  • An acquisition module used to acquire information of a target network element corresponding to a machine learning ML model, wherein the target network element supports a function of a user plane transmission model
  • the first sending module is used to send the information of the target network element to the terminal.
  • an information transmission device comprising:
  • the sending module is used to send information of a target network element corresponding to the ML model to the access mobility management function or the terminal, where the target network element supports the function of the user plane transmission model.
  • a terminal comprising a processor and a memory, wherein the memory stores programs or instructions that can be run on the processor, and when the programs or instructions are executed by the processor, the steps of the information transmission method on the terminal side provided in the embodiment of the present application are implemented.
  • a terminal comprising a processor and a communication interface, wherein the communication interface is used to obtain information of a target network element corresponding to a machine learning ML model, and the target network element supports the function of a user plane transmission model; the model is transmitted between the target network element and the user plane session or tunnel, wherein the tunnel is a tunnel between the terminal and the target network element established based on the user plane session, and the model transmission includes at least one of the following: receiving information of the ML model from the target network element; sending information of the ML model to the target network element.
  • a network side device is provided, wherein the network side device is a target network element, including a processor and a memory.
  • a memory wherein the memory stores a program or instruction that can be run on the processor, and when the program or instruction is executed by the processor, the steps of the information transmission method on the target network element side provided in the embodiment of the present application are implemented.
  • a network side device which is a target network element, including a processor and a communication interface, wherein the communication interface is used to transmit a model with a terminal using a user plane session or a tunnel, wherein the target network element supports the function of user plane transmission model, and the tunnel is a tunnel between the target network element and the terminal established based on the user plane session, and the model transmission includes at least one of the following: sending information of the ML model to the terminal; receiving information of the ML model sent by the terminal.
  • a network side device which is an access mobility management function, including a processor and a memory, the memory storing programs or instructions that can be run on the processor, and the program or instructions, when executed by the processor, implement the steps of the information transmission method on the access mobility management function side provided in the embodiment of the present application.
  • a network side device which is an access mobility management function, including a processor and a communication interface, wherein the communication interface is used to obtain information about a target network element corresponding to a machine learning ML model, and the target network element supports the function of a user plane transmission model; and send the information about the target network element to a terminal.
  • a network side device which is a model training logic network element, including a processor and a memory, the memory storing programs or instructions that can be run on the processor, and the program or instructions, when executed by the processor, implement the steps of the information transmission method on the model training logic network element side as provided in the embodiment of the present application.
  • a network side device which is a model training logic network element, including a processor and a communication interface, wherein the communication interface is used to send information of a target network element corresponding to the ML model to an access mobility management function or a terminal, and the target network element supports the function of a user plane transmission model.
  • an information transmission system including: a terminal, a target network element, an access mobility management function and a model training logic network element
  • the terminal can be used to execute the steps of the information transmission method on the terminal side as provided in the embodiments of the present application
  • the target network element can be used to execute the steps of the information transmission method on the target network element side as provided in the embodiments of the present application
  • the access mobility management function can be used to execute the steps of the information transmission method on the access mobility management function side as provided in the embodiments of the present application
  • the model training logic network element can be used to execute the steps of the information transmission method on the model training logic network element side as provided in the embodiments of the present application.
  • a readable storage medium on which a program or instruction is stored.
  • the steps of the information transmission method on the terminal side as provided in the embodiment of the present application are implemented, or the steps of the information transmission method on the target network element side as provided in the embodiment of the present application are implemented, or the steps of the information transmission method on the access mobility management function side as provided in the embodiment of the present application are implemented, or the steps of the information transmission method on the model training logic network element side as provided in the embodiment of the present application are implemented.
  • a chip comprising a processor and a communication interface, the communication interface is coupled to the processor, the processor is used to run a program or instruction, implement the information transmission method on the terminal side as provided in the embodiment of the present application, or implement the information transmission method on the target network element side as provided in the embodiment of the present application, or implement the information transmission method on the access mobility management function side as provided in the embodiment of the present application, or implement the model training method as provided in the embodiment of the present application.
  • a computer program/program product is provided, which is stored in a storage medium, and is executed by at least one processor to implement the steps of the information transmission method on the terminal side as provided in the embodiments of the present application, or the computer program/program product is executed by at least one processor to implement the steps of the information transmission method on the target network element side as provided in the embodiments of the present application, or the computer program/program product is executed by at least one processor to implement the steps of the information transmission method on the access mobility management function side as provided in the embodiments of the present application, or the computer program/program product is executed by at least one processor to implement the steps of the information transmission method on the model training logic network element side as provided in the embodiments of the present application.
  • the terminal obtains information of a target network element corresponding to an ML model, and the target network element supports the function of transmitting the model on the user plane; the terminal transmits the model with the target network element using a user plane session or a tunnel, wherein the tunnel is a tunnel between the terminal and the target network element established based on the user plane session, and the model transmission includes at least one of the following: downloading information of the ML model from the target network element; uploading information of the ML model to the target network element.
  • the information of the ML model can be transmitted between the terminal and the target network element, thereby improving the performance of the transmission model of the terminal.
  • FIG1 is a block diagram of a wireless communication system to which an embodiment of the present application can be applied;
  • FIG2 is a schematic diagram of a model application provided in an embodiment of the present application.
  • FIG3 is a flow chart of an information transmission method provided in an embodiment of the present application.
  • FIG4 is a flow chart of another information transmission method provided in an embodiment of the present application.
  • FIG5 is a flow chart of another information transmission method provided in an embodiment of the present application.
  • FIG6 is a flow chart of another information transmission method provided in an embodiment of the present application.
  • FIG7 is a schematic diagram of an information transmission method provided in an embodiment of the present application.
  • FIG8 is a schematic diagram of another information transmission method provided in an embodiment of the present application.
  • FIG9 is a structural diagram of an information transmission device provided in an embodiment of the present application.
  • FIG10 is a structural diagram of another information transmission device provided in an embodiment of the present application.
  • FIG11 is a structural diagram of another information transmission device provided in an embodiment of the present application.
  • FIG12 is a structural diagram of another information transmission device provided in an embodiment of the present application.
  • FIG13 is a structural diagram of a communication device provided in an embodiment of the present application.
  • FIG14 is a structural diagram of another terminal provided in an embodiment of the present application.
  • FIG. 15 is a structural diagram of another network-side device provided in an embodiment of the present application.
  • first, second, etc. in the specification and claims of the present application are used to distinguish similar objects, and are not used to describe a specific order or sequence. It should be understood that the terms used in this way are interchangeable under appropriate circumstances, so that the embodiments of the present application can be implemented in an order other than those illustrated or described here, and the objects distinguished by “first” and “second” are generally of the same type, and the number of objects is not limited.
  • the first object can be one or more.
  • “and/or” in the specification and claims represents at least one of the connected objects, and the character “/" generally represents that the objects associated with each other are in an "or” relationship.
  • instruction in the specification and claims of this application can be either an explicit instruction or an implicit instruction.
  • An explicit instruction can be understood as the sender explicitly informing the receiver of the operation to be performed or the request result in the instruction sent; an implicit instruction can be understood as the receiver making a judgment based on the instruction sent by the sender and determining the operation to be performed or the request result based on the judgment result.
  • LTE Long Term Evolution
  • LTE-Advanced, LTE-A Long Term Evolution
  • CDMA Code Division Multiple Access
  • TDMA Time Division Multiple Access
  • FDMA Frequency Division Multiple Access
  • OFDMA Orthogonal Frequency Division Multiple Access
  • SC-FDMA Single-carrier Frequency Division Multiple Access
  • system and “network” in the embodiments of the present application are often used interchangeably, and the described technology can be used for the above-mentioned systems and radio technologies as well as for other systems and radio technologies.
  • 5G 5th Generation
  • 5G 5th Generation
  • FIG1 shows a block diagram of a wireless communication system applicable to an embodiment of the present application.
  • the wireless communication system includes a terminal 11 and a network side device 12 .
  • the terminal 11 may be a mobile phone, a tablet computer, a laptop computer or a notebook computer, a personal digital assistant (PDA), a handheld computer, a netbook, an ultra-mobile personal computer (UMPC), a mobile Internet device (MID), an augmented reality (AR)/virtual reality (VR) device, a robot, a wearable device, a vehicle user equipment (VUE), a pedestrian terminal (PUE), a smart home (a home appliance with wireless communication function, such as a refrigerator, a television, a washing machine or furniture, etc.), a game console, a personal computer (PC), a teller machine or a self-service machine and other terminal side devices, and the wearable device includes: a smart watch, a smart bracelet, a smart headset, a smart glasses, smart jewelry (smart bracelet, smart bracelet, smart ring
  • the network side device 12 may include access network equipment or core network equipment, wherein the access network equipment may also be referred to as wireless access network equipment, wireless access network (RAN), wireless access network function or wireless access network unit.
  • the access network equipment may include base stations, wireless local area networks (WLANs), wireless local area networks (WLANs), wireless access network functions, wireless access network units, ...
  • the base station may be referred to as a Node B, an evolved Node B (eNB), an access point, a base transceiver station (Base Transceiver Station, BTS), a radio base station, a radio transceiver, a basic service set (Basic Service Set, BSS), an extended service set (Extended Service Set, ESS), a home Node B, a home evolved Node B, a transmitting and receiving point (Transmitting Receiving Point, TRP) or some other suitable term in the field.
  • the base station is not limited to a specific technical vocabulary.
  • the core network equipment may include but is not limited to at least one of the following: core network node, core network function, mobility management entity (Mobility Management Entity, MME), access mobility management function (Access and Mobility Management Function, AMF), session management function (Session Management Function, SMF), user plane function (User Plane Function, UPF), policy control function (Policy Control Function, PCF), policy and charging rules function unit (Policy and Charging Rules Function, PCRF), edge application service discovery function (Edge Application Server Discovery Function, EASDF), unified data management (Unified Data Management, UDM), unified data storage (Unified Data Repository, UDR), home user server (Home Subscriber Server, HSS), centralized network configuration (CNC), network storage function (Network Repository Function, NRF), network exposure function (Network Exposure Function, NEF), local NEF (Local N
  • the overall process of communication network optimization based on ML model can be shown in Figure 2, where the model training function generates ML model based on training data, and after completing the model validity test, the model is deployed to the model reasoning function.
  • the model training function needs to obtain and analyze a large amount of data, which has high requirements on hardware performance and computing power, and is mainly deployed on network side devices, such as operator servers, or third-party servers.
  • the model inference function is based on the above ML model, takes the inference data as input, and obtains the inference output, such as the prediction of air interface indicators.
  • the model inference function has lower requirements on hardware performance and computing power than the model training function.
  • model inference is performed on the terminal side, and even part of the model training is completed on the terminal side.
  • model training is done on the communication network side (e.g., model training logical network element, or AF), and model inference is done on the terminal side.
  • the terminal can download the model from the network side;
  • local model training is completed on the terminal side, and model inference is performed on the network side or other terminal sides.
  • the terminal can pass the trained model to the network side.
  • FIG. 3 is a flow chart of an information transmission method provided in an embodiment of the present application. As shown in FIG. 3, the method includes the following steps:
  • Step 301 The terminal obtains information of a target network element corresponding to an ML model, where the target network element supports a function of a user plane transmission model.
  • the target network element may be a user-plane network element, which is a network element or functional module that stores a model or can provide a model instance.
  • the user-plane network element may be an independently deployed database network element, a data analysis repository network element (Analytics Data Repository Function, ADRF), or a model application platform, a model store, etc.; or, the user-plane network element may be a functional module, network element or device that is integrated with or plug-in to the model training logic network element.
  • ADRF Analytics Data Repository Function
  • the target network element may also be a network-side device that supports user plane transmission and control plane transmission.
  • the above-mentioned target network element can store the above-mentioned ML model.
  • the model generated by the model training logical function (MTLF) or other devices can be stored in the target network element.
  • MTLF model training logical function
  • the above-mentioned ML model can be provided by a model providing network element (also referred to as a model provider), and the above-mentioned model providing network element can be a model training function network element, a model storage function network element, an application server or an open application model (Open Application Model, OAM) and other network elements or devices that can provide ML models.
  • a model providing network element also referred to as a model provider
  • the above-mentioned model providing network element can be a model training function network element, a model storage function network element, an application server or an open application model (Open Application Model, OAM) and other network elements or devices that can provide ML models.
  • OAM Open Application Model
  • the ML model may be an AI model.
  • Step 302 The terminal performs model transmission with the target network element by using a user plane session or a tunnel, wherein the tunnel is a tunnel between the terminal and the target network element established based on the user plane session, and the model transmission includes at least one of the following:
  • the above-mentioned user plane session can be a protocol data unit (PDU) session.
  • PDU protocol data unit
  • the above tunnel may be a secure tunnel established between the terminal and the target network element based on the user session.
  • the receiving of the information of the ML model from the target network element may also be referred to as downloading the information of the ML model from the target network element.
  • the sending of the ML model information to the target network element may also be referred to as uploading or uploading the ML model information to the target network element.
  • the above steps can realize the transmission of ML model information between the terminal and the target network element by using the user plane session, thereby improving the performance of the transmission model of the terminal.
  • the terminal can perform model inference based on the information of the ML model, thereby improving the performance of the terminal's communication services or other services, such as predicting air interface indicators based on the information of the ML model, such as predicting channel information based on the information of the ML model, etc., without limitation.
  • the information of the ML model includes at least one of the following:
  • Identification information of the ML model Identification information of the ML model, function information of the ML model, type information of the ML model, requirement information of the ML model, model structure information of the ML model, and model parameter information of the ML model.
  • model identification is used to uniquely identify the requested model instance within a certain range (such as within the Public Land Mobile Network (PLMN)), that is, to identify which specific model is requested;
  • the model function and type information are used to characterize the function or purpose of the ML model. For example, it may include model type, data analysis task type (analytics ID), model functionality ID, etc.
  • the requirement information of the ML model may include at least one of the following:
  • Model transmission delay requirement information used to indicate that the model needs to support sharing
  • identity limitation information used to limit the network element identity provided by the model
  • model expression method limitation information used to limit the network element identity provided by the model
  • model performance requirement information used to limit the model needs to support sharing
  • the above-mentioned model transmission delay requirement information can instruct the other end to feedback the deadline, maximum delay and other information of the above-mentioned ML model transmission.
  • the model size requirement information may indicate storage space requirement information for the ML model transmission.
  • the sharing indication information for indicating that the model needs to support sharing may indicate that the ML model needs to support sharing between devices from different manufacturers or with different functions.
  • the identity limiting information used to limit the identity of the network element provided by the model may include manufacturer information, such as limiting the manufacturer that generates the model to one or more specific manufacturers, or limiting the network element provided by the model to one or more specific devices.
  • model expression limitation information can limit the ML model to be expressed in certain specific languages or based on certain specific AI frameworks (AI framework).
  • AI framework include Open Neural Network Exchange (ONNX), PyTorch neural network exchange (PNNX), etc.
  • AI frameworks include TensorFlow and Pytorch, etc.
  • the above-mentioned model performance requirement information may indicate the minimum and maximum requirements of the terminal for the accuracy of the model.
  • the accuracy of the model may be expressed in terms of mean absolute error (MAE), minimum mean square error (MMSE) or other forms of the model prediction results.
  • MAE mean absolute error
  • MMSE minimum mean square error
  • model usage scope requirement information can indicate the model's valid area, applicable data network name (Data Network Name, DNN), applicable slices, effective time and other scope information.
  • applicable data network name Data Network Name, DNN
  • applicable slices effective time and other scope information.
  • the information of the target network element includes at least one of the following:
  • the address information of the target network element is the address information of the target network element
  • the target network element stores a Uniform Resource Locator (URL) of the ML model
  • the target network element transmits the DNN of the ML model
  • the target network element transmits Single Network Slice Selection Assistance Information (S-NSSAI) of the ML model;
  • S-NSSAI Single Network Slice Selection Assistance Information
  • Radio Access Technology of the target network element transmitting the ML model
  • the access type of the target network element transmitting the ML model
  • the target network element transmits security information related to the user plane tunnel of the ML model.
  • the URL may indicate the location of the file where the target network element stores the ML model.
  • the above-mentioned DNN may indicate obtaining the DNN corresponding to the user plane session of the ML model.
  • the above-mentioned S-NSSAI may indicate obtaining the S-NSSAI corresponding to the user plane session of the ML model.
  • the above RAT type may indicate the RAT type corresponding to the user plane session of the model.
  • the access type may indicate the access type corresponding to the user plane session for obtaining the ML model.
  • the above security information may include security information related to the user plane tunnel such as a security certificate.
  • the information of the target network element can enable the terminal to know the destination address of the other end to which the model is transmitted and other related information, so that the terminal can subsequently complete the ML model transmission with the other end using the user plane session.
  • the target network element information may also include at least one of the following:
  • the identifier of the target network element, the temporary identifier of the terminal corresponding to the target network element, and the temporary session identifier corresponding to the target network element may be an identifier temporarily allocated by the target network element to the terminal and used to identify the terminal, and the temporary session identifier corresponding to the target network element may be an identifier temporarily allocated by the target network element to the user plane session and used to identify the session.
  • the method further includes:
  • the terminal sends a first message to an access mobility management function or a model training logic network element, where the first message is used to request to obtain the ML model, or the first message is used to request to upload the ML model.
  • the first message may be a non-access stratum (NAS) message, such as requesting to obtain the ML model through a NAS message, or requesting to upload the ML model through a NAS message.
  • NAS non-access stratum
  • the NAS message includes at least one of the following:
  • Uplink NAS transport (UL NAS transport) message.
  • the first message can be a direct message between the terminal and the model training logical network element, that is, the information transmitted between the terminal and the model training logical network element is transparent to the access mobility management function, and the access mobility management function does not perform any analysis.
  • the first message may include at least one of the following:
  • the user plane session corresponds to Internet Protocol (IP) address information of the terminal, model capability information of the terminal, identification information of the ML model, function information of the ML model, type information of the ML model, requirement information of the ML model, and first indication information, wherein the first indication information is used to indicate that the ML model needs to be transmitted using a user plane session.
  • IP Internet Protocol
  • the user plane session may be a user plane session selected by the terminal from among user plane sessions of existing data services for transmitting the ML model, or may be a user plane session newly established by the terminal for transmitting the ML model.
  • the model capability information of the terminal may include at least one of the following:
  • the terminal supports the indication information received by the model
  • the terminal supports the indication information sent by the model
  • the terminal transmits capability information of the model via the user plane
  • the model of the terminal stores spatial information.
  • the model identification information supported by the above terminal e.g., UE supported model ID
  • UE supported model ID can be used to uniquely identify the model instance within a certain range
  • the functional information of the model supported by the terminal may be a model functional identifier supported by the terminal (UE supported model type/UE supported analytics ID/supported model functionality ID), which may describe the function or purpose of the supported model, wherein the functional information of the model supported by the terminal may also be other descriptive information, such as purpose information, etc.;
  • the type information of the model supported by the terminal may indicate the model type supported by the terminal.
  • the above-mentioned indication information of the terminal supporting model reception may be the capability indication information of the terminal supporting model downloading or receiving, for example, whether the terminal supports the capability of model downloading or receiving. Further optionally, the capability indication information may also be different for different model identifiers or model types, that is, the terminal may support downloading or receiving models corresponding to some or all model identifiers or model types;
  • the above-mentioned indication information that the terminal supports model transmission may be the capability indication information that the terminal supports model upload or transmission, for example, whether the terminal supports the capability of model upload or transmission. Further optionally, the capability indication information may also be different for different model identifiers or model types, that is, the terminal may support uploading or transmitting models corresponding to some or all model identifiers or model types;
  • the capability information of the terminal transmitting the model through the user plane may be whether the terminal supports receiving or sending the ML model through the user plane session. Further optionally, the capability information may also be different for different model identifiers or model types, that is, the terminal may receive or send models corresponding to some or all model identifiers or model types through the user plane session;
  • the model storage space information of the terminal may be the initial or remaining storage space size of the model that can be stored in the terminal.
  • the identification information of the ML model may indicate the ML model that the terminal requests to obtain or the ML model instance that the terminal requests to upload.
  • the functional information of the ML model may indicate the functional information of the ML model that the terminal requests to obtain or upload.
  • the type information of the ML model may indicate the type information of the ML model that the terminal requests to obtain or upload.
  • the requirement information of the ML model may indicate the requirement information of the ML model that the terminal requests to obtain or upload, wherein the requirement information refers to the above implementation manner and is not described in detail here.
  • the first indication information is used to indicate that the ML model requested to be obtained or uploaded needs to be transmitted using a user plane session.
  • the first message includes at least one of the above items, so that the terminal can request to obtain or upload a more accurate ML model.
  • the method further includes:
  • the terminal sends a second message to the access mobility management function or the model training logical network element, where the second message is used to report or update the model capability information of the terminal.
  • the above-mentioned second message can be sent to the access mobility management function or the model training logical network element during the registration process, PDU session establishment, PDU session modification, etc., or when the network side sends a capability request (such as the subsequent third message contains the acquisition indication information of the user plane transmission model capability).
  • the above-mentioned second message can be a NAS message, or a direct message between the terminal and the model training logical network element.
  • the model capability information of the terminal may include at least one of the following:
  • the terminal supports the indication information received by the model
  • the terminal supports the indication information sent by the model
  • the terminal transmits capability information of the model via the user plane
  • the model of the terminal stores spatial information.
  • the model capability information of the terminal is reported or updated through the second message, so that the network side can perform corresponding model transmission based on the model capability information of the terminal.
  • the terminal obtains information of a target network element corresponding to the ML model, including:
  • the terminal receives a third message sent by an access mobility management function or a model training logic network element, where the third message includes information about a target network element corresponding to the ML model.
  • the above-mentioned third message can be a NAS message, or a direct message between the terminal and the model training logical network element.
  • the third message may further include at least one of the following:
  • the first indication information is used to indicate that the ML model needs to be transmitted using a user plane session
  • the acquisition indication information of the user plane transmission model capability may instruct the terminal to report its own capability of utilizing the user plane session transmission model.
  • At least one of the first indication information and the indication information for obtaining the user plane transmission model capability may instruct the terminal to transmit the ML model through the user plane.
  • the ML model may be transmitted through the user plane by protocol agreement or pre-configuration.
  • the method further includes:
  • the terminal sends a fourth message to the access mobility management function or the model training logical network element, and the fourth message includes IP address information of the terminal corresponding to the user plane session.
  • the fourth message may be sent after determining that the user plane session transmission model needs to be used, for example: After receiving the third message, it is determined that the user plane session transmission model needs to be used, and then the IP address, capability information of the user plane transmission model, etc. are reported through the fourth message.
  • the user plane session may be that after receiving the third message, the terminal selects a user plane session of an existing data service, or creates a new user plane session for transmitting the ML model, for transmitting the ML model, and sending corresponding IP address information.
  • the IP address information of the terminal corresponding to the user plane session may also be sent before receiving the information of the target network, such as the first message described in the above implementation includes the IP address information of the terminal corresponding to the user plane session.
  • the fourth message further includes:
  • Capability information of a user plane transmission model of the terminal
  • the capability information of the user plane transmission model of the above terminal may indicate that the terminal supports the user plane transmission model, and further, may support model functions, model types, etc.
  • the capability information of the user plane transmission model of the above terminal can enable the network side to adopt the user plane session transmission ML model based on the capability information.
  • the terminal uses a user plane session or a tunnel to perform model transmission with the target network element, including:
  • the terminal determines, according to the information of the target network element, a destination address of the opposite end corresponding to transmitting the ML model
  • the terminal uses the user plane session or tunnel to perform model transmission with the opposite end destination address.
  • the terminal determines, according to the information of the target network element, the destination address of the other end corresponding to the ML model to be transmitted, which may include at least one of the following:
  • the terminal determines the IP address of the target network element according to the FQDN of the target network element, and the terminal uses the IP address of the target network element as the opposite end destination address;
  • the terminal uses the address of the target network element as the opposite end destination address
  • the terminal determines the opposite-end destination address according to the URL of the target network element storing the ML model.
  • the terminal determines the opposite destination address based on the URL of the target network element storing the ML model, which may be, based on the URL of the target network element storing the ML model, determining the IP address corresponding to the location where the model is stored, and using the IP address as the opposite destination address.
  • the target network element may be a model storage network element, and the information of the target network element is the URL of the model storage network element storing the model.
  • model transmission can be implemented between the user plane session or tunnel and the above-mentioned opposite end destination address.
  • the target network element information may also be directly the opposite destination address corresponding to the ML model, that is, the terminal directly transmits the model based on the target network element information and the opposite destination address.
  • the terminal obtains information of a target network element corresponding to the ML model, and the target network element supports the function of the user plane transmission model; the terminal uses a user plane session or a tunnel to transmit the model with the target network element, wherein the tunnel is a tunnel between the terminal and the target network element established based on the user plane session, and the Model transmission includes at least one of the following: downloading the information of the ML model from the target network element; uploading the information of the ML model to the target network element. In this way, the information of the ML model can be transmitted between the terminal and the target network element, thereby improving the performance of the transmission model of the terminal.
  • FIG. 4 is a flowchart of another information transmission method provided in an embodiment of the present application. As shown in FIG. 4, the method includes the following steps:
  • Step 401 The target network element performs model transmission with the terminal by using a user plane session or a tunnel, wherein the target network element supports a function of the user plane transmission model, the tunnel is a tunnel between the target network element and the terminal established based on the user plane session, and the model transmission includes at least one of the following:
  • the above-mentioned target network element uses the user plane session or tunnel to transmit the model with the terminal. Please refer to the corresponding description of the embodiment shown in Figure 3, which will not be repeated here.
  • the method further includes:
  • the target network element obtains at least one of the following:
  • the user plane session corresponds to the IP address information of the terminal, the model capability information of the terminal, the identification information of the ML model, the function information of the ML model, the type information of the ML model, the requirement information of the ML model, and the first indication information; wherein the first indication information is used to indicate that the ML model needs to be transmitted using the user plane session.
  • the above-mentioned target network element acquisition may be receiving at least one of the above-mentioned items sent by other devices or network elements.
  • the target network element receives at least one of the following items sent by the access mobility management function or the model training logic network element:
  • the user plane session corresponds to the IP address information of the terminal, the model capability information of the terminal, the identification information of the ML model, the function information of the ML model, the type information of the ML model, the requirement information of the ML model, and the first indication information; wherein the first indication information is used to indicate that the ML model needs to be transmitted using the user plane session.
  • the at least one item may be the at least one item sent to the target network element after the access mobility management function or the model training logic network element receives the first message sent by the terminal, wherein the first message includes at least one of the following:
  • the user plane session corresponds to the IP address information of the terminal, the model capability information of the terminal, the identification information of the ML model, the function information of the ML model, the type information of the ML model, the requirement information of the ML model, and the first indication information, wherein the first indication information is used to indicate that the ML model needs to be transmitted using the user plane session.
  • the access mobility management function or the model training logic network element may send at least one of the above items via one or more messages.
  • the following information is sent through a message: the model capability information of the terminal, the identification information of the ML model, and the
  • the user plane session includes at least one of the function information of the ML model, the type information of the ML model, the requirement information of the ML model, and the first indication information, and then sending the IP address information of the terminal corresponding to the user plane session through another message.
  • At least one of the above items can enable the target network element and the terminal to perform more accurate model transmission.
  • the target network element transmits the model to the terminal by using a user plane session or a tunnel, including:
  • the target network element transmits the model with the terminal by using a user plane session or a tunnel according to the IP address information of the terminal.
  • the target network element may transmit the model with the terminal by using a user plane session or a tunnel according to the IP address information of the terminal.
  • the IP address information of the terminal may be used as the opposite destination address corresponding to the ML model, and the target network element may transmit the model with the opposite destination address by using the user plane session or the tunnel.
  • the information of the ML model includes at least one of the following:
  • Identification information of the ML model Identification information of the ML model, function information of the ML model, type information of the ML model, requirement information of the ML model, model structure information of the ML model, and model parameter information of the ML model.
  • the requirement information of the ML model includes at least one of the following:
  • Model transmission delay requirement information used to indicate that the model needs to support sharing
  • identity limitation information used to limit the network element identity provided by the model
  • model expression method limitation information used to limit the network element identity provided by the model
  • model performance requirement information used to limit the model needs to support sharing
  • this embodiment is an implementation method of the target network element corresponding to the embodiment shown in Figure 3. Its specific implementation method can refer to the relevant description of the embodiment shown in Figure 3. In order to avoid repeated description, this embodiment will not be repeated.
  • FIG. 5 is a flowchart of another information transmission method provided in an embodiment of the present application. As shown in FIG. 5, the method includes the following steps:
  • Step 501 Access the mobility management function to obtain information of a target network element corresponding to the machine learning ML model, where the target network element supports a function of a user plane transmission model;
  • Step 502 The access mobility management function sends the information of the target network element to the terminal.
  • the information of the above-mentioned target network element and the role of the above-mentioned target network element information refer to the corresponding description of the embodiment shown in Figure 3, which will not be repeated here.
  • the information of the target network element includes at least one of the following:
  • the address information of the target network element is the address information of the target network element
  • the target network element stores a uniform resource locator URL of the ML model
  • the target network element transmits the data network name DNN of the ML model
  • the target network element transmits single network slice selection auxiliary information S-NSSAI of the ML model
  • the target network element transmits the radio access technology RAT type of the ML model
  • the access type of the target network element transmitting the ML model
  • the target network element transmits security information related to the user plane tunnel of the ML model.
  • the method further includes:
  • the access mobility management function receives a first message sent by the terminal, where the first message is used to request to obtain the ML model, or the first message is used to request to upload the ML model.
  • the first message includes at least one of the following:
  • the user plane session corresponds to the IP address information of the terminal, the model capability information of the terminal, the identification information of the ML model, the function information of the ML model, the type information of the ML model, the requirement information of the ML model, and the first indication information, wherein the first indication information is used to indicate that the ML model needs to be transmitted using the user plane session.
  • the access mobility management function obtains information of a target network element corresponding to the ML model, including:
  • the access mobility management function obtains information of a target network element corresponding to the ML model according to the first message.
  • the access mobility management function may obtain the information of the target network element corresponding to the ML model according to the first message by determining the ML model according to the first message and then obtaining the information of the target network element corresponding to the ML model.
  • the access mobility management function obtains information of a target network element corresponding to the ML model according to the first message, including:
  • the access mobility management function sends a discovery message for determining a target network element corresponding to the ML model to the network storage function, where the discovery message includes at least one of the following items carried by the first message: identification information of the ML model, function information of the ML model, type information of the ML model, and indication information of a user plane transmission model;
  • the access mobility management function receives the information of the target network element sent by the network storage function.
  • the embodiments of the present application are not limited to the information of the target network element stored in the above-mentioned network.
  • the above-mentioned access mobility management function may be pre-configured with information of the target network elements corresponding to multiple ML models, so that the access mobility management function can directly determine the information of the above-mentioned target network element.
  • the method further includes:
  • the access mobility management function determines, according to the first message, a model providing network element of the ML model.
  • the above-mentioned determination of the model providing network element of the ML model may be to select the model providing network element of the ML model from the devices or network elements known to the access mobility management function, or the access mobility management function may obtain the model providing network element of the ML model from other devices or network elements.
  • the access mobility management function determines the model providing network element of the ML model according to the first message, including:
  • the access mobility management function sends a discovery message for determining a model providing network element of the ML model to the network storage function, where the discovery message includes at least one of the following items carried by the first message:
  • identification information of the ML model identification information of the ML model, function information of the ML model, type information of the ML model, and indication information of a user plane transmission model;
  • the access mobility management function receives the information of the model providing network element sent by the network storage function.
  • the model providing network element includes: a model training logic network element or a model training application function, and the method further includes:
  • the access mobility management function sends an indication message to the model training logical network element or the model training application function, where the indication information is used to indicate the ML model;
  • the access mobility management function obtains information of a target network element corresponding to the ML model, including:
  • the access mobility management function receives a response message sent by the model training logical network element or the model training application function, and the response message includes information of the target network element corresponding to the ML model.
  • the above indication information may include at least one of the following:
  • Model capability information of the terminal identification information of the ML model, function information of the ML model, type information of the ML model, requirement information of the ML model, indication information of the user plane transmission model, analytical identification (Analytic ID), and notification target address (Notification Target Address) of the ML model;
  • the analysis identifier is used to indicate that the ML model is used for the data analysis task corresponding to the analysis identifier
  • the notification target address of the ML model is empty, or the notification target address of the ML model is the IP address information of the terminal corresponding to the user plane session, or the notification target address is the address information of the access mobility management function.
  • the above analysis identifier can indicate that the requested model is used for a data analysis task corresponding to a certain analytics ID.
  • the analysis identifier is determined or mapped by AMF based on the model function type of the ML model requested by the terminal.
  • the notification target address of the above ML model may be empty because the access mobility management function does not know the IP address information of the terminal.
  • the notification target address may be set to empty or set to the address information corresponding to the access mobility management function.
  • the above response message also includes:
  • First indication information wherein the first indication information is used to indicate that the ML model needs to be transmitted using a user plane session.
  • the access mobility management function obtains information of a target network element corresponding to the ML model, including:
  • the access mobility management function obtains information of a target network element corresponding to the ML model from a network storage function.
  • the network storage function may directly obtain the information of the target network element corresponding to the ML model.
  • the method further includes:
  • the access mobility management function receives a second message sent by the terminal, where the second message is used to report or update model capability information of the terminal.
  • the model capability information of the terminal includes at least one of the following:
  • the terminal supports the indication information received by the model
  • the terminal supports the indication information sent by the model
  • the terminal transmits capability information of the model via the user plane
  • the model of the terminal stores spatial information.
  • the access mobility management function sends the information of the target network element to the terminal, including:
  • the access mobility management function sends a third message to the terminal, where the third message includes information of a target network element corresponding to the ML model, and also includes at least one of the following:
  • the first indication information is used to indicate that the ML model needs to be transmitted using a user plane session
  • the method further includes:
  • the access mobility management function receives a fourth message sent by the terminal, where the fourth message includes IP address information of the user plane session corresponding to the terminal.
  • the fourth message further includes:
  • the method further comprises:
  • the access mobility management function sends a fifth message to the model training logic network element or the target network element, and the fifth message includes at least one of the following:
  • the user plane session corresponds to the IP address information of the terminal
  • Capability information of a user plane transmission model of the terminal
  • this embodiment is an implementation method of the access mobility management function corresponding to the embodiment shown in Figure 3. Its specific implementation method can refer to the relevant description of the embodiment shown in Figure 3. In order to avoid repeated description, this embodiment will not be repeated.
  • FIG. 6 is a flowchart of another information transmission method provided in an embodiment of the present application. As shown in FIG. 6, the method includes the following steps:
  • Step 601 The model training logic network element sends information of a target network element corresponding to the ML model to the access mobility management function or the terminal, where the target network element supports the function of the user plane transmission model.
  • the information of the above-mentioned target network element and the role of the above-mentioned target network element information refer to the corresponding description of the embodiment shown in Figure 3, which will not be repeated here.
  • the information of the target network element includes at least one of the following:
  • the address information of the target network element is the address information of the target network element
  • the target network element stores a uniform resource locator URL of the ML model
  • the target network element transmits the data network name DNN of the ML model
  • the target network element transmits single network slice selection auxiliary information S-NSSAI of the ML model
  • the target network element transmits the radio access technology RAT type of the ML model
  • the access type of the target network element transmitting the ML model
  • the target network element transmits security information related to the user plane tunnel of the ML model.
  • the method further includes:
  • the model training logic network element receives a first message from the terminal, where the first message is used to request to obtain the ML model, or the first message is used to trigger uploading of the ML model.
  • the above-mentioned first message can be sent directly by the terminal to the model training logical network element, or it can be transferred by accessing the mobile management function, and the message may be changed during the transfer.
  • the first message includes at least one of the following:
  • the user plane session corresponds to the IP address information of the terminal, the model capability information of the terminal, the identification information of the ML model, the function information of the ML model, the type information of the ML model, the requirement information of the ML model, and the first indication information, wherein the first indication information is used to indicate that the ML model needs to be transmitted using the user plane session.
  • the method further includes:
  • the model training logic network element receives an indication message sent by the access mobility management function, where the indication information is used to indicate the ML model;
  • the model training logic network element sends information of a target network element corresponding to the ML model to the access mobility management function, including:
  • the model training logic network element sends a response message to the access mobility management function, and the response message includes information of a target network element corresponding to the ML model.
  • the indication information includes at least one of the following:
  • Model capability information of the terminal identification information of the ML model, function information of the ML model, type information of the ML model, requirement information of the ML model, indication information of the user plane transmission model, analysis identifier, and notification target address of the ML model;
  • the analysis identifier is used to indicate that the ML model is used for the data analysis task corresponding to the analysis identifier
  • the notification target address of the ML model is empty, or the notification target address of the ML model is the IP address information of the terminal corresponding to the user plane session, or the notification target address is the address information of the access mobility management function.
  • the method further includes:
  • the model training logic network element determines the information of the target network element according to the first message or the indication information.
  • determining the information of the target network element according to the first message or the indication information can be by selecting the above-mentioned target network element from the devices or network elements known to the model training logical network element, or the model training logical network element obtains the information of the above-mentioned target network element from other devices or network elements.
  • the model training logic network element determines the information of the target network element according to the first message or the indication information, including:
  • the model training logic network element sends a discovery message for determining a target network element corresponding to the ML model to the network storage function, wherein the discovery message includes at least one of the following: identification information of the ML model, function information of the ML model, type information of the ML model, and indication information of a user plane transmission model;
  • the model training logic network element receives the information of the target network element sent by the network storage function.
  • the model training logic network element determines the information of the target network element according to the first message or the indication information, including:
  • the model training logic network element sends an information request message to the model storage network element, where the information request message is used to request the model storage network element to feedback information of the target network element corresponding to the ML model;
  • the model training logic network element receives information of a target network element corresponding to the ML model sent by the model storage network element, where the information of the target network element includes a URL where the model storage network element stores the ML model.
  • the model storage network element may be the target network element, and the information of the target network element may be the URL of the target network element storage model.
  • the response message also includes:
  • First indication information wherein the first indication information is used to indicate that the ML model needs to be transmitted using a user plane session.
  • the method further includes at least one of the following:
  • the model training logic network element determines to transmit the ML model in a user plane manner
  • the model training logic network element obtains the information of the target network element
  • the model training logic network element receives a fifth message sent by the access mobility management function, where the fifth message includes at least one of the following:
  • the IP address information of the terminal corresponding to the user plane session
  • Capability information of a user plane transmission model of the terminal
  • the above-mentioned model training logic network element can determine whether to use the user plane mode to transmit the above-mentioned ML model based on at least one factor such as local strategy, control plane load, user plane load, terminal capability information of transmitting the model through the user plane, size requirement of the requested model, and transmission delay requirement.
  • model training logic network element obtains the information of the target network element, which may be that the model training logic network element directly obtains the information of the target network element.
  • this embodiment is an implementation of the network side device corresponding to the embodiment shown in FIG. 3.
  • the relevant description of the embodiment shown in FIG. 3 please refer to the relevant description of the embodiment shown in FIG. 3 to avoid repeated description. No longer.
  • Embodiment 1 is a diagrammatic representation of Embodiment 1:
  • This embodiment is shown in FIG7 and includes the following steps:
  • Step 1 the UE initiates a NAS message to the AMF, which is used to report/update its own ML model capability information.
  • the ML model capability information of the UE may include:
  • Model identification information supported by UE e.g. UE supported model ID
  • UE supported model ID which is used to uniquely identify a model instance within a certain range
  • Model function types or other descriptive information supported by the UE for example, UE supported model type/supported analytics ID/supported model functionality ID, which is used to describe the functions or uses of the supported models;
  • Capability indication information of the UE supporting the model receiving and/or sending for example, whether the UE supports the capability of the model receiving and/or sending.
  • the capability indication information may also be different for different model IDs or model types;
  • the capability information of the UE to transmit (receive/send) the model via the user plane for example, whether the UE supports sending or receiving the ML model via the user plane PDU session.
  • the capability information may also be different for different model IDs or model types.
  • Model storage space information of the UE for example, the initial or remaining storage space size that can store models in the UE.
  • Step 1 UE initiates a NAS message to AMF, which is used to request the ML model.
  • the NAS message of the model acquisition request includes at least one of the following:
  • the model ID of the ML model requested by the UE is the model ID of the ML model requested by the UE.
  • the model capability type of the ML model requested by the UE is the model capability type of the ML model requested by the UE.
  • the indication information of the user plane transmission model is used to indicate to the network that the requested model needs to be transmitted using the user plane PDU session.
  • the requirement information of the ML model includes at least one of the following requirements:
  • Transmission delay requirement information used to indicate the deadline, maximum delay, and other information of the network feedback model
  • Model size requirement information used to indicate the storage space requirement of the model fed back by the network
  • Sharable indication information is used to indicate that the model needs to support sharing between different manufacturers or different devices
  • Manufacturer or provider identity qualification information used to limit the manufacturer that generates the model to one or more specific manufacturers, or limit the network elements provided by the model to one or more specific devices;
  • Model expression limitation information is used to limit the model to be expressed in certain specific languages or based on certain specific AI frameworks.
  • commonly used model languages include ONNX, PNNX, etc.
  • AI frameworks include TensorFlow, Pytorch, etc.;
  • Model performance requirement information used to indicate the minimum and maximum requirements of the UE for the accuracy of the model
  • the model uses scope-limited information to indicate the model's valid area, applicable DNN, applicable slice, valid time, and other scope information.
  • Step 2 Optionally, AMF selects a model providing network element based on the model request message sent by the UE.
  • the model providing network elements in the network can specifically be model training functional network elements, or model storage functional network elements, application servers, OAM and other network elements or devices that can provide AI/ML models.
  • the AMF discovers and selects the model providing network element from the NRF based on at least one of the following information, and the model providing network element matches the following information.
  • the AMF may provide at least one of the following information to the NRF:
  • Model ID of the ML model requested by the UE Used to indicate that the selected model providing network element can provide the ML model corresponding to the model ID;
  • the model function type of the ML model requested by the UE which is used for the selected model providing network element to provide the ML model corresponding to the model function type;
  • the indication information of the user plane transmission model is used to indicate to the network that the requested model needs to be transmitted using the user plane PDU session, and the selected model provides that the network element can support the transmission model using the user plane PDU session.
  • Step 3 Assuming that the model providing network element in step 2 is MTLF, AMF sends a first request message to MTLF to indicate to MTLF the model requested by the UE.
  • the content carried in the first request message refers to the NAS message content in step 1.
  • the first request message may also carry other information related to model transmission, such as at least one of the following:
  • Analytic ID used to indicate that the requested model is for a data analysis task corresponding to a certain analytics ID.
  • the analytics ID is determined or mapped by the AMF based on the model function type of the ML model requested by the UE.
  • Notification target address (A Notification Target Address):
  • the AMF since the AMF does not yet know the IP address information of the UE, it is optional to set the Notification Target Address to empty or set it to the address information corresponding to the AMF.
  • Step 4 MTLF returns a first response message to AMF, which includes information about the user plane network element corresponding to the requested model.
  • the user plane network element refers to a network element or functional module that stores models or can provide model instances.
  • the user plane network element supports the function of transmitting ML models between terminals through user plane sessions.
  • the user plane network element can be an independently deployed database network element (such as ADRF, or model application platform, model store, etc.), and the models generated by MTLF or other devices can be stored in the database.
  • the user plane network element can be a functional module that is co-located with MTLF or plug-in.
  • the information of the user plane network element includes at least one of the following information of the user plane network element:
  • the FQDN of the user plane network element is the FQDN of the user plane network element
  • IP address IP address, Media Access Control (MAC) address, etc.
  • URL for example, used to indicate the location where the user plane network element stores the model file
  • DNN used to indicate the DNN corresponding to the user plane session of the model
  • S-NSSAI used to indicate the S-NSSAI corresponding to the user plane session of the model
  • RAT Type used to indicate the RAT Type corresponding to the user plane session of the model
  • Access Type used to indicate the Access Type corresponding to the user plane session for obtaining the model
  • Security information related to the user plane tunnel including security certificates, etc.
  • the first response message may include indication information of the user plane transmission model, which is used to indicate the model requested for transmission using the user plane PDU session.
  • the MTLF before sending the first response message, the MTLF also determines to use the user plane mode to transmit the model based on at least one factor such as local policy, control plane load, user plane load, UE's ability to transmit (receive/send) the model through the user plane, size requirements of the requested model, and transmission delay requirements.
  • at least one factor such as local policy, control plane load, user plane load, UE's ability to transmit (receive/send) the model through the user plane, size requirements of the requested model, and transmission delay requirements.
  • Step 3a Optionally, before sending the first response message, if the requested model is stored in the user plane network element, the MTLF requests the user plane network element for model storage information to first determine the information of the user plane network element corresponding to the model.
  • the user plane network element storing the ML model here is ADRF
  • MTLF sends a request message to ADRF to obtain the storage address information of the model file.
  • ADRF feeds back the URL where the model file is stored to MTLF, and the URL corresponds to the information of the user plane network element.
  • Step 5 AMF sends a NAS message to the UE based on the first response message, which includes information about the user plane network element corresponding to the model.
  • the NAS message also includes at least one of the following information:
  • the user plane transmission model indication information is used to indicate the model requested by the user plane PDU session transmission
  • the UE user plane transmission model capability acquisition indication information is used to instruct the UE to report its own capability of utilizing the user plane PDU session transmission model.
  • Step 6 The UE selects an existing PDU session, or creates a new PDU session, for transmitting the model.
  • the UE may first determine its own capability to support the user plane transmission model.
  • Step 7 The UE sends a NAS message to the AMF, which includes the UE IP address information corresponding to the PDU session selected in step 6.
  • the NAS message also includes user plane transmission model capability information of the UE.
  • Step 8 AMF sends the UE IP address information and, optionally, the UE’s user plane transport model capability information to MTLF.
  • AMF can set the A Notification Target Address in step 3 to the UE IP address information.
  • the MTLF forwards the UE IP address information and, optionally, the UE's user plane transmission model capability information to the user plane network element (ADRF).
  • ADRF user plane network element
  • Step 9 Optionally, based on the PDU session, a secure tunnel is established between the UE and the user plane network element.
  • Step 10 Using the PDU session or the secure tunnel, the UE and the user plane network element interact with the model-related signaling and/or transmission model.
  • the interaction-related signaling refers to at least one of the following information between the UE and the user plane network element using the PDU session interaction model:
  • the transmission model refers to the use of the PDU session UE to download or upload model information from ADRF or to ADRF, which includes model structure information and/or model parameter information.
  • Embodiment 2 is a diagrammatic representation of Embodiment 1:
  • the basic process of this embodiment is similar to that of the first embodiment, except that before requesting the model, the UE has determined that it has the capability of the user plane transmission model and the PDU session of the transmission model.
  • the UE can directly report its own IP address information and, optionally, the capability information of the UE user plane transmission model. As shown in FIG8 , the process includes the following steps:
  • Step 0 The UE selects an existing PDU session, or creates a new PDU session, for transmitting the model.
  • the UE can first determine its ability to support the user plane transmission model
  • Step 1 UE initiates a NAS message to AMF, which is used to request the ML model.
  • the NAS message here also includes the UE IP address information corresponding to the PDU session in step 0.
  • Step 2 Same as step 2 of Example 1.
  • Step 3 is the same as step 3 in embodiment 1, except that the first request message also includes the UE IP address information.
  • AMF may set the Notification Target Address in the first request message to the UE IP address.
  • Step 3a is the same as step 3a in Example 1, except that the request message sent by MTLF to ADRF also includes the UE IP address information.
  • Step 4 Same as step 4 of embodiment 1.
  • Step 5 Same as step 5 of embodiment 1.
  • a secure tunnel is established between the UE and the user plane network element based on the PDU session.
  • the UE can obtain the download address or location information of the AI model in the network, and dynamically download the model from the network by means of a user plane PDU session.
  • the network can also obtain the IP address information of the UE and dynamically upload the model from the UE using the user plane PDU session. type.
  • FIG. 9 is a structural diagram of an information transmission device provided in an embodiment of the present application.
  • the information transmission device 900 includes:
  • An acquisition module 901 is used to acquire information of a target network element corresponding to a machine learning ML model, where the target network element supports a function of a user plane transmission model;
  • the transmission module 902 is configured to perform model transmission with the target network element by using a user plane session or a tunnel, wherein the tunnel is a tunnel between the terminal and the target network element established based on the user plane session, and the model transmission includes at least one of the following:
  • the information of the ML model includes at least one of the following:
  • Identification information of the ML model Identification information of the ML model, function information of the ML model, type information of the ML model, requirement information of the ML model, model structure information of the ML model, and model parameter information of the ML model.
  • the requirement information of the ML model includes at least one of the following:
  • Model transmission delay requirement information used to indicate that the model needs to support sharing
  • identity limitation information used to limit the network element identity provided by the model
  • model expression method limitation information used to limit the network element identity provided by the model
  • model performance requirement information used to limit the model needs to support sharing
  • the information of the target network element includes at least one of the following:
  • the address information of the target network element is the address information of the target network element
  • the target network element stores a uniform resource locator URL of the ML model
  • the target network element transmits the data network name DNN of the ML model
  • the target network element transmits single network slice selection auxiliary information S-NSSAI of the ML model
  • the target network element transmits the radio access technology RAT type of the ML model
  • the access type of the target network element transmitting the ML model
  • the target network element transmits security information related to the user plane tunnel of the ML model.
  • the device further comprises:
  • the first sending module is used to send a first message to an access mobility management function or a model training logic network element, where the first message is used to request to obtain the ML model, or the first message is used to request to upload the ML model.
  • the first message includes at least one of the following:
  • the user plane session corresponds to the IP address information of the terminal, the model capability information of the terminal, the identification information of the ML model, the function information of the ML model, the type information of the ML model, the requirement information of the ML model, and the first indication information, wherein the first indication information is used to indicate that the ML model needs to be transmitted using the user plane session.
  • the device further comprises:
  • the second sending module is used to send a second message to the access mobility management function or the model training logic network element, and the second message is used to report or update the model capability information of the terminal.
  • the model capability information of the terminal includes at least one of the following:
  • the terminal supports the indication information received by the model
  • the terminal supports the indication information sent by the model
  • the terminal transmits capability information of the model via the user plane
  • the model of the terminal stores spatial information.
  • the acquisition module 901 is used to:
  • the third message further includes at least one of the following:
  • the first indication information is used to indicate that the ML model needs to be transmitted using a user plane session
  • the device further comprises:
  • the third sending module is used to send a fourth message to the access mobility management function or the model training logic network element, and the fourth message includes the IP address information of the terminal corresponding to the user plane session.
  • the fourth message further includes:
  • Capability information of a user plane transmission model of the terminal
  • the transmission module 902 is used to:
  • the model is transmitted between the user plane session or tunnel and the opposite end destination address.
  • determining, according to the information of the target network element, a destination address of a peer end corresponding to transmitting the ML model includes at least one of the following:
  • the opposite end destination address is determined according to the URL of the target network element storing the ML model.
  • the above-mentioned information transmission device can improve the performance of the transmission model of the terminal.
  • the information transmission device in the embodiment of the present application may be an electronic device, such as an electronic device with an operating system, or a component in an electronic device, such as an integrated circuit or a chip.
  • the electronic device may be a terminal, or may be other devices other than a terminal.
  • the terminal may include but is not limited to the terminals listed in the embodiment of the present application.
  • the other devices may be servers, network attached storage (NAS), etc., which are not specifically limited in the embodiments of the present application.
  • the information transmission device provided in the embodiment of the present application can implement each process implemented by the method embodiment shown in Figure 3 and achieve the same technical effect. To avoid repetition, it will not be repeated here.
  • FIG. 10 is a structural diagram of an information transmission device provided in an embodiment of the present application.
  • the information transmission device 1000 includes:
  • the transmission module 1001 is used to perform model transmission between the terminal and the target network element corresponding to the device by using a user plane session or a tunnel, wherein the target network element corresponding to the device supports the function of the user plane transmission model, the tunnel is a tunnel between the target network element and the terminal established based on the user plane session, and the model transmission includes at least one of the following:
  • the target network element corresponding to the above-mentioned device may be the target network element to which the above-mentioned device belongs, or the above-mentioned device is the target network element.
  • the device further comprises:
  • the acquisition module is used to obtain at least one of the following:
  • the user acquisition module surface session corresponds to the IP address information of the terminal, the model capability information of the terminal, the identification information of the ML model, the function information of the ML model, the type information of the ML model, the requirement information of the ML model, and the first indication information; wherein the first indication information is used to indicate that the ML model needs to be transmitted using a user surface session.
  • the acquisition module is used to receive at least one of the following items sent by the access mobility management function or the model training logic network element:
  • the user plane session corresponds to the IP address information of the terminal, the model capability information of the terminal, the identification information of the ML model, the function information of the ML model, the type information of the ML model, the requirement information of the ML model, and the first indication information; wherein the first indication information is used to indicate that the ML model needs to be transmitted using the user plane session.
  • the transmission module 1001 is used for:
  • the model is transmitted between the terminal and the user plane session or tunnel.
  • the information of the ML model includes at least one of the following:
  • Identification information of the ML model Identification information of the ML model, function information of the ML model, type information of the ML model, requirement information of the ML model, model structure information of the ML model, and model parameter information of the ML model.
  • the requirement information of the ML model includes at least one of the following:
  • Model transmission delay requirement information used to indicate that the model needs to support sharing
  • identity limitation information used to limit the network element identity provided by the model
  • model expression method limitation information used to limit the network element identity provided by the model
  • model performance requirement information used to limit the model needs to support sharing
  • the above-mentioned information transmission device can improve the performance of the transmission model of the terminal.
  • the information transmission device in the embodiment of the present application may be an electronic device, such as an electronic device with an operating system, or a component in the electronic device, such as an integrated circuit or a chip.
  • the electronic device may be a terminal or a network side device.
  • the information transmission device provided in the embodiment of the present application can implement each process implemented by the method embodiment shown in Figure 4 and achieve the same technical effect. To avoid repetition, it will not be repeated here.
  • FIG. 11 is a structural diagram of an information transmission device provided in an embodiment of the present application.
  • the information transmission device 1100 includes:
  • An acquisition module 1101 is used to acquire information of a target network element corresponding to a machine learning ML model, where the target network element supports a function of a user plane transmission model;
  • the first sending module 1102 is configured to send the information of the target network element to the terminal.
  • the information of the target network element includes at least one of the following:
  • the address information of the target network element is the address information of the target network element
  • the target network element stores a uniform resource locator URL of the ML model
  • the target network element transmits the data network name DNN of the ML model
  • the target network element transmits single network slice selection auxiliary information S-NSSAI of the ML model
  • the target network element transmits the radio access technology RAT type of the ML model
  • the access type of the target network element transmitting the ML model
  • the target network element transmits security information related to the user plane tunnel of the ML model.
  • the device further comprises:
  • the first receiving module is used to receive a first message sent by the terminal, where the first message is used to request to obtain the ML model, or the first message is used to request to upload the ML model.
  • the first message includes at least one of the following:
  • the user plane session corresponds to the IP address information of the terminal, the model capability information of the terminal, the identification information of the ML model, the function information of the ML model, the type information of the ML model, the requirement information of the ML model, and the first indication information, wherein the first indication information is used to indicate that the ML model needs to be transmitted using the user plane session.
  • the acquisition module 1101 is used to:
  • the information of the target network element corresponding to the ML model can be obtained according to the first message.
  • the acquisition module 1101 is used to:
  • the discovery message includes at least one of the following items carried by the first message: identification information of the ML model, function information of the ML model, type information of the ML model, and indication information of a user plane transmission model;
  • the device further comprises:
  • a determination module is used to determine a model providing network element of the ML model according to the first message.
  • the determining module is used to:
  • the discovery message includes at least one of the following items carried by the first message:
  • identification information of the ML model identification information of the ML model, function information of the ML model, type information of the ML model, and indication information of a user plane transmission model;
  • the model providing network element includes: a model training logic network element or a model training application function, and the device further includes:
  • a second sending module is used to send an indication message to the model training logic network element or the model training application function, where the indication information is used to indicate the ML model;
  • the acquisition module 1101 is used for:
  • the indication information includes at least one of the following:
  • Model capability information of the terminal identification information of the ML model, function information of the ML model, type information of the ML model, requirement information of the ML model, indication information of the user plane transmission model, analysis identifier, and notification target address of the ML model;
  • the analysis identifier is used to indicate that the ML model is used for the data analysis task corresponding to the analysis identifier
  • the notification target address of the ML model is empty, or the notification target address of the ML model is the IP address information of the terminal corresponding to the user plane session, or the notification target address is the address information of the access mobility management function.
  • the response message also includes:
  • First indication information wherein the first indication information is used to indicate that the ML model needs to be transmitted using a user plane session.
  • the acquisition module 1101 is used to:
  • the information of the target network element corresponding to the ML model is obtained from the network storage function.
  • the device further comprises:
  • the second receiving module is used to receive a second message sent by the terminal, where the second message is used to report or update the model capability information of the terminal.
  • the model capability information of the terminal includes at least one of the following:
  • the terminal supports the indication information received by the model
  • the terminal supports the indication information sent by the model
  • the terminal transmits capability information of the model via the user plane
  • the model of the terminal stores spatial information.
  • the first sending module 1102 is used to:
  • the third message including information of a target network element corresponding to the ML model, and also including at least one of the following:
  • the first indication information is used to indicate that the ML model needs to be transmitted using a user plane session
  • the device further comprises:
  • the third receiving module is used to receive a fourth message sent by the terminal, where the fourth message includes IP address information of the terminal corresponding to the user plane session.
  • the fourth message further includes:
  • the device also includes:
  • the third sending module is configured to send a fifth message to the model training logic network element or the target network element, where the fifth message includes at least one of the following:
  • the user plane session corresponds to the IP address information of the terminal
  • Capability information of a user plane transmission model of the terminal
  • the above-mentioned information transmission device can improve the performance of the transmission model of the terminal.
  • the information transmission device in the embodiment of the present application may be an electronic device, such as an electronic device with an operating system, or a component in the electronic device, such as an integrated circuit or a chip.
  • the electronic device may be a terminal or a network side device.
  • the information transmission device provided in the embodiment of the present application can implement each process implemented by the method embodiment shown in Figure 5 and achieve the same technical effect. To avoid repetition, it will not be repeated here.
  • FIG. 12 is a structural diagram of an information transmission device provided in an embodiment of the present application.
  • the information transmission device 1200 includes:
  • the sending module 1201 is used to send information of a target network element corresponding to the ML model to the access mobility management function or the terminal, where the target network element supports the function of the user plane transmission model.
  • the information of the target network element includes at least one of the following:
  • the address information of the target network element is the address information of the target network element
  • the target network element stores a uniform resource locator URL of the ML model
  • the target network element transmits the data network name DNN of the ML model
  • the target network element transmits single network slice selection auxiliary information S-NSSAI of the ML model
  • the target network element transmits the radio access technology RAT type of the ML model
  • the access type of the target network element transmitting the ML model
  • the target network element transmits security information related to the user plane tunnel of the ML model.
  • the device further comprises:
  • the first receiving module is used to receive a first message from the terminal, where the first message is used to request to obtain the ML model, or the first message is used to trigger uploading of the ML model.
  • the first message includes at least one of the following:
  • the user plane session corresponds to the IP address information of the terminal, the model capability information of the terminal, the identification information of the ML model, the function information of the ML model, the type information of the ML model, the requirement information of the ML model, and the first indication information, wherein the first indication information is used to indicate that the ML model needs to be transmitted using the user plane session.
  • the device further comprises:
  • a second receiving module configured to receive an indication message sent by the access mobility management function, wherein the indication information is used to indicate the ML model;
  • the sending module 1201 is used for:
  • a response message is sent to the access mobility management function, where the response message includes information of a target network element corresponding to the ML model.
  • the indication information includes at least one of the following:
  • Model capability information of the terminal identification information of the ML model, function information of the ML model, type information of the ML model, requirement information of the ML model, indication information of the user plane transmission model, analysis identifier, and notification target address of the ML model;
  • the analysis identifier is used to indicate that the ML model is used for the data analysis task corresponding to the analysis identifier
  • the notification target address of the ML model is empty, or the notification target address of the ML model is the IP address information of the terminal corresponding to the user plane session, or the notification target address is the address information of the access mobility management function.
  • the device further comprises:
  • the first determining module is used to determine the information of the target network element according to the first message or the indication information.
  • the first determining module is used to:
  • the discovery message including at least one of the following: identification information of the ML model, function information of the ML model, type information of the ML model, and indication information of a user plane transmission model;
  • the first determining module is used to:
  • the response message also includes:
  • First indication information wherein the first indication information is used to indicate that the ML model needs to be transmitted using a user plane session.
  • the device further comprises at least one of the following:
  • a second determining module is used to determine whether to transmit the ML model in a user plane manner
  • An acquisition module used to acquire information of the target network element
  • the third receiving module is configured to receive a fifth message sent by the access mobility management function, where the fifth message includes at least one of the following:
  • the IP address information of the terminal corresponding to the user plane session
  • Capability information of a user plane transmission model of the terminal
  • the above-mentioned information transmission device can improve the performance of the transmission model of the terminal.
  • the information transmission device in the embodiment of the present application may be an electronic device, such as an electronic device with an operating system, or a component in the electronic device, such as an integrated circuit or a chip.
  • the electronic device may be a terminal or a network side device.
  • the information transmission device provided in the embodiment of the present application can implement each process implemented by the method embodiment shown in Figure 6 and achieve the same technical effect. To avoid repetition, it will not be repeated here.
  • an embodiment of the present application further provides a communication device 1300, including a processor 1301 and a memory 1302, wherein the memory 1302 stores a program or instruction that can be run on the processor 1301.
  • the communication device 1300 is a terminal
  • the program or instruction is executed by the processor 1301 to implement the various steps of the resource allocation method embodiment on the terminal side, and the same technical effect can be achieved.
  • the communication device 1300 is a network side device
  • the program or instruction is executed by the processor 1301 to implement the various steps of the resource allocation method embodiment on the network side, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.
  • the embodiment of the present application also provides a communication device, including a processor and a communication interface, wherein the communication interface is used to obtain information of a target network element corresponding to a machine learning ML model, and the target network element supports the function of a user plane transmission model; a transmission module is used to transmit the model to the target network element using a user plane session or tunnel, wherein the tunnel is a tunnel between a terminal and the target network element established based on a user plane session, and the model transmission includes at least one of the following: receiving information of the ML model from the target network element; sending information of the ML model to the target network element.
  • This communication device embodiment corresponds to the above-mentioned information transmission method embodiment, and each implementation process and implementation method of the above-mentioned method embodiment can be applied to this communication device embodiment, and can achieve the same technical effect.
  • FIG14 is a schematic diagram of the hardware structure of a terminal implementing an embodiment of the present application.
  • the terminal 1400 includes but is not limited to: a radio frequency unit 1401, a network module 1402, an audio output unit 1403, an input unit 1404, a sensor 1405, a display unit 1406, a user input unit 1407, an interface unit 1408, a memory 1409 and at least some of the components of the processor 1410.
  • the terminal 1400 may also include a power source (such as a battery) for supplying power to various components, and the power source may be logically connected to the processor 1410 through a power management system, so that the power management system can manage charging, discharging, and power consumption.
  • a power source such as a battery
  • the present invention may include more or fewer components than those shown in the figure, or some components may be combined, or the components may be arranged differently, which will not be described in detail here.
  • the input unit 1404 may include a graphics processing unit (GPU) 14041 and a microphone 14042, and the graphics processing unit 14041 processes the image data of the static picture or video obtained by the image capture device (such as a camera) in the video capture mode or the image capture mode.
  • the display unit 1406 may include a display panel 14061, and the display panel 14061 may be configured in the form of a liquid crystal display, an organic light emitting diode, etc.
  • the user input unit 1407 includes a touch panel 14071 and at least one of other input devices 14072.
  • the touch panel 14071 is also called a touch screen.
  • the touch panel 14071 may include two parts: a touch detection device and a touch controller.
  • Other input devices 14072 may include, but are not limited to, a physical keyboard, function keys (such as a volume control key, a switch key, etc.), a trackball, a mouse, and a joystick, which will not be repeated here.
  • the radio frequency unit 1401 can transmit the data to the processor 1410 for processing; in addition, the radio frequency unit 1401 can send uplink data to the network side device.
  • the radio frequency unit 1401 includes but is not limited to an antenna, an amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, etc.
  • the memory 1409 can be used to store software programs or instructions and various data.
  • the memory 1409 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first storage area may store an operating system, an application program or instruction required for at least one function (such as a sound playback function, an image playback function, etc.), etc.
  • the memory 1409 may include a volatile memory or a non-volatile memory, or the memory 1409 may include both volatile and non-volatile memories.
  • the non-volatile memory may be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), or a flash memory.
  • the volatile memory may be a random access memory (RAM), a static random access memory (SRAM), a dynamic random access memory (DRAM), a synchronous dynamic random access memory (SDRAM), a double data rate synchronous dynamic random access memory (DDRSDRAM), an enhanced synchronous dynamic random access memory (ESDRAM), a synchronous link dynamic random access memory (SLDRAM) and a direct memory bus random access memory (DRRAM).
  • the memory 1409 in the embodiment of the present application includes but is not limited to these and any other suitable types of memory.
  • the processor 1410 may include one or more processing units; optionally, the processor 1410 integrates an application processor and a modem processor, wherein the application processor mainly processes operations related to an operating system, a user interface, and application programs, and the modem processor mainly processes wireless communication signals, such as a baseband processor. It is understandable that the modem processor may not be integrated into the processor 1410.
  • the radio frequency unit 1401 is used for:
  • the model is transmitted between the target network element and the user plane session or tunnel, wherein the tunnel is based on The tunnel between the terminal and the target network element established by the user plane session, wherein the model transmission includes at least one of the following:
  • the information of the ML model includes at least one of the following:
  • Identification information of the ML model Identification information of the ML model, function information of the ML model, type information of the ML model, requirement information of the ML model, model structure information of the ML model, and model parameter information of the ML model.
  • the requirement information of the ML model includes at least one of the following:
  • Model transmission delay requirement information used to indicate that the model needs to support sharing
  • identity limitation information used to limit the network element identity provided by the model
  • model expression method limitation information used to limit the network element identity provided by the model
  • model performance requirement information used to limit the model needs to support sharing
  • the information of the target network element includes at least one of the following:
  • the address information of the target network element is the address information of the target network element
  • the target network element stores a uniform resource locator URL of the ML model
  • the target network element transmits the data network name DNN of the ML model
  • the target network element transmits single network slice selection auxiliary information S-NSSAI of the ML model
  • the target network element transmits the radio access technology RAT type of the ML model
  • the access type of the target network element transmitting the ML model
  • the target network element transmits security information related to the user plane tunnel of the ML model.
  • the radio frequency unit 1401 is further configured to:
  • the first message includes at least one of the following:
  • the user plane session corresponds to the IP address information of the terminal, the model capability information of the terminal, the identification information of the ML model, the function information of the ML model, the type information of the ML model, the requirement information of the ML model, and the first indication information, wherein the first indication information is used to indicate that the ML model needs to be transmitted using the user plane session.
  • the radio frequency unit 1401 is further configured to:
  • a second message is sent to the access mobility management function or the model training logic network element, where the second message is used to report or update the model capability information of the terminal.
  • the model capability information of the terminal includes at least one of the following:
  • the terminal supports the indication information received by the model
  • the terminal supports the indication information sent by the model
  • the terminal transmits capability information of the model via the user plane
  • the model of the terminal stores spatial information.
  • the obtaining information of a target network element corresponding to the machine learning ML model includes:
  • the third message further includes at least one of the following:
  • the first indication information is used to indicate that the ML model needs to be transmitted using a user plane session
  • the radio frequency unit 1401 is further configured to:
  • a fourth message is sent to the access mobility management function or the model training logic network element, where the fourth message includes IP address information of the terminal corresponding to the user plane session.
  • the fourth message further includes:
  • Capability information of a user plane transmission model of the terminal
  • the using a user plane session or tunnel to transmit the model to the target network element includes:
  • the model is transmitted between the user plane session or tunnel and the opposite end destination address.
  • determining, according to the information of the target network element, a destination address of a peer end corresponding to transmitting the ML model includes at least one of the following:
  • the opposite end destination address is determined according to the URL of the target network element storing the ML model.
  • the above terminal can improve the data transmission performance of the terminal.
  • the embodiment of the present application also provides a communication device, including a processor and a communication interface, wherein the communication interface is used to perform model transmission with a terminal using a user plane session or a tunnel, wherein the target network element supports the function of the user plane transmission model, and the tunnel is a tunnel between the target network element and the terminal established based on the user plane session, and the model transmission includes at least one of the following: sending information of the ML model to the terminal; receiving information of the ML model sent by the terminal.
  • the communication interface is used to obtain information of the target network element corresponding to the machine learning ML model, and the target network element supports the function of the user plane transmission model; and sending information of the target network element to the terminal.
  • the communication interface is used to send information of the target network element corresponding to the ML model to the access mobility management function or the terminal, and the target network element supports the function of the user plane transmission model.
  • This communication device embodiment corresponds to the above-mentioned information transmission method embodiment, and each implementation process and implementation method of the above-mentioned method embodiment can be applied to the communication device embodiment and can achieve the same technical effect.
  • the embodiment of the present application further provides a network side device.
  • the network side device 1500 includes: a processor 1501, a network interface 1502, and a memory 1503.
  • the network interface 1502 is, for example, a common public radio interface (CPRI).
  • CPRI common public radio interface
  • the network side device 1500 of an embodiment of the present invention also includes: instructions or programs stored in the memory 1503 and executable on the processor 1501.
  • the processor 1501 calls the instructions or programs in the memory 1503 to execute the method executed by each module shown in any one of Figures 10 to 12, and achieves the same technical effect. To avoid repetition, it will not be repeated here.
  • the network interface 1502 is used to perform model transmission with the terminal using a user plane session or a tunnel, wherein the target network element supports the function of the user plane transmission model, the tunnel is a tunnel between the target network element and the terminal established based on the user plane session, and the model transmission includes at least one of the following:
  • the network interface 1502 is further used for:
  • the user plane session corresponds to the IP address information of the terminal, the model capability information of the terminal, the identification information of the ML model, the function information of the ML model, the type information of the ML model, the requirement information of the ML model, and the first indication information; wherein the first indication information is used to indicate that the ML model needs to be transmitted using the user plane session.
  • the network interface 1502 is used to receive at least one of the following items sent by the access mobility management function or the model training logic network element:
  • the user plane session corresponds to the IP address information of the terminal, the model capability information of the terminal, the identification information of the ML model, the function information of the ML model, the type information of the ML model, the requirement information of the ML model, and the first indication information; wherein the first indication information is used to indicate that the ML model needs to be transmitted using the user plane session.
  • the transmitting the model between the terminal and the user plane session or tunnel includes:
  • the model is transmitted between the terminal and the user plane session or tunnel.
  • the information of the ML model includes at least one of the following:
  • Identification information of the ML model Identification information of the ML model, function information of the ML model, type information of the ML model, requirement information of the ML model, model structure information of the ML model, and model parameter information of the ML model.
  • the requirement information of the ML model includes at least one of the following:
  • Model transmission delay requirement information used to indicate that the model needs to support sharing
  • identity limitation information used to limit the network element identity provided by the model
  • model expression method limitation information used to limit the network element identity provided by the model
  • model performance requirement information used to limit the model needs to support sharing
  • the network interface 1502 is used to: obtain information of a target network element corresponding to a machine learning ML model, wherein the target network element supports the function of a user plane transmission model; and send information of the target network element to a terminal.
  • the information of the target network element includes at least one of the following:
  • the address information of the target network element is the address information of the target network element
  • the target network element stores a uniform resource locator URL of the ML model
  • the target network element transmits the data network name DNN of the ML model
  • the target network element transmits single network slice selection auxiliary information S-NSSAI of the ML model
  • the target network element transmits the radio access technology RAT type of the ML model
  • the access type of the target network element transmitting the ML model
  • the target network element transmits security information related to the user plane tunnel of the ML model.
  • the network interface 1502 is further used for:
  • a first message sent by the terminal is received, where the first message is used to request to obtain the ML model, or the first message is used to request to upload the ML model.
  • the first message includes at least one of the following:
  • the user plane session corresponds to the IP address information of the terminal, the model capability information of the terminal, the identification information of the ML model, the function information of the ML model, the type information of the ML model, the requirement information of the ML model, and the first indication information, wherein the first indication information is used to indicate that the ML model needs to be transmitted using the user plane session.
  • the obtaining information of the target network element corresponding to the ML model includes:
  • the method of acquiring information of a target network element corresponding to the ML model according to the first message includes:
  • the discovery message includes at least one of the following items carried by the first message: identification information of the ML model, function information of the ML model, type information of the ML model, and indication information of a user plane transmission model;
  • the network interface 1502 is further used for:
  • a model providing network element of the ML model is determined.
  • the determining, according to the first message, a model providing network element of the ML model includes:
  • the discovery message includes at least one of the following items carried by the first message:
  • identification information of the ML model identification information of the ML model, function information of the ML model, type information of the ML model, and indication information of a user plane transmission model;
  • the access mobility management function receives the information of the model providing network element sent by the network storage function.
  • model providing network element includes: model training logic network element or model training application function, network interface 1502 is also used for:
  • the obtaining of information of a target network element corresponding to the ML model includes:
  • the indication information includes at least one of the following:
  • Model capability information of the terminal identification information of the ML model, function information of the ML model, type information of the ML model, requirement information of the ML model, indication information of the user plane transmission model, analysis identifier, and notification target address of the ML model;
  • the analysis identifier is used to indicate that the ML model is used for the data analysis task corresponding to the analysis identifier
  • the notification target address of the ML model is empty, or the notification target address of the ML model is the IP address information of the terminal corresponding to the user plane session, or the notification target address is the address information of the access mobility management function.
  • the response message also includes:
  • First indication information wherein the first indication information is used to indicate that the ML model needs to be transmitted using a user plane session.
  • the obtaining information of the target network element corresponding to the ML model includes:
  • the information of the target network element corresponding to the ML model is obtained from the network storage function.
  • the network interface 1502 is further used for:
  • a second message sent by the terminal is received, where the second message is used to report or update model capability information of the terminal.
  • the model capability information of the terminal includes at least one of the following:
  • the terminal supports the indication information received by the model
  • the terminal supports the indication information sent by the model
  • the terminal transmits capability information of the model via the user plane
  • the model of the terminal stores spatial information.
  • the sending the information of the target network element to the terminal includes:
  • the third message including information of a target network element corresponding to the ML model, and also including at least one of the following:
  • the first indication information is used to indicate that the ML model needs to be transmitted using a user plane session
  • the network interface 1502 is further used for:
  • a fourth message sent by the terminal is received, where the fourth message includes IP address information of the user plane session corresponding to the terminal.
  • the fourth message further includes:
  • the network interface 1502 is also used for:
  • the user plane session corresponds to the IP address information of the terminal
  • Capability information of a user plane transmission model of the terminal
  • the network interface 1502 is used to send information of a target network element corresponding to the ML model to the access mobility management function or the terminal, where the target network element supports the function of the user plane transmission model.
  • the information of the target network element includes at least one of the following:
  • the address information of the target network element is the address information of the target network element
  • the target network element stores a uniform resource locator URL of the ML model
  • the target network element transmits the data network name DNN of the ML model
  • the target network element transmits single network slice selection auxiliary information S-NSSAI of the ML model
  • the target network element transmits the radio access technology RAT type of the ML model
  • the access type of the target network element transmitting the ML model
  • the target network element transmits security information related to the user plane tunnel of the ML model.
  • the network interface 1502 is further used for:
  • a first message is received from the terminal, where the first message is used to request to obtain the ML model, or the first message is used to trigger uploading of the ML model.
  • the first message includes at least one of the following:
  • the user plane session corresponds to the IP address information of the terminal, the model capability information of the terminal, the identification information of the ML model, the function information of the ML model, the type information of the ML model, the requirement information of the ML model, and the first indication information, wherein the first indication information is used to indicate that the ML model needs to be transmitted using the user plane session.
  • the network interface 1502 is further used for:
  • the sending information of the target network element corresponding to the ML model to the access mobility management function includes:
  • the access mobility management function Send a response message to the access mobility management function, the response message including the target corresponding to the ML model Information about network elements.
  • the indication information includes at least one of the following:
  • Model capability information of the terminal identification information of the ML model, function information of the ML model, type information of the ML model, requirement information of the ML model, indication information of the user plane transmission model, analysis identifier, and notification target address of the ML model;
  • the analysis identifier is used to indicate that the ML model is used for the data analysis task corresponding to the analysis identifier
  • the notification target address of the ML model is empty, or the notification target address of the ML model is the IP address information of the terminal corresponding to the user plane session, or the notification target address is the address information of the access mobility management function.
  • the network interface 1502 is further used for:
  • the information of the target network element is determined according to the first message or the indication information.
  • the determining the information of the target network element according to the first message or the indication information includes:
  • the discovery message includes at least one of the following: identification information of the ML model, function information of the ML model, type information of the ML model, and indication information of a user plane transmission model;
  • the model training logic network element receives the information of the target network element sent by the network storage function.
  • the determining the information of the target network element according to the first message or the indication information includes:
  • the response message also includes:
  • First indication information wherein the first indication information is used to indicate that the ML model needs to be transmitted using a user plane session.
  • the network interface 1502 is further used for at least one of the following:
  • the IP address information of the terminal corresponding to the user plane session
  • Capability information of a user plane transmission model of the terminal
  • An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored.
  • the program or instruction is executed by a processor, the steps of the information transmission method on the terminal side provided in the embodiment of the present application are implemented, or the steps of the information transmission method on the target network element side provided in the embodiment of the present application are implemented, or the steps of the information transmission method on the access mobility management function side provided in the embodiment of the present application are implemented, or the steps of the information transmission method on the model training logic network element side provided in the embodiment of the present application are implemented.
  • the processor is the processor in the terminal described in the above embodiment.
  • the readable storage medium includes a computer readable storage medium, such as a computer read-only memory ROM, a random access memory RAM, a magnetic disk or an optical disk.
  • An embodiment of the present application further provides a chip, which includes a processor and a communication interface, wherein the communication interface is coupled to the processor, and the processor is used to run a program or instruction to implement an information transmission method on the terminal side as provided in an embodiment of the present application, or to implement an information transmission method on the target network element side as provided in an embodiment of the present application, or to implement an information transmission method on the access mobility management function side as provided in an embodiment of the present application, or to implement an information transmission method on the model training logic network element side as provided in an embodiment of the present application.
  • the chip mentioned in the embodiments of the present application can also be called a system-level chip, a system chip, a chip system or a system-on-chip chip, etc.
  • the embodiments of the present application further provide a computer program/program product, which is stored in a storage medium, and is executed by at least one processor to implement the steps of the information transmission method on the terminal side as provided in the embodiments of the present application, or the computer program/program product is executed by at least one processor to implement the steps of the information transmission method on the target network element side as provided in the embodiments of the present application, or the computer program/program product is executed by at least one processor to implement the steps of the information transmission method on the access mobility management function side as provided in the embodiments of the present application, or the computer program/program product is executed by at least one processor to implement the steps of the information transmission method on the model training logic network element side as provided in the embodiments of the present application.
  • An embodiment of the present application also provides an information transmission system, including: a terminal, a target network element, an access mobility management function and a model training logic network element, wherein the terminal can be used to execute the steps of the information transmission method on the terminal side provided in the embodiment of the present application, the target network element can be used to execute the steps of the information transmission method on the target network element side provided in the embodiment of the present application, the access mobility management function can be used to execute the steps of the information transmission method on the access mobility management function side provided in the embodiment of the present application, and the model training logic network element can be used to execute the steps of the information transmission method on the model training logic network element side provided in the embodiment of the present application.
  • the disclosed part may be embodied in the form of a computer software product, which is stored in a storage medium (such as ROM/RAM, magnetic disk, optical disk), and includes a number of instructions for enabling a terminal (which may be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in the various embodiments of the present application.
  • a storage medium such as ROM/RAM, magnetic disk, optical disk

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Databases & Information Systems (AREA)
  • Artificial Intelligence (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Computer Security & Cryptography (AREA)
  • Mobile Radio Communication Systems (AREA)

Abstract

本申请公开了一种信息传输方法、装置、终端及网络侧设备，属于通信技术领域，本申请实施例的信息传输方法包括：终端获取机器学习ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；所述终端利用用户面会话或者隧道与所述目标网元之间进行模型传输，其中，所述隧道为基于用户面会话建立的所述终端和所述目标网元之间的隧道，所述模型传输包括如下至少一项：从所述目标网元接收所述ML模型的信息；向所述目标网元发送所述ML模型的信息。

Description

信息传输方法、装置、终端及网络侧设备
相关申请的交叉引用
本申请主张在2022年12月07日在中国提交的中国专利申请No.202211567268.4的优先权，其全部内容通过引用包含于此。
技术领域
本申请属于通信技术领域，具体涉及一种信息传输方法、装置、终端及网络侧设备。
背景技术
目前通信技术领域正在研究基于机器学习(Machine Learning，ML)模型的通信性能增强，从而进一步提升终端的通信性能。目前对于终端如何下载或者上传ML模型的信息尚无解决方案，也就是说，目前终端还无法传输ML模型的信息，如无法从其他设备获取ML模型的信息，无法向其他设备发送ML模型的信息，进而导致终端的传输模型的性能比较差。
发明内容
本申请实施例提供一种信息传输方法、装置、终端及网络侧设备，能够解决终端的传输模型的性能比较差的问题。
第一方面，提供了一种信息传输方法，包括：
终端获取机器学习ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；
所述终端利用用户面会话或者隧道与所述目标网元之间进行模型传输，其中，所述隧道为基于用户面会话建立的所述终端和所述目标网元之间的隧道，所述模型传输包括如下至少一项：
从所述目标网元接收所述ML模型的信息；
向所述目标网元发送所述ML模型的信息。
第二方面，提供了一种信息传输方法，包括：
目标网元利用用户面会话或者隧道与终端之间进行模型传输，其中，所述目标网元支持用户面传输模型的功能，所述隧道为基于用户面会话建立的所述目标网元和终端之间的隧道，所述模型传输包括如下至少一项：
向所述终端发送所述ML模型的信息；
接收所述终端发送的所述ML模型的信息。
第三方面，提供了一种信息传输方法，包括：
接入移动管理功能获取机器学习ML模型对应的目标网元的信息，所述目标网元支持 用户面传输模型的功能；
所述接入移动管理功能向终端发送所述目标网元的信息。
第四方面，提供了一种信息传输方法，包括：
模型训练逻辑网元向接入移动管理功能或者终端发送ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能。
第五方面，提供了一种信息传输装置，包括：
获取模块，用于获取机器学习ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；
传输模块，用于利用用户面会话或者隧道与所述目标网元之间进行模型传输，其中，所述隧道为基于用户面会话建立的终端和所述目标网元之间的隧道，所述模型传输包括如下至少一项：
从所述目标网元接收所述ML模型的信息；
向所述目标网元发送所述ML模型的信息。
第六方面，提供了一种信息传输装置，包括：
传输模块，用于利用用户面会话或者隧道与终端之间进行模型传输，其中，所述装置对应的目标网元支持用户面传输模型的功能，所述隧道为基于用户面会话建立的所述目标网元和终端之间的隧道，所述模型传输包括如下至少一项：
向所述终端发送所述ML模型的信息；
接收所述终端发送的所述ML模型的信息。
第七方面，提供了一种信息传输装置，包括：
获取模块，用于获取机器学习ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；
第一发送模块，用于向终端发送所述目标网元的信息。
第八方面，提供了一种信息传输装置，包括：
发送模块，用于向接入移动管理功能或者终端发送ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能。
第九方面，提供了一种终端，包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如本申请实施例提供的终端侧的信息传输方法的步骤。
第十方面，提供了一种终端，包括处理器及通信接口，其中，所述通信接口用于获取机器学习ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；利用用户面会话或者隧道与所述目标网元之间进行模型传输，其中，所述隧道为基于用户面会话建立的终端和所述目标网元之间的隧道，所述模型传输包括如下至少一项：从所述目标网元接收所述ML模型的信息；向所述目标网元发送所述ML模型的信息。
第十一方面，提供了一种网络侧设备，所述网络侧设备为目标网元，包括处理器和存 储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现本申请实施例提供的目标网元侧的信息传输方法的步骤。
第十二方面，提供了一种网络侧设备，所述网络侧设备为目标网元，包括处理器及通信接口，其中，所述通信接口用于利用用户面会话或者隧道与终端之间进行模型传输，其中，所述目标网元支持用户面传输模型的功能，所述隧道为基于用户面会话建立的所述目标网元和终端之间的隧道，所述模型传输包括如下至少一项：向所述终端发送所述ML模型的信息；接收所述终端发送的所述ML模型的信息。
第十三方面，提供了一种网络侧设备，所述网络侧设备为接入移动管理功能，包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现本申请实施例提供的接入移动管理功能侧的信息传输方法的步骤。
第十四方面，提供了一种网络侧设备，所述网络侧设备为接入移动管理功能，包括处理器及通信接口，其中，所述通信接口用于获取机器学习ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；向终端发送所述目标网元的信息。
第十五方面，提供了一种网络侧设备，所述网络侧设备为模型训练逻辑网元，包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如本申请实施例提供的模型训练逻辑网元侧的信息传输方法的步骤。
第十六方面，提供了一种网络侧设备，所述网络侧设备为模型训练逻辑网元，包括处理器及通信接口，其中，所述通信接口用于向接入移动管理功能或者终端发送ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能。
第十七方面，提供了一种信息传输系统，包括：终端、目标网元、接入移动管理功能和模型训练逻辑网元，所述终端可用于执行如本申请实施例提供的终端侧的信息传输方法的步骤，所述目标网元可用于执行如本申请实施例提供的目标网元侧的信息传输方法的步骤，所述接入移动管理功能可用于执行如本申请实施例提供的接入移动管理功能侧的信息传输方法的步骤，所述模型训练逻辑网元可用于执行如本申请实施例提供的模型训练逻辑网元侧的信息传输方法的步骤。
第十八方面，提供了一种可读存储介质，所述可读存储介质上存储程序或指令，所述程序或指令被处理器执行时实现如本申请实施例提供的终端侧的信息传输方法的步骤，或者实现如本申请实施例提供的目标网元侧的信息传输方法的步骤，或者实现如本申请实施例提供的接入移动管理功能侧的信息传输方法的步骤，或者实现如本申请实施例提供的模型训练逻辑网元侧的信息传输方法的步骤。
第十九方面，提供了一种芯片，所述芯片包括处理器和通信接口，所述通信接口和所述处理器耦合，所述处理器用于运行程序或指令，实现如本申请实施例提供的终端侧的信息传输方法，或实现如本申请实施例提供的目标网元侧的信息传输方法，或实现如本申请实施例提供的接入移动管理功能侧的信息传输方法，或实现如本申请实施例提供的模型训 练逻辑网元侧的信息传输方法。
第二十方面，提供了一种计算机程序/程序产品，所述计算机程序/程序产品被存储在存储介质中，所述计算机程序/程序产品被至少一个处理器执行以实现如本申请实施例提供的终端侧的信息传输方法的步骤，或所述计算机程序/程序产品被至少一个处理器执行以实现如本申请实施例提供的目标网元侧的信息传输方法的步骤，或所述计算机程序/程序产品被至少一个处理器执行以实现如本申请实施例提供的接入移动管理功能侧的信息传输方法的步骤，或所述计算机程序/程序产品被至少一个处理器执行以实现如本申请实施例提供的模型训练逻辑网元侧的信息传输方法的步骤。
在本申请实施例中，终端获取ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；所述终端利用用户面会话或者隧道与所述目标网元之间进行模型传输，其中，所述隧道为基于用户面会话建立的所述终端和所述目标网元之间的隧道，所述模型传输包括如下至少一项：从所述目标网元下载所述ML模型的信息；向所述目标网元上传所述ML模型的信息。这样可以实现终端与目标网元之间传输ML模型的信息，进而提高终端的传输模型的性能。
附图说明
图1是本申请实施例可应用的一种无线通信系统的框图；
图2是本申请实施例提供的一种模型应用的示意图；
图3是本申请实施例提供的一种信息传输方法的流程图；
图4是本申请实施例提供的另一种信息传输方法的流程图；
图5是本申请实施例提供的另一种信息传输方法的流程图；
图6是本申请实施例提供的另一种信息传输方法的流程图；
图7是本申请实施例提供的一种信息传输方法的示意图；
图8是本申请实施例提供的另一种信息传输方法的示意图；
图9是本申请实施例提供的一种信息传输装置的结构图；
图10是本申请实施例提供的另一种信息传输装置的结构图；
图11是本申请实施例提供的另一种信息传输装置的结构图；
图12是本申请实施例提供的另一种信息传输装置的结构图；
图13是本申请实施例提供的一种通信设备的结构图；
图14是本申请实施例提供的另一种终端的结构图；
图15是本申请实施例提供的另一种网络侧设备的结构图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施 例，本领域普通技术人员所获得的所有其他实施例，都属于本申请保护的范围。
本申请的说明书和权利要求书中的术语“第一”、“第二”等是用于区别类似的对象，而不用于描述特定的顺序或先后次序。应该理解这样使用的术语在适当情况下可以互换，以便本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施，且“第一”、“第二”所区别的对象通常为一类，并不限定对象的个数，例如第一对象可以是一个，也可以是多个。此外，说明书以及权利要求中“和/或”表示所连接对象的至少其中之一，字符“/”一般表示前后关联对象是一种“或”的关系。
本申请的说明书和权利要求书中的术语“指示”既可以是一个明确的指示，也可以是一个隐含的指示。其中，明确的指示可以理解为，发送方在发送的指示中明确告知了接收方需要执行的操作或请求结果；隐含的指示可以理解为，接收方根据发送方发送的指示进行判断，根据判断结果确定需要执行的操作或请求结果。
值得指出的是，本申请实施例所描述的技术不限于长期演进型(Long Term Evolution，LTE)/LTE的演进(LTE-Advanced，LTE-A)系统，还可用于其他无线通信系统，诸如码分多址(Code Division Multiple Access，CDMA)、时分多址(Time Division Multiple Access，TDMA)、频分多址(Frequency Division Multiple Access，FDMA)、正交频分多址(Orthogonal Frequency Division Multiple Access，OFDMA)、单载波频分多址(Single-carrier Frequency Division Multiple Access，SC-FDMA)和其他系统。本申请实施例中的术语“系统”和“网络”常被可互换地使用，所描述的技术既可用于以上提及的系统和无线电技术，也可用于其他系统和无线电技术。以下描述出于示例目的描述了第5代(5thGeneration，5G)系统，并且在以下大部分描述中使用5G术语，但是这些技术也可应用于NR系统应用以外的应用，如第6代(6thGeneration，6G)通信系统。
图1示出本申请实施例可应用的一种无线通信系统的框图。无线通信系统包括终端11和网络侧设备12。其中，终端11可以是手机、平板电脑(Tablet Personal Computer)、膝上型电脑(Laptop Computer)或称为笔记本电脑、个人数字助理(Personal Digital Assistant，PDA)、掌上电脑、上网本、超级移动个人计算机(ultra-mobile personal computer，UMPC)、移动上网装置(Mobile Internet Device，MID)、增强现实(augmented reality，AR)/虚拟现实(virtual reality，VR)设备、机器人、可穿戴式设备(Wearable Device)、车载设备(Vehicle User Equipment，VUE)、行人终端(Vehicle User Equipment，PUE)、智能家居(具有无线通信功能的家居设备，如冰箱、电视、洗衣机或者家具等)、游戏机、个人计算机(personal computer，PC)、柜员机或者自助机等终端侧设备，可穿戴式设备包括：智能手表、智能手环、智能耳机、智能眼镜、智能首饰(智能手镯、智能手链、智能戒指、智能项链、智能脚镯、智能脚链等)、智能腕带、智能服装等。需要说明的是，在本申请实施例并不限定终端11的具体类型。网络侧设备12可以包括接入网设备或核心网设备，其中，接入网设备也可以称为无线接入网设备、无线接入网(Radio Access Network,RAN)、无线接入网功能或无线接入网单元。接入网设备可以包括基站、无线局域网(Wireless Local Area  Network，WLAN)接入点或WiFi节点等，基站可被称为节点B、演进节点B(eNB)、接入点、基收发机站(Base Transceiver Station，BTS)、无线电基站、无线电收发机、基本服务集(Basic Service Set，BSS)、扩展服务集(Extended Service Set，ESS)、家用B节点、家用演进型B节点、发送接收点(Transmitting Receiving Point，TRP)或所述领域中其他某个合适的术语，只要达到相同的技术效果，所述基站不限于特定技术词汇，需要说明的是，在本申请实施例中仅以NR系统中的基站为例进行介绍，并不限定基站的具体类型。核心网设备可以包含但不限于如下至少一项：核心网节点、核心网功能、移动管理实体(Mobility Management Entity，MME)、接入移动管理功能(Access and Mobility Management Function，AMF)、会话管理功能(Session Management Function，SMF)、用户平面功能(User Plane Function，UPF)、策略控制功能(Policy Control Function，PCF)、策略与计费规则功能单元(Policy and Charging Rules Function，PCRF)、边缘应用服务发现功能(Edge Application Server Discovery Function，EASDF)、统一数据管理(Unified Data Management，UDM)，统一数据仓储(Unified Data Repository，UDR)、归属用户服务器(Home Subscriber Server，HSS)、集中式网络配置(Centralized network configuration，CNC)、网络存储功能(Network Repository Function，NRF)，网络开放功能(Network Exposure Function，NEF)、本地NEF(Local NEF，或L-NEF)、绑定支持功能(Binding Support Function，BSF)、应用功能(Application Function，AF)等。需要说明的是，在本申请实施例中仅以NR系统中的核心网设备为例进行介绍，并不限定核心网设备的具体类型。
在一些实施例中，基于ML模型的通信网络优化整体流程可以如图2所示，其中，模型训练功能基于训练数据生成ML模型，完成模型有效性测试后，将模型部署至模型推理功能。在生成ML模型过程中，模型训练功能需要获取并分析大量数据，对硬件性能和算力的要求很高，主要部署在网络侧设备上，如运营商服务器，或第三方服务器。
模型推理功能基于上述ML模型，以推理数据为输入，获得推理输出，如空口指标预测。模型推理功能对硬件性能和算力的要求相比模型训练功能更低。
在一些实施例中，在终端侧进行模型推理，甚至部分模型训练在终端侧完成。
在一些场景中，模型训练在通信网络侧(例如模型训练逻辑网元，或AF)完成，模型推理在终端侧。此种场景下，终端可以从网络侧下载模型；
在另一些场景中，局部模型训练在终端侧完成，模型推理在网络侧或其他端侧，终端可以将所训练的模型传递给网络侧。
下面结合附图，通过一些实施例及其应用场景对本申请实施例提供的一种信息传输方法、装置、终端及网络侧设备进行详细地说明。
请参见图3，图3是本申请实施例提供的一种信息传输方法的流程图，如图3所示，包括以下步骤：
步骤301、终端获取ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能。
在一些实施方式中，上述目标网元可以是用户面网元，该用户面网元是存储模型或者可以提供模型实例的网元或功能模块。该用户面网元可以是独立部署的一个数据库网元，数据分析存储库网元(Analytics Data Repository Function，ADRF)，或者模型应用平台，模型商店(model store)等；或者，上述用户面网元可以是与模型训练逻辑网元合设或外挂的一个功能模块、网元或者设备。
在一些实施方式中，上述目标网元还可以是支持用户面传输和控制面传输的网络侧设备。
上述目标网元可以存储上述ML模型，例如：模型训练逻辑网元(Model Training logical function，MTLF)或者其他设备产生的模型可以存储在该目标网元中。
另外，上述ML模型可以是由模型提供网元(也可以称作模型提供者)提供，上述模型提供网元可以是模型训练功能网元、模型存储功能网元、应用服务器或开放应用模型(Open Application Model，OAM)等可以提供ML模型的网元或设备。
本申请实施例中，ML模型可以是AI模型。
步骤302、所述终端利用用户面会话或者隧道与所述目标网元之间进行模型传输，其中，所述隧道为基于用户面会话建立的所述终端和所述目标网元之间的隧道，所述模型传输包括如下至少一项：
从所述目标网元接收所述ML模型的信息；
向所述目标网元发送所述ML模型的信息。
其中，上述用户面会话可以是协议数据单元(Protocol Data Unit，PDU)会话。
上述隧道可以是，在用户会话的基础上，终端和目标网元之间建立安全隧道。
上述从所述目标网元接收所述ML模型的信息也可以称作，从目标网元下载ML模型的信息。
上述向所述目标网元发送所述ML模型的信息也可以称作，向目标网元上传或者上载ML模型的信息。
本申请实施例中，通过上述步骤可以实现终端与目标网元之间利用用户面会话传输ML模型的信息，进而提高终端的传输模型的性能。
另外，终端在获取上述ML模型的信息后，可以基于ML模型的信息进行模型推理，进而提高终端的通信业务或其他业务的性能，如基于ML模型的信息进行空口指标预测、如基于ML模型的信息进行信道信息预测等，对此不作限定。
作为一种可选的实施方式，所述ML模型的信息包括如下至少一项：
所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、所述ML模型的模型结构信息、所述ML模型的模型参数信息。
需要说明的是，所述模型标识(Identity，ID)用于在一定范围内(如公共陆地移动网(Public Land Mobile Network，PLMN)内)唯一地标识所请求的模型实例，也即标识所请求的模型具体是哪个；所述模型功能、类型信息用于表征所述ML模型的功能或用途， 例如可以是包括模型类型(model type)、数据分析任务类型(analytics ID)、模型功能标识(model functionality ID)等。
在一些实施方式中，上述ML模型的要求信息可以包括如下至少一项：
模型传输时延要求信息、模型大小要求信息、用于指示模型需支持共享的共享指示信息、用于限定模型提供网元身份的身份限定信息、模型表述方式限定信息、模型性能要求信息、模型使用范围要求信息。
其中，上述模型传输时延要求信息可以指示对端反馈上述ML模型传输的截止时间、最大时延等信息。
上述模型大小要求信息可以指示上述ML模型传输对存储空间的要求信息。
上述用于指示模型需支持共享的共享指示信息可以指示上述ML模型需要支持在不同厂商或不同功能设备之间共享。
上述用于限定模型提供网元身份的身份限定信息可以包括厂商信息，如限定产生模型的厂商为某个或某些特定厂商，或限定模型提供网元为某个或某些特定的设备。
上述模型表述方式限定信息可以限定ML模型用某些特定的语言或基于某些特定的AI框架(AI framework)来表述，例如，常用的模型语言有开放式神经网络交换(Open Neural Network Exchange，ONNX)、PyTorch模神经网络交换(pytorch neural network exchange，PNNX)等，AI framework有TensorFlow和Pytorch等。
上述模型性能要求信息可以指示终端对模型的准确度的最低、最高要求。其中模型的准确度可以用模型预测结果的平均绝对误差(Mean absolute error，MAE)、最小均方误差(Minimum mean square error，MMSE)或其他形式表示。
上述模型使用范围要求信息可以指示模型的有效区域、适用数据网络名称(Data Network Name，DNN)、适用切片、有效时间等范围信息。
作为一种可选的实施方式，所述目标网元的信息包括如下至少一项：
所述目标网元的全量域名(Fully Qualified Domain Name，FQDN)；
所述目标网元的地址信息；
所述目标网元存储所述ML模型的统一资源定位符(Uniform Resource Locator，URL)；
所述目标网元传输所述ML模型的DNN；
所述目标网元传输所述ML模型的单一网络切片选择辅助信息(Single Network Slice Selection Assistance Information，S-NSSAI)；
所述目标网元传输所述ML模型的无线接入技术(Radio Access Technology，RAT)类型；
所述目标网元传输所述ML模型的接入类型；
所述目标网元传输所述ML模型的用户面隧道相关的安全信息。
其中，URL可以指示目标网元存储ML模型的文件的位置。
上述DNN可以指示获取ML模型的用户面会话对应的DNN。
上述S-NSSAI可以指示获取ML模型的用户面会话对应的S-NSSAI。
上述RAT类型可以指示获取该模型的用户面会话对应的RAT类型。
上述接入类型可以指示获取ML模型的用户面会话对应的接入类型。
上述安全信息可以包括安全证书等用户面隧道相关的安全信息。
该实施方式中，通过上述目标网元的信息可以使得终端获知进行模型传输的对端目的地址以及其他相关信息，以便于终端后续利用用户面会话与对端完成ML模型传输。
在一些实施方式中，上述目标网元的信息也可以包括如下至少一项：
所述目标网元的标识、所述目标网元对应的终端临时标识、所述目标网元对应的临时会话标识。其中，上述目标网元对应的终端临时标识可以是目标网元为终端临时分配的，用于标识该终端的标识，上述目标网元对应的临时会话标识可以是目标网元为该用户面会话临时分配的，用于标识该会话的标识。
作为一种可选的实施方式，所述方法还包括：
所述终端向接入移动管理功能或者模型训练逻辑网元发送第一消息，所述第一消息用于请求获取所述ML模型，或者，所述第一消息用于请求上传所述ML模型。
上述第一消息可以是非接入层(Non-Access Stratum，NAS)消息，如通过NAS消息请求获取所述ML模型，或者，通过NAS消息请求上传所述ML模型。
上述NAS消息包括如下至少一项：
终端注册请求；
模型获取/订阅请求消息；
上行NAS传输(UL NAS transport)消息。
在一些实施方式中，若终端和模型训练逻辑网元之间存在逻辑接口或架设专有协议栈，这样，第一消息可以是终端和模型训练逻辑网元之间的直传消息，即终端和模型训练逻辑网元间传输的信息对于接入移动管理功能而言是透明的，接入移动管理功能并不做解析。
在一些实施方式中，所述第一消息可以包括如下至少一项：
所述用户面会话对应所述终端的互联网协议(Internet Protocol，IP)地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
其中，上述用户面会话可以是终端在已有数据业务的用户面会话中选择用于传输上述ML模型的用户面会话，也可以是终端为传输该ML模型而新建立的用户面会话。
其中，上述终端的模型能力信息可以包括如下至少一项：
所述终端支持的模型的标识信息；
所述终端支持的模型的功能信息；
所述终端支持的模型的类型信息；
所述终端支持模型接收的指示信息；
所述终端支持模型发送的指示信息；
所述终端通过用户面传输模型的能力信息；
所述终端的模型存储空间信息。
上述终端支持的模型标识信息(例如：UE supported model ID)可以用于在一定范围内唯一地标识模型实例；
所述终端支持的模型的功能信息上述可以是，终端支持的模型功能标识(UE supported model type/UE supported analytics ID/supported model functionality ID)，该信息可以说明所支持模型的功能或用途，其中，上述终端支持的模型的功能信息也可以是其他描述信息可以是终端支持的模型的其他描述信息，如用途信息等；
所述终端支持的模型的类型信息上述可以指示终端支持的模型类型。
上述终端支持模型接收的指示信息可以是，终端支持模型下载或接收的能力指示信息，例如：终端是否支持模型下载或接收的能力，进一步可选的，该能力指示信息还可以分不同的模型标识或模型类型而不同，即终端可以是支持下载或者接收部分或者全部模型标识或模型类型对应的模型；
上述终端支持模型发送的指示信息可以是，终端支持模型上传或发送的能力指示信息，例如：终端是否支持模型上传或发送的能力，进一步可选的，该能力指示信息还可以分不同的模型标识或模型类型而不同，即终端可以是支持下上传或发送部分或者全部模型标识或模型类型对应的模型；
上述终端通过用户面传输模型的能力信息可以是，终端是否支持通过用户面会话来接收或者发送ML模型，进一步可选的，该能力信息还可以分不同的模型标识或模型类型而不同，即终端可以通过用户面会话来接收或者发送部分或者全部模型标识或模型类型对应的模型；
上述终端的模型存储空间信息可以是，终端中初始或剩余的可存储模型的存储空间大小。
上述ML模型的标识信息可以指示终端请求获取的ML模型或者请求上传的ML模型实例。
上述ML模型的功能信息可以指示终端请求获取或者上传的ML模型的功能信息。
上述ML模型的类型信息可以指示终端请求获取或者上传的ML模型的类型信息。
上述ML模型的要求信息可以指示终端请求获取或者上传的ML模型的要求信息，其中，该要求信息参见上述实施方式，此处不作赘述。
上述第一指示信息用于指示请求获取或者上传的ML模型需要利用用户面会话进行传输。
该实施方式中，通过上述第一消息包括上述至少一项可以实现终端请求获取或者上传更加精确的ML模型。
作为一种可选的实施方式，所述方法还包括：
所述终端向接入移动管理功能或者模型训练逻辑网元或发送第二消息，所述第二消息用于上报或者更新所述终端的模型能力信息。
其中，上述第二消息可以是在注册过程中、PDU会话建立、PDU会话修改等过程中，或者网络侧下发能力请求(如后面的第三消息包含用户面传输模型能力的获取指示信息)时，向接入移动管理功能或者模型训练逻辑网元发送第二消息。
上述第二消息可以是NAS消息，或者终端和模型训练逻辑网元之间的直传消息。
其中，上述终端的模型能力信息可以包括如下至少一项：
所述终端支持的模型的标识信息；
所述终端支持的模型的功能信息；
所述终端支持的模型的类型信息；
所述终端支持模型接收的指示信息；
所述终端支持模型发送的指示信息；
所述终端通过用户面传输模型的能力信息；
所述终端的模型存储空间信息。
该实施方式中，通过上述第二消息上报或者更新所述终端的模型能力信息，这样可以使得网络侧可以基于终端的模型能力信息进行相应的模型传输。
作为一种可选的实施方式，所述终端获取ML模型对应的目标网元的信息，包括：
所述终端接收接入移动管理功能或者模型训练逻辑网元发送的第三消息，所述第三消息包括所述ML模型对应的目标网元的信息。
上述第三消息可以是NAS消息，或者终端和模型训练逻辑网元之间的直传消息。
该实施方式中，可以实现通过接入移动管理功能或者模型训练逻辑网元获取ML模型对应的目标网元的信息。
可选的，上述第三消息还可以包括如下至少一项：
第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输；
用户面传输模型能力的获取指示信息。
用户面传输模型能力的获取指示信息可以指示终端上报自身的关于利用用户面会话传输模型的能力。
通过上述第一指示信息和用户面传输模型能力的获取指示信息中的至少一项可以指示终端通过用户面传输上述ML模型，需要说明的是，在一些实施方式中，也可以是协议约定或者预先配置通过用户面传输上述ML模型。
可选的，所述方法还包括：
所述终端向所述接入移动管理功能或者模型训练逻辑网元发送第四消息，所述第四消息包括所述用户面会话对应所述终端的IP地址信息。
其中，上述第四消息可以是在确定需要利用用户面会话传输模型后发送的，例如：在 接收到上述第三消息后，确定需要利用用户面会话传输模型，则将IP地址、用户面传输模型的能力信息等通过第四消息上报。
上述用户面会话可以是，终端在接收到上述第三消息后，终端选择已有数据业务的用户面会话，或为传输该ML模型而新建用户面会话，用于传输上述ML模型，并发送对应的IP地址信息。
需要说明的是，在一些实施方式中，上述用户面会话对应所述终端的IP地址信息也可以在接收到目标网络的信息之前发送，如在上述实施方式描述的第一消息包括用户面会话对应所述终端的IP地址信息。
可选的，所述第四消息还包括：
所述终端的用户面传输模型的能力信息。
上述终端的用户面传输模型的能力信息可以表示终端支持用户面传输模型，进一步，还可以支持模型功能、模型类型等。
通过上述终端的用户面传输模型的能力信息可以使得网络侧基于该能力信息采用用户面会话传输ML模型。
作为一种可选的实施方式，所述终端利用用户面会话或者隧道与所述目标网元之间进行模型传输，包括：
所述终端根据所述目标网元的信息，确定传输所述ML模型对应的对端目的地址；
所述终端利用所述用户面会话或者隧道与所述对端目的地址之间进行模型传输。
其中，上述终端根据所述目标网元的信息，确定传输所述ML模型对应的对端目的地址，可以包括以下至少一项：
所述终端根据所述目标网元的FQDN，确定所述目标网元的IP地址，所述终端将所述目标网元的IP地址作为所述对端目的地址；
所述终端将所述目标网元的地址作为所述对端目的地址；
所述终端根据所述目标网元存储所述ML模型的URL，确定所述对端目的地址。
上述终端根据所述目标网元存储所述ML模型的URL，确定所述对端目的地址可以是，可以是根据上述目标网元存储所述ML模型的URL，确定存储模型位置对应的IP地址，将该IP地址作为上述对端目的地址。该情况下，目标网元可以是模型存储网元，目标网元的信息是模型存储网元存储模型的URL。
该实施方式中，可以实现基于用户面会话或者隧道与上述对端目的地址之间进行模型传输。
需要说明的是，在一些实施方式中，上述目标网元的信息也可以直接是上述ML模型对应的对端目的地址，即终端直接基于目标网元的信息与对端目的地址之间进行模型传输。
在本申请实施例中，终端获取ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；所述终端利用用户面会话或者隧道与所述目标网元之间进行模型传输，其中，所述隧道为基于用户面会话建立的所述终端和所述目标网元之间的隧道，所述 模型传输包括如下至少一项：从所述目标网元下载所述ML模型的信息；向所述目标网元上传所述ML模型的信息。这样可以实现终端与目标网元之间传输ML模型的信息，进而提高终端的传输模型的性能。
请参见图4，图4是本申请实施例提供的另一种信息传输方法的流程图，如图4所示，包括以下步骤：
步骤401、目标网元利用用户面会话或者隧道与终端之间进行模型传输，其中，所述目标网元支持用户面传输模型的功能，所述隧道为基于用户面会话建立的所述目标网元和终端之间的隧道，所述模型传输包括如下至少一项：
向所述终端发送所述ML模型的信息；
接收所述终端发送的所述ML模型的信息。
其中，上述目标网元利用用户面会话或者隧道与终端之间进行模型传输可以参见图3所示的实施例的相应说明，此处不作赘述。
可选的，所述方法还包括：
所述目标网元获取如下至少一项：
所述用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息；其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
其中，上述目标网元获取可以是接收其他设备或者网元发送的上述至少一项。
在一些实施方式中，上述目标网元接收接入移动管理功能或者模型训练逻辑网元发送的如下至少一项：
所述用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息；其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
其中，上述至少一项可以是在接入移动管理功能或者模型训练逻辑网元接收到上述终端发送的第一消息后，向上述目标网元发送的上述至少一项，其中，上述第一消息包括如下至少一项：
所述用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
需要说明的是，在一些实施方式中，接入移动管理功能或者模型训练逻辑网元可以通过一个或者多个消息发送上述至少一项。
通过一消息发送：所述终端的模型能力信息、所述ML模型的标识信息、所述ML模 型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息中的至少一项，再通过另一消息发送所述用户面会话对应所述终端的IP地址信息。
通过上述至少一项可以使得目标网元与上述终端进行更加精确的模型传输。
可选的，所述目标网元利用用户面会话或者隧道与终端之间进行模型传输，包括：
所述目标网元根据所述终端的IP地址信息，利用用户面会话或者隧道与终端之间进行模型传输。
上述目标网元根据所述终端的IP地址信息，利用用户面会话或者隧道与终端之间进行模型传输可以是，将上述终端的IP地址信息作为ML模型对应的对端目的地址，目标网元利用所述用户面会话或者隧道与该对端目的地址之间进行模型传输。
可选的，所述ML模型的信息包括如下至少一项：
所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、所述ML模型的模型结构信息、所述ML模型的模型参数信息。
可选的，所述ML模型的要求信息包括如下至少一项：
模型传输时延要求信息、模型大小要求信息、用于指示模型需支持共享的共享指示信息、用于限定模型提供网元身份的身份限定信息、模型表述方式限定信息、模型性能要求信息、模型使用范围要求信息。
需要说明的是，本实施例作为与图3所示的实施例中对应的目标网元的实施方式，其具体的实施方式可以参见图3所示的实施例的相关说明，以为避免重复说明，本实施例不再赘述。
请参见图5，图5是本申请实施例提供的另一种信息传输方法的流程图，如图5所示，包括以下步骤：
步骤501、接入移动管理功能获取机器学习ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；
步骤502、所述接入移动管理功能向终端发送所述目标网元的信息。
其中，上述目标网元的信息，以及上述目标网元的信息的作用参见图3所示实施例的相应说明，此处不作赘述。
可选的，所述目标网元的信息包括如下至少一项：
所述目标网元的全量域名FQDN；
所述目标网元的地址信息；
所述目标网元存储所述ML模型的统一资源定位符URL；
所述目标网元传输所述ML模型的数据网络名称DNN；
所述目标网元传输所述ML模型的单一网络切片选择辅助信息S-NSSAI；
所述目标网元传输所述ML模型的无线接入技术RAT类型；
所述目标网元传输所述ML模型的接入类型；
所述目标网元传输所述ML模型的用户面隧道相关的安全信息。
可选的，所述方法还包括：
所述接入移动管理功能接收所述终端发送的第一消息，所述第一消息用于请求获取所述ML模型，或者，所述第一消息用于请求上传所述ML模型。
可选的，所述第一消息包括如下至少一项：
所述用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
其中，上述第一消息参见图3所示实施例的相应说明，此处不作赘述。
可选的，所述接入移动管理功能获取ML模型对应的目标网元的信息，包括：
所述接入移动管理功能根据所述第一消息，获取所述ML模型对应的目标网元的信息。
上述接入移动管理功能根据所述第一消息，获取所述ML模型对应的目标网元的信息可以是，根据上述第一消息确定上述ML模型，再获取ML模型对应的目标网元的信息。
可选的，所述接入移动管理功能根据所述第一消息获取所述ML模型对应的目标网元的信息，包括：
所述接入移动管理功能向网络存储功能发送用于确定所述ML模型的对应的目标网元的发现消息，所述发现消息包括所述第一消息携带的如下至少一项：所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、用户面传输模型的指示信息；
所述接入移动管理功能接收所述网络存储功能发送的所述目标网元的信息。
需要说明的是，本申请实施例中并不限定通过上述网络存储功能目标网元的信息，例如：在一些实施方式中，也可以是上述接入移动管理功能预先配置有多个ML模型对应的目标网元的信息，从而接入移动管理功能可以直接确定上述目标网元的信息。
可选的，所述方法还包括：
所述接入移动管理功能根据所述第一消息，确定所述ML模型的模型提供网元。
其中，上述模型提供网元可以参见图3所示实施例的相应说明，此处不作赘述。
上述确定所述ML模型的模型提供网元可以是，在接入移动管理功能已知的设备或者网元中选择上述ML模型的模型提供网元，也可以是接入移动管理功能向其他设备或者网元获取上述ML模型的模型提供网元。
例如：上述接入移动管理功能根据所述第一消息，确定所述ML模型的模型提供网元，包括：
所述接入移动管理功能向网络存储功能发送用于确定所述ML模型的模型提供网元的发现消息，所述发现消息包括所述第一消息携带的如下至少一项：
所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、用户面传输模型的指示信息；
所述接入移动管理功能接收所述网络存储功能发送的所述模型提供网元的信息。
可选的，所述模型提供网元包括：模型训练逻辑网元或模型训练应用功能，所述方法还包括：
所述接入移动管理功能向所述模型训练逻辑网元或模型训练应用功能发送指示消息，所述指示信息用于指示所述ML模型；
所述接入移动管理功能获取ML模型对应的目标网元的信息，包括：
所述接入移动管理功能接收所述模型训练逻辑网元或模型训练应用功能发送响应消息，所述响应消息包括所述ML模型对应的目标网元的信息。
其中，上述指示信息可以包括如下至少一项：
所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息，用户面传输模型的指示信息、分析标识(Analytic ID)、所述ML模型的通知目标地址(Notification Target Address)；
其中，所述分析标识用于指示所述ML模型用于所述分析标识对应的数据分析任务；
所述ML模型的通知目标地址为空，或者，所述ML模型的通知目标地址为用户面会话对应所述终端的IP地址信息，或者，所述通知目标地址为所述接入移动管理功能的地址信息。
上述分析标识可以表示所请求的模型是用于某种analytics ID对应的数据分析任务，可选的，该分析标识是AMF根据终端所请求的ML模型的模型功能类型确定或映射而来。
上述ML模型的通知目标地址为空可以是，因接入移动管理功能还未知道终端的IP地址信息，此时，可以将通知目标地址置为空或者设置为接入移动管理功能对应的地址信息。
上述实施方式中，可以实现通过模型训练逻辑网元或模型训练应用功能获取目标网元的信息。
可选的，上述响应消息还包括：
第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
可选的，所述接入移动管理功能获取ML模型对应的目标网元的信息，包括：
所述接入移动管理功能从网络存储功能获取所述ML模型对应的目标网元的信息。
该实施方式中，可以是直接网络存储功能获取ML模型对应的目标网元的信息。
可选的，所述方法还包括：
所述接入移动管理功能接收所述终端发送的第二消息，所述第二消息用于上报或者更新所述终端的模型能力信息。
可选的，所述终端的模型能力信息包括如下至少一项：
所述终端支持的模型的标识信息；
所述终端支持的模型的功能信息；
所述终端支持的模型的类型信息；
所述终端支持模型接收的指示信息；
所述终端支持模型发送的指示信息；
所述终端通过用户面传输模型的能力信息；
所述终端的模型存储空间信息。
其中，上述第二消息参见图3所示实施例的相应说明，此处不作赘述。
可选的，所述接入移动管理功能向终端发送所述目标网元的信息，包括：
所述接入移动管理功能向所述终端发送第三消息，所述第三消息包括所述ML模型对应的目标网元的信息，以及还包括如下至少一项：
第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输；
用户面传输模型能力的获取指示信息。
其中，上述第三消息参见图3所示实施例的相应说明，此处不作赘述。
可选的，所述方法还包括：
所述接入移动管理功能接收所述终端发送的第四消息，所述第四消息包括用户面会话对应所述终端的IP地址信息。
其中，上述第四消息参见图3所示实施例的相应说明，此处不作赘述。
可选的，所述第四消息还包括：
所述终端的用户面传输模型的能力信息；
所述方法还包括：
所述接入移动管理功向模型训练逻辑网元或者所述目标网元发送第五消息，所述第五消息包括如下至少一项：
所述用户面会话对应所述终端的IP地址信息；
所述终端的用户面传输模型的能力信息。
需要说明的是，本实施例作为与图3所示的实施例中对应的接入移动管理功的实施方式，其具体的实施方式可以参见图3所示的实施例的相关说明，以为避免重复说明，本实施例不再赘述。
请参见图6，图6是本申请实施例提供的另一种信息传输方法的流程图，如图6所示，包括以下步骤：
步骤601、模型训练逻辑网元向接入移动管理功能或者终端发送ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能。
其中，上述目标网元的信息，以及上述目标网元的信息的作用参见图3所示实施例的相应说明，此处不作赘述。
可选的，所述目标网元的信息包括如下至少一项：
所述目标网元的全量域名FQDN；
所述目标网元的地址信息；
所述目标网元存储所述ML模型的统一资源定位符URL；
所述目标网元传输所述ML模型的数据网络名称DNN；
所述目标网元传输所述ML模型的单一网络切片选择辅助信息S-NSSAI；
所述目标网元传输所述ML模型的无线接入技术RAT类型；
所述目标网元传输所述ML模型的接入类型；
所述目标网元传输所述ML模型的用户面隧道相关的安全信息。
可选的，所述方法还包括：
所述模型训练逻辑网元接收来自所述终端的第一消息，所述第一消息用于请求获取所述ML模型，或者，所述第一消息用于触发上传所述ML模型。
其中，上述第一消息可以是终端直接发送给模型训练逻辑网元，也可以是通过接入移动管理功能中转的，且中转的时候消息可能发生了改变。
可选的，所述第一消息包括如下至少一项：
用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
其中，上述第一消息参见图3所示实施例的相应说明，此处不作赘述。
可选的，所述方法还包括：
所述模型训练逻辑网元接收所述接入移动管理功能发送的指示消息，所述指示信息用于指示所述ML模型；
所述模型训练逻辑网元向接入移动管理功能发送ML模型对应的目标网元的信息，包括：
所述模型训练逻辑网元向所述接入移动管理功能发送响应消息，所述响应消息包括所述ML模型对应的目标网元的信息。
可选的，所述指示信息包括如下至少一项：
所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息，用户面传输模型的指示信息、分析标识、所述ML模型的通知目标地址；
其中，所述分析标识用于指示所述ML模型用于所述分析标识对应的数据分析任务；
所述ML模型的通知目标地址为空，或者，所述ML模型的通知目标地址为用户面会话对应所述终端的IP地址信息，或者，所述通知目标地址为所述接入移动管理功能的地址信息。
其中，上述指示消息和响应响应参见图5所示实施例的相应说明，此处不作赘述。
可选的，所述方法还包括：
所述模型训练逻辑网元根据所述第一消息或所述指示信息确定所述目标网元的信息。
其中，根据所述第一消息或所述指示信息确定所述目标网元的信息可以是，在模型训练逻辑网元已知的设备或者网元中选择上述目标网元，也可以是模型训练逻辑网元向其他设备或者网元获取上述目标网元的信息。
例如：所述模型训练逻辑网元根据所述第一消息或所述指示信息确定所述目标网元的信息，包括：
所述模型训练逻辑网元向网络存储功能发送用于确定所述ML模型的对应的目标网元的发现消息，所述发现消息包括如下至少一项：所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、用户面传输模型的指示信息；
所述模型训练逻辑网元接收所述网络存储功能发送的所述目标网元的信息。
又例如：所述模型训练逻辑网元根据所述第一消息或所述指示信息确定所述目标网元的信息，包括：
所述模型训练逻辑网元向模型存储网元发送信息请求消息，所述信息请求消息用于请求所述模型存储网元反馈所述ML模型对应的目标网元的信息；
所述模型训练逻辑网元接收所述模型存储网元发送所述ML模型对应的目标网元的信息，所述目标网元的信息包括所述模型存储网元存储所述ML模型的URL。
其中，上述模型存储网元可以是上述目标网元，上述目标网元的信息可以是，目标网元存储模型的URL。
可选的，所述响应消息还包括：
第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
可选的，所述方法还包括如下至少一项：
所述模型训练逻辑网元确定使用用户面方式传输所述ML模型；
所述模型训练逻辑网元获取所述目标网元的信息；
所述模型训练逻辑网元接收所述接入移动管理功能向发送的第五消息，所述第五消息包括如下至少一项：
用户面会话对应所述终端的IP地址信息；
所述终端的用户面传输模型的能力信息。
上述模型训练逻辑网元可以根据本地策略、控制面负载、用户面负载、终端通过用户面传输模型的能力信息、所请求模型的大小要求、传输时延要求等至少一项因素，确定使用用户面方式传输上述ML模型。
上述模型训练逻辑网元获取所述目标网元的信息可以是，模型训练逻辑网元直接获取目标网元的信息。
需要说明的是，本实施例作为与图3所示的实施例中对应的网络侧设备的实施方式，其具体的实施方式可以参见图3所示的实施例的相关说明，以为避免重复说明，本实施例 不再赘述。
下面通过多个实施例对本申请实施例提供的信息传输方法进行举例说明：
实施例一：
该实施例如图7所示，包括如下步骤：
步骤0、可选的，在执行步骤1之前，UE向AMF发起NAS消息，该NAS消息用于上报/更新自身的ML模型能力信息。
其中，此处UE的ML模型能力信息可以包括：
UE支持的模型标识信息(例如UE supported model ID)，该模型ID用于在一定范围内唯一地标识模型实例；
UE支持的模型功能类型或其他描述信息；例如UE支持模型类型/模型功能标识(UE supported model type/supported analytics ID/supported model functionality ID)，用于说明所支持模型的功能或用途；
UE支持模型接收和/或发送的能力指示信息；例如UE是否支持模型接收和/或发送的能力，进一步可选的，该能力指示信息还可以分不同的model ID或model type而不同；
UE通过用户面传输(接收/发送)模型的能力信息；例如，UE是否支持通过用户面PDU会话来发送或接收ML模型，进一步可选的，该能力信息还可以分不同的model ID或model type而不同；
UE的模型存储空间信息，例如，UE中初始或剩余的可存储模型的存储空间大小。
步骤1、UE向AMF发起NAS消息，该NAS消息用于请求获取ML模型。
该模型获取请求的NAS消息中，包括以下至少一项：
UE的ML模型能力信息，具体含义参考步骤0中描述。
UE所请求的ML模型的model ID，
UE所请求的ML模型的模型功能类型。
UE所请求的ML模型的要求信息
用户面传输模型的指示信息，用于向网络指明所请求的模型需要利用用户面PDU会话进行传输。
其中，ML模型的要求信息包括以下至少一项要求：
传输时延要求信息，用于指示网络反馈模型的截止时间、最大时延等信息；
模型的大小要求信息，用于指示网络反馈的模型对存储空间的要求信息；
可共享指示信息，用于指示模型需支持不同厂商或不同设备之间的共享；
厂商或提供者身份限定信息，用于限定产生模型的厂商为某个或某些特定厂商，或限定模型提供网元为某个或某些特定的设备；
模型表述方式限定信息，用于限定模型用某些特定的语言或基于某些特定的AI framework来表述。例如，常用的模型语言有ONNX，PNNX等，AI framework有TensorFlow，Pytorch等；
模型性能要求信息，用于指示UE对模型的准确度的最低、最高要求；
模型使用范围限定信息，用于指示模型的有效区域、适用DNN、适用切片、有效时间等范围信息。
步骤2、可选的，AMF根据UE发送的模型请求消息，选择模型提供网元。
其中，网络中的模型提供网元具体可以是模型训练功能网元、或模型存储功能网元、应用服务器、OAM等可以提供AI/ML model的网元或设备。
具体地，一种方式中，AMF根据以下至少一项信息从NRF中发现选择所述模型提供网元，所述模型提供网元与以下信息匹配。其中，AMF可向NRF提供以下信息的至少一项：
UE所请求的ML模型的model ID。用于指示所选模型提供网元可以提供model ID对应的ML模型；
UE所请求的ML模型的模型功能类型，用于所选模型提供网元可以提供模型功能类型对应的ML模型；
用户面传输模型的指示信息，用于向网络指明所请求的模型需要利用用户面PDU会话进行传输，所选模型提供网元可以支持用用户面PDU会话进行传输模型。
步骤3、假设步骤2中模型提供网元是MTLF，AMF向MTLF发送第一请求消息，用于向MTLF指示UE所请求的模型。
可选的，第一请求消息中携带的内容参考步骤1中的NAS消息内容。
除此之外，第一请求消息中还可以携带模型传输相关的其他信息，如以下至少一项：
分析标识(Analytic ID)：用于表征所请求的模型是用于某种analytics ID对应的数据分析任务。可选的，该analytics ID是AMF根据UE所请求的ML模型的模型功能类型确定或映射而来。
通知目标地址(A Notification Target Address)：在本实施例中，因还AMF还未知道UE的IP地址信息，此时可选的，可以将Notification Target Address置为空或者设置为AMF对应的地址信息。
步骤4、MTLF向AMF返回第一响应消息，其中包括所请求模型对应的用户面网元的信息。
其中，该用户面网元是指存储模型或者可以提供模型实例的网元或功能模块，该用户面网元支持通过用户面会话与终端之间传输ML模型的功能。在实际部署中，该用户面网元可以是独立部署的一个数据库网元(如ADRF，或者模型应用平台，模型商店model store等)，MTLF或者其他设备产生的模型可以存储在该数据库中。另一种部署方式中，该用户面网元可以是与MTLF合设或外挂的一个功能模块。
具体地，用户面网元的信息包括该用户面网元的以下至少一项信息：
该用户面网元的FQDN；
该用户面网元的地址信息(IP地址，媒体访问控制(Media Access Control，MAC) 地址等)；
URL，例如用于指示用户面网元存储模型文件的位置；
DNN：用于指示获取该模型的用户面会话对应的DNN；
S-NSSAI：用于指示获取该模型的用户面会话对应的S-NSSAI；
RAT Type:用于指示获取该模型的用户面会话对应的RAT Type；
Access Type：用于指示获取该模型的用户面会话对应的Access Type；
用户面隧道相关的安全信息：包括安全证书等。
可选的，第一响应消息中可包括用户面传输模型的指示信息，用于指示所需要利用用户面PDU会话传输所请求的模型。
可选的，在发送第一响应消息之前，MTLF还根据本地策略、控制面负载、用户面负载、UE通过用户面传输(接收/发送)模型的能力信息、所请求模型的大小要求、传输时延要求等至少一项因素，确定使用用户面方式传输所述模型。
步骤3a、可选的，在发送第一响应消息之前，若所请求的模型在用户面网元中存储，MTLF向该用户面网元请求模型存储信息，用于先确定模型对应的用户面网元的信息。
例如，此处的存储ML模型的用户面网元是ADRF，MTLF向ADRF发送请求消息，用于获取模型文件的存储地址信息。ADRF向MTLF反馈模型文件存储的URL，则该URL对应为用户面网元的信息。
步骤5、AMF根据第一响应消息，向UE发送NAS消息，其中包括所述模型对应的用户面网元的信息。
可选的，NAS消息中还包括以下至少一项信息：
用户面传输模型的指示信息，用于指示所需要利用用户面PDU会话传输所请求的模型；
UE用户面传输模型能力获取指示信息，用于指示UE上报自身的关于利用用户面PDU会话传输模型的能力。
步骤6、UE选择已有的PDU会话，或新建PDU会话，用于传输所述模型。
在此之前，UE可先确定自身支持用户面传输模型的能力。
步骤7、UE向AMF发送NAS消息，其中包括步骤6中所选择的PDU会话对应的UE IP地址信息。
可选的，该NAS消息中还包括UE的用户面传输模型能力信息。
步骤8、AMF将UE IP地址信息，以及可选的，UE的用户面传输模型能力信息发送给MTLF。
具体地，此时AMF可将步骤3中的A Notification Target Address设置为UE IP地址信息。
可选的，MTLF向所述用户面网元(ADRF)转发UE IP地址信息，以及可选的，UE的用户面传输模型能力信息。
步骤9、可选的，在所述PDU会话的基础上，UE和所述用户面网元之间建立安全隧道。
步骤10、利用所述PDU会话或所述安全隧道，UE和该用户面网元之间交互模型相关信令和/或传输模型。
其中，交互相关信令是指UE和用户面网元之间利用PDU会话交互模型的以下至少一项信息：
ML模型的model ID；
ML模型的模型功能类型；
ML模型的要求信息(具体见步骤1的含义解释)
另外，传输模型是指利用PDU会话UE从ADRF下载或向ADRF上载模型信息，其包括模型结构信息和/或模型参数信息。
实施例二：
该实施例基本流程类似于实施例一，不同点在于UE在请求模型之前，已经确定了自身具备用户面传输模型的能力以及传输模型的PDU会话。UE在请求模型时可直接上报自身的IP地址信息，以及可选的，UE用户面传输模型的能力信息。如图8所示，包括如下步骤：
步骤0、UE选择已有的PDU会话，或新建PDU会话，用于传输所述模型。
在此之前，UE可先确定自身支持用户面传输模型的能力
步骤1、UE向AMF发起NAS消息，该NAS消息用于请求获取ML模型。
具体可参照实施例的NAS消息，但此处NAS消息中还包括步骤0中PDU会话对应的UE IP地址信息。
步骤2、同实施例一步骤2.
步骤3、同实施例一步骤3，不同点在于第一请求消息中还包括所述UE IP地址信息。
具体地，AMF可以将第一请求消息中的Notification Target Address设置为UE IP地址。
步骤3a、同实施例1步骤3a，不同点在于MTLF向ADRF发送请求消息中还包括所述UE IP地址信息。
步骤4、同实施例一步骤4。
步骤5、同实施例一步骤5。
步骤6、实施例一步骤10，利用所述PDU会话，UE和该用户面网元之间交互模型相关信令和/或传输模型。
在此之前，可选的，在所述PDU会话的基础上，UE和所述用户面网元之间建立安全隧道。
本申请实施例中：UE可以获得网络中AI模型的下载地址或位置信息，利用用户面PDU会话的方式从网络动态下载模型。
网络也可以获取UE的IP地址信息，利用用户面PDU会话的方式从UE动态上载模 型。
请参见图9，图9是本申请实施例提供的一种信息传输装置的结构图，如图9所示，信息传输装置900包括：
获取模块901，用于获取机器学习ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；
传输模块902，用于利用用户面会话或者隧道与所述目标网元之间进行模型传输，其中，所述隧道为基于用户面会话建立的终端和所述目标网元之间的隧道，所述模型传输包括如下至少一项：
从所述目标网元接收所述ML模型的信息；
向所述目标网元发送所述ML模型的信息。
可选的，所述ML模型的信息包括如下至少一项：
所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、所述ML模型的模型结构信息、所述ML模型的模型参数信息。
可选的，所述ML模型的要求信息包括如下至少一项：
模型传输时延要求信息、模型大小要求信息、用于指示模型需支持共享的共享指示信息、用于限定模型提供网元身份的身份限定信息、模型表述方式限定信息、模型性能要求信息、模型使用范围要求信息。
可选的，所述目标网元的信息包括如下至少一项：
所述目标网元的全量域名FQDN；
所述目标网元的地址信息；
所述目标网元存储所述ML模型的统一资源定位符URL；
所述目标网元传输所述ML模型的数据网络名称DNN；
所述目标网元传输所述ML模型的单一网络切片选择辅助信息S-NSSAI；
所述目标网元传输所述ML模型的无线接入技术RAT类型；
所述目标网元传输所述ML模型的接入类型；
所述目标网元传输所述ML模型的用户面隧道相关的安全信息。
可选的，所述装置还包括：
第一发送模块，用于向接入移动管理功能或者模型训练逻辑网元发送第一消息，所述第一消息用于请求获取所述ML模型，或者，所述第一消息用于请求上传所述ML模型。
可选的，所述第一消息包括如下至少一项：
所述用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
可选的，所述装置还包括：
第二发送模块，用于向接入移动管理功能或者模型训练逻辑网元发送第二消息，所述第二消息用于上报或者更新所述终端的模型能力信息。
可选的，所述终端的模型能力信息包括如下至少一项：
所述终端支持的模型的标识信息；
所述终端支持的模型的功能信息；
所述终端支持的模型的类型信息；
所述终端支持模型接收的指示信息；
所述终端支持模型发送的指示信息；
所述终端通过用户面传输模型的能力信息；
所述终端的模型存储空间信息。
可选的，获取模块901用于：
接收接入移动管理功能或者模型训练逻辑网元发送的第三消息，所述第三消息包括所述ML模型对应的目标网元的信息。
可选的，所述第三消息还包括如下至少一项：
第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输；
用户面传输模型能力的获取指示信息。
可选的，所述装置还包括：
第三发送模块，用于向所述接入移动管理功能或者模型训练逻辑网元发送第四消息，所述第四消息包括所述用户面会话对应所述终端的IP地址信息。
可选的，所述第四消息还包括：
所述终端的用户面传输模型的能力信息。
可选的，所述传输模块902用于：
根据所述目标网元的信息，确定传输所述ML模型对应的对端目的地址；
利用所述用户面会话或者隧道与所述对端目的地址之间进行模型传输。
可选的，所述根据所述目标网元的信息，确定传输所述ML模型对应的对端目的地址，包括以下至少一项：
根据所述目标网元的FQDN，确定所述目标网元的IP地址，所述终端将所述目标网元的IP地址作为所述对端目的地址；
将所述目标网元的地址作为所述对端目的地址；
根据所述目标网元存储所述ML模型的URL，确定所述对端目的地址。
上述信息传输装置可以提高终端的传输模型的性能。
本申请实施例中的信息传输装置可以是电子设备，例如具有操作系统的电子设备，也可以是电子设备中的部件，例如集成电路或芯片。例如：该电子设备可以是终端，也可以为除终端之外的其他设备。示例性的，终端可以包括但不限于本申请实施例所列举的终端 的类型，其他设备可以为服务器、网络附属存储器(Network Attached Storage，NAS)等，本申请实施例不作具体限定。
本申请实施例提供的信息传输装置能够实现图3所示的方法实施例实现的各个过程，并达到相同的技术效果，为避免重复，这里不再赘述。
请参见图10，图10是本申请实施例提供的一种信息传输装置的结构图，如图10所示，信息传输装置1000包括：
传输模块1001，用于利用用户面会话或者隧道与终端之间进行模型传输，其中，所述装置对应的目标网元支持用户面传输模型的功能，所述隧道为基于用户面会话建立的所述目标网元和终端之间的隧道，所述模型传输包括如下至少一项：
向所述终端发送所述ML模型的信息；
接收所述终端发送的所述ML模型的信息。
上述装置对应的目标网元可以是，上述装置所属的目标网元，或者，上述装置为目标网元。
可选的，所述装置还包括：
获取模块，用于获取如下至少一项：
所述用户获取模块面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息；其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
可选的，获取模块用于接收接入移动管理功能或者模型训练逻辑网元发送的如下至少一项：
所述用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息；其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
可选的，传输模块1001用于：
根据所述终端的IP地址信息，利用用户面会话或者隧道与终端之间进行模型传输。
可选的，所述ML模型的信息包括如下至少一项：
所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、所述ML模型的模型结构信息、所述ML模型的模型参数信息。
可选的，所述ML模型的要求信息包括如下至少一项：
模型传输时延要求信息、模型大小要求信息、用于指示模型需支持共享的共享指示信息、用于限定模型提供网元身份的身份限定信息、模型表述方式限定信息、模型性能要求信息、模型使用范围要求信息。
上述信息传输装置可以提高终端的传输模型的性能。
本申请实施例中的信息传输装置可以是电子设备，例如具有操作系统的电子设备，也可以是电子设备中的部件，例如集成电路或芯片。该电子设备可以是终端或网络侧设备。
本申请实施例提供的信息传输装置能够实现图4所示的方法实施例实现的各个过程，并达到相同的技术效果，为避免重复，这里不再赘述。
请参见图11，图11是本申请实施例提供的一种信息传输装置的结构图，如图11所示，信息传输装置1100包括：
获取模块1101，用于获取机器学习ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；
第一发送模块1102，用于向终端发送所述目标网元的信息。
可选的，所述目标网元的信息包括如下至少一项：
所述目标网元的全量域名FQDN；
所述目标网元的地址信息；
所述目标网元存储所述ML模型的统一资源定位符URL；
所述目标网元传输所述ML模型的数据网络名称DNN；
所述目标网元传输所述ML模型的单一网络切片选择辅助信息S-NSSAI；
所述目标网元传输所述ML模型的无线接入技术RAT类型；
所述目标网元传输所述ML模型的接入类型；
所述目标网元传输所述ML模型的用户面隧道相关的安全信息。
可选的，所述装置还包括：
第一接收模块，用于接收所述终端发送的第一消息，所述第一消息用于请求获取所述ML模型，或者，所述第一消息用于请求上传所述ML模型。
可选的，所述第一消息包括如下至少一项：
所述用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
可选的，获取模块1101用于：
能根据所述第一消息，获取所述ML模型对应的目标网元的信息。
可选的，获取模块1101用于：
向网络存储功能发送用于确定所述ML模型的对应的目标网元的发现消息，所述发现消息包括所述第一消息携带的如下至少一项：所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、用户面传输模型的指示信息；
接收所述网络存储功能发送的所述目标网元的信息。
可选的，所述装置还包括：
确定模块，用于根据所述第一消息，确定所述ML模型的模型提供网元。
可选的，所述确定模块用于：
向网络存储功能发送用于确定所述ML模型的模型提供网元的发现消息，所述发现消息包括所述第一消息携带的如下至少一项：
所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、用户面传输模型的指示信息；
接收所述网络存储功能发送的所述模型提供网元的信息。
可选的，所述模型提供网元包括：模型训练逻辑网元或模型训练应用功能，所述装置还包括：
第二发送模块，用于向所述模型训练逻辑网元或模型训练应用功能发送指示消息，所述指示信息用于指示所述ML模型；
获取模块1101用于：
接收所述模型训练逻辑网元或模型训练应用功能发送响应消息，所述响应消息包括所述ML模型对应的目标网元的信息。
可选的，所述指示信息包括如下至少一项：
所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息，用户面传输模型的指示信息、分析标识、所述ML模型的通知目标地址；
其中，所述分析标识用于指示所述ML模型用于所述分析标识对应的数据分析任务；
所述ML模型的通知目标地址为空，或者，所述ML模型的通知目标地址为用户面会话对应所述终端的IP地址信息，或者，所述通知目标地址为所述接入移动管理功能的地址信息。
可选的，所述响应消息还包括：
第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
可选的，获取模块1101用于：
从网络存储功能获取所述ML模型对应的目标网元的信息。
可选的，所述装置还包括：
第二接收模块，用于接收所述终端发送的第二消息，所述第二消息用于上报或者更新所述终端的模型能力信息。
可选的，所述终端的模型能力信息包括如下至少一项：
所述终端支持的模型的标识信息；
所述终端支持的模型的功能信息；
所述终端支持的模型的类型信息；
所述终端支持模型接收的指示信息；
所述终端支持模型发送的指示信息；
所述终端通过用户面传输模型的能力信息；
所述终端的模型存储空间信息。
可选的，所述第一发送模块1102用于：
向所述终端发送第三消息，所述第三消息包括所述ML模型对应的目标网元的信息，以及还包括如下至少一项：
第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输；
用户面传输模型能力的获取指示信息。
可选的，所述装置还包括：
第三接收模块，用于接收所述终端发送的第四消息，所述第四消息包括用户面会话对应所述终端的IP地址信息。
可选的，所述第四消息还包括：
所述终端的用户面传输模型的能力信息；
所述装置还包括：
第三发送模块，用于向模型训练逻辑网元或者所述目标网元发送第五消息，所述第五消息包括如下至少一项：
所述用户面会话对应所述终端的IP地址信息；
所述终端的用户面传输模型的能力信息。
上述信息传输装置可以提高终端的传输模型的性能。
本申请实施例中的信息传输装置可以是电子设备，例如具有操作系统的电子设备，也可以是电子设备中的部件，例如集成电路或芯片。该电子设备可以是终端或网络侧设备。
本申请实施例提供的信息传输装置能够实现图5所示的方法实施例实现的各个过程，并达到相同的技术效果，为避免重复，这里不再赘述。
请参见图12，图12是本申请实施例提供的一种信息传输装置的结构图，如图10所示，信息传输装置1200包括：
发送模块1201，用于向接入移动管理功能或者终端发送ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能。
可选的，所述目标网元的信息包括如下至少一项：
所述目标网元的全量域名FQDN；
所述目标网元的地址信息；
所述目标网元存储所述ML模型的统一资源定位符URL；
所述目标网元传输所述ML模型的数据网络名称DNN；
所述目标网元传输所述ML模型的单一网络切片选择辅助信息S-NSSAI；
所述目标网元传输所述ML模型的无线接入技术RAT类型；
所述目标网元传输所述ML模型的接入类型；
所述目标网元传输所述ML模型的用户面隧道相关的安全信息。
可选的，所述装置还包括：
第一接收模块，用于接收来自所述终端的第一消息，所述第一消息用于请求获取所述ML模型，或者，所述第一消息用于触发上传所述ML模型。
可选的，所述第一消息包括如下至少一项：
用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
可选的，所述装置还包括：
第二接收模块，用于接收所述接入移动管理功能发送的指示消息，所述指示信息用于指示所述ML模型；
发送模块1201用于：
向所述接入移动管理功能发送响应消息，所述响应消息包括所述ML模型对应的目标网元的信息。
可选的，所述指示信息包括如下至少一项：
所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息，用户面传输模型的指示信息、分析标识、所述ML模型的通知目标地址；
其中，所述分析标识用于指示所述ML模型用于所述分析标识对应的数据分析任务；
所述ML模型的通知目标地址为空，或者，所述ML模型的通知目标地址为用户面会话对应所述终端的IP地址信息，或者，所述通知目标地址为所述接入移动管理功能的地址信息。
可选的，所述装置还包括：
第一确定模块，用于根据所述第一消息或所述指示信息确定所述目标网元的信息。
可选的，所述第一确定模块用于：
向网络存储功能发送用于确定所述ML模型的对应的目标网元的发现消息，所述发现消息包括如下至少一项：所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、用户面传输模型的指示信息；
接收所述网络存储功能发送的所述目标网元的信息。
可选的，所述第一确定模块用于：
向模型存储网元发送信息请求消息，所述信息请求消息用于请求所述模型存储网元反馈所述ML模型对应的目标网元的信息；
接收所述模型存储网元发送所述ML模型对应的目标网元的信息，所述目标网元的信息包括所述模型存储网元存储所述ML模型的URL。
可选的，所述响应消息还包括：
第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
可选的，所述装置还包括如下至少一项：
第二确定模块，用于确定使用用户面方式传输所述ML模型；
获取模块，用于获取所述目标网元的信息；
第三接收模块，用于接收所述接入移动管理功能向发送的第五消息，所述第五消息包括如下至少一项：
用户面会话对应所述终端的IP地址信息；
所述终端的用户面传输模型的能力信息。
上述信息传输装置可以提高终端的传输模型的性能。
本申请实施例中的信息传输装置可以是电子设备，例如具有操作系统的电子设备，也可以是电子设备中的部件，例如集成电路或芯片。该电子设备可以是终端或网络侧设备。
本申请实施例提供的信息传输装置能够实现图6所示的方法实施例实现的各个过程，并达到相同的技术效果，为避免重复，这里不再赘述。
可选的，如图13所示，本申请实施例还提供一种通信设备1300，包括处理器1301和存储器1302，存储器1302上存储有可在所述处理器1301上运行的程序或指令，例如，该通信设备1300为终端时，该程序或指令被处理器1301执行时实现上述终端侧的资源分配方法实施例的各个步骤，且能达到相同的技术效果。该通信设备1300为网络侧设备时，该程序或指令被处理器1301执行时实现上述网络侧的资源分配方法实施例的各个步骤，且能达到相同的技术效果，为避免重复，这里不再赘述。
本申请实施例还提供一种通信设备，包括处理器及通信接口，其中，所述通信接口用于获取机器学习ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；传输模块，用于利用用户面会话或者隧道与所述目标网元之间进行模型传输，其中，所述隧道为基于用户面会话建立的终端和所述目标网元之间的隧道，所述模型传输包括如下至少一项：从所述目标网元接收所述ML模型的信息；向所述目标网元发送所述ML模型的信息。该通信设备实施例与上述信息传输方法实施例对应，上述方法实施例的各个实施过程和实现方式均可适用于该通信设备实施例中，且能达到相同的技术效果。
具体地，图14为实现本申请实施例的一种终端的硬件结构示意图。
该终端1400包括但不限于：射频单元1401、网络模块1402、音频输出单元1403、输入单元1404、传感器1405、显示单元1406、用户输入单元1407、接口单元1408、存储器1409以及处理器1410等中的至少部分部件。
本领域技术人员可以理解，终端1400还可以包括给各个部件供电的电源(比如电池)，电源可以通过电源管理系统与处理器1410逻辑相连，从而通过电源管理系统实现管理充电、放电、以及功耗管理等功能。图14中示出的终端结构并不构成对终端的限定，终端 可以包括比图示更多或更少的部件，或者组合某些部件，或者不同的部件布置，在此不再赘述。
应理解的是，本申请实施例中，输入单元1404可以包括图形处理单元(Graphics Processing Unit，GPU)14041和麦克风14042，图形处理单元14041对在视频捕获模式或图像捕获模式中由图像捕获装置(如摄像头)获得的静态图片或视频的图像数据进行处理。显示单元1406可包括显示面板14061，可以采用液晶显示器、有机发光二极管等形式来配置显示面板14061。用户输入单元1407包括触控面板14071以及其他输入设备14072中的至少一种。触控面板14071，也称为触摸屏。触控面板14071可包括触摸检测装置和触摸控制器两个部分。其他输入设备14072可以包括但不限于物理键盘、功能键(比如音量控制按键、开关按键等)、轨迹球、鼠标、操作杆，在此不再赘述。
本申请实施例中，射频单元1401接收来自网络侧设备的下行数据后，可以传输给处理器1410进行处理；另外，射频单元1401可以向网络侧设备发送上行数据。通常，射频单元1401包括但不限于天线、放大器、收发信机、耦合器、低噪声放大器、双工器等。
存储器1409可用于存储软件程序或指令以及各种数据。存储器1409可主要包括存储程序或指令的第一存储区和存储数据的第二存储区，其中，第一存储区可存储操作系统、至少一个功能所需的应用程序或指令(比如声音播放功能、图像播放功能等)等。此外，存储器1409可以包括易失性存储器或非易失性存储器，或者，存储器1409可以包括易失性和非易失性存储器两者。其中，非易失性存储器可以是只读存储器(Read-Only Memory，ROM)、可编程只读存储器(Programmable ROM，PROM)、可擦除可编程只读存储器(Erasable PROM，EPROM)、电可擦除可编程只读存储器(Electrically EPROM，EEPROM)或闪存。易失性存储器可以是随机存取存储器(Random Access Memory，RAM)，静态随机存取存储器(Static RAM，SRAM)、动态随机存取存储器(Dynamic RAM，DRAM)、同步动态随机存取存储器(Synchronous DRAM，SDRAM)、双倍数据速率同步动态随机存取存储器(Double Data Rate SDRAM，DDRSDRAM)、增强型同步动态随机存取存储器(Enhanced SDRAM，ESDRAM)、同步连接动态随机存取存储器(Synch link DRAM，SLDRAM)和直接内存总线随机存取存储器(Direct Rambus RAM，DRRAM)。本申请实施例中的存储器1409包括但不限于这些和任意其它适合类型的存储器。
处理器1410可包括一个或多个处理单元；可选的，处理器1410集成应用处理器和调制解调处理器，其中，应用处理器主要处理涉及操作系统、用户界面和应用程序等的操作，调制解调处理器主要处理无线通信信号，如基带处理器。可以理解的是，上述调制解调处理器也可以不集成到处理器1410中。
其中，射频单元1401用于：
获取机器学习ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；
利用用户面会话或者隧道与所述目标网元之间进行模型传输，其中，所述隧道为基于 用户面会话建立的所述终端和所述目标网元之间的隧道，所述模型传输包括如下至少一项：
从所述目标网元接收所述ML模型的信息；
向所述目标网元发送所述ML模型的信息。
可选的，所述ML模型的信息包括如下至少一项：
所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、所述ML模型的模型结构信息、所述ML模型的模型参数信息。
可选的，所述ML模型的要求信息包括如下至少一项：
模型传输时延要求信息、模型大小要求信息、用于指示模型需支持共享的共享指示信息、用于限定模型提供网元身份的身份限定信息、模型表述方式限定信息、模型性能要求信息、模型使用范围要求信息。
可选的，述目标网元的信息包括如下至少一项：
所述目标网元的全量域名FQDN；
所述目标网元的地址信息；
所述目标网元存储所述ML模型的统一资源定位符URL；
所述目标网元传输所述ML模型的数据网络名称DNN；
所述目标网元传输所述ML模型的单一网络切片选择辅助信息S-NSSAI；
所述目标网元传输所述ML模型的无线接入技术RAT类型；
所述目标网元传输所述ML模型的接入类型；
所述目标网元传输所述ML模型的用户面隧道相关的安全信息。
可选的，射频单元1401还用于：
向接入移动管理功能或者模型训练逻辑网元发送第一消息，所述第一消息用于请求获取所述ML模型，或者，所述第一消息用于请求上传所述ML模型。
可选的，所述第一消息包括如下至少一项：
所述用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
可选的，射频单元1401还用于：
向接入移动管理功能或者模型训练逻辑网元发送第二消息，所述第二消息用于上报或者更新所述终端的模型能力信息。
可选的，所述终端的模型能力信息包括如下至少一项：
所述终端支持的模型的标识信息；
所述终端支持的模型的功能信息；
所述终端支持的模型的类型信息；
所述终端支持模型接收的指示信息；
所述终端支持模型发送的指示信息；
所述终端通过用户面传输模型的能力信息；
所述终端的模型存储空间信息。
可选的，所述获取机器学习ML模型对应的目标网元的信息，包括：
接收接入移动管理功能或者模型训练逻辑网元发送的第三消息，所述第三消息包括所述ML模型对应的目标网元的信息。
可选的，所述第三消息还包括如下至少一项：
第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输；
用户面传输模型能力的获取指示信息。
可选的，射频单元1401还用于：
向所述接入移动管理功能或者模型训练逻辑网元发送第四消息，所述第四消息包括所述用户面会话对应所述终端的IP地址信息。
可选的，所述第四消息还包括：
所述终端的用户面传输模型的能力信息。
可选的，所述利用用户面会话或者隧道与所述目标网元之间进行模型传输，包括：
根据所述目标网元的信息，确定传输所述ML模型对应的对端目的地址；
利用所述用户面会话或者隧道与所述对端目的地址之间进行模型传输。
可选的，所述根据所述目标网元的信息，确定传输所述ML模型对应的对端目的地址，包括以下至少一项：
根据所述目标网元的FQDN，确定所述目标网元的IP地址，所述终端将所述目标网元的IP地址作为所述对端目的地址；
将所述目标网元的地址作为所述对端目的地址；
根据所述目标网元存储所述ML模型的URL，确定所述对端目的地址。
上述终端可以提高终端的数据传输性能。
本申请实施例还提供一种通信设备，包括处理器及通信接口，其中，所述通信接口用于利用用户面会话或者隧道与终端之间进行模型传输，其中，所述目标网元支持用户面传输模型的功能，所述隧道为基于用户面会话建立的所述目标网元和终端之间的隧道，所述模型传输包括如下至少一项：向所述终端发送所述ML模型的信息；接收所述终端发送的所述ML模型的信息。或者，所述通信接口用于获取机器学习ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；向终端发送所述目标网元的信息。或者，所述通信接口用于向接入移动管理功能或者终端发送ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能。该通信设备实施例与上述信息传输方法实施例对应，上述方法实施例的各个实施过程和实现方式均可适用于该通信设备实施例中，且能达到相同的技术效果。
具体地，本申请实施例还提供了一种网络侧设备。如图15所示，该网络侧设备1500包括：处理器1501、网络接口1502和存储器1503。其中，网络接口1502例如为通用公共无线接口(common public radio interface，CPRI)。
具体地，本发明实施例的网络侧设备1500还包括：存储在存储器1503上并可在处理器1501上运行的指令或程序，处理器1501调用存储器1503中的指令或程序执行图10至12中任一项所示各模块执行的方法，并达到相同的技术效果，为避免重复，故不在此赘述。
在上述网络侧设备为目标网元的实施例中：
网络接口1502用于利用用户面会话或者隧道与终端之间进行模型传输，其中，所述目标网元支持用户面传输模型的功能，所述隧道为基于用户面会话建立的所述目标网元和终端之间的隧道，所述模型传输包括如下至少一项：
向所述终端发送所述ML模型的信息；
接收所述终端发送的所述ML模型的信息。
可选的，网络接口1502还用于：
获取如下至少一项：
所述用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息；其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
可选的，网络接口1502用于接收接入移动管理功能或者模型训练逻辑网元发送的如下至少一项：
所述用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息；其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
可选的，所述利用用户面会话或者隧道与终端之间进行模型传输，包括：
根据所述终端的IP地址信息，利用用户面会话或者隧道与终端之间进行模型传输。
可选的，所述ML模型的信息包括如下至少一项：
所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、所述ML模型的模型结构信息、所述ML模型的模型参数信息。
可选的，所述ML模型的要求信息包括如下至少一项：
模型传输时延要求信息、模型大小要求信息、用于指示模型需支持共享的共享指示信息、用于限定模型提供网元身份的身份限定信息、模型表述方式限定信息、模型性能要求信息、模型使用范围要求信息。
在上述网络侧设备为接入移动管理功能的实施例中：
网络接口1502用于：获取机器学习ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；向终端发送所述目标网元的信息。
可选的，所述目标网元的信息包括如下至少一项：
所述目标网元的全量域名FQDN；
所述目标网元的地址信息；
所述目标网元存储所述ML模型的统一资源定位符URL；
所述目标网元传输所述ML模型的数据网络名称DNN；
所述目标网元传输所述ML模型的单一网络切片选择辅助信息S-NSSAI；
所述目标网元传输所述ML模型的无线接入技术RAT类型；
所述目标网元传输所述ML模型的接入类型；
所述目标网元传输所述ML模型的用户面隧道相关的安全信息。
可选的，网络接口1502还用于：
接收所述终端发送的第一消息，所述第一消息用于请求获取所述ML模型，或者，所述第一消息用于请求上传所述ML模型。
可选的，所述第一消息包括如下至少一项：
所述用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
可选的，所述获取ML模型对应的目标网元的信息，包括：
根据所述第一消息，获取所述ML模型对应的目标网元的信息。
可选的，所述根据所述第一消息获取所述ML模型对应的目标网元的信息，方法包括：
向网络存储功能发送用于确定所述ML模型的对应的目标网元的发现消息，所述发现消息包括所述第一消息携带的如下至少一项：所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、用户面传输模型的指示信息；
接收所述网络存储功能发送的所述目标网元的信息。
可选的，网络接口1502还用于：
根据所述第一消息，确定所述ML模型的模型提供网元。
可选的，所述根据所述第一消息，确定所述ML模型的模型提供网元，包括：
向网络存储功能发送用于确定所述ML模型的模型提供网元的发现消息，所述发现消息包括所述第一消息携带的如下至少一项：
所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、用户面传输模型的指示信息；
所述接入移动管理功能接收所述网络存储功能发送的所述模型提供网元的信息。
可选的，所述模型提供网元包括：模型训练逻辑网元或模型训练应用功能，网络接口 1502还用于：
向所述模型训练逻辑网元或模型训练应用功能发送指示消息，所述指示信息用于指示所述ML模型；
所述获取ML模型对应的目标网元的信息，包括：
接收所述模型训练逻辑网元或模型训练应用功能发送响应消息，所述响应消息包括所述ML模型对应的目标网元的信息。
可选的，所述指示信息包括如下至少一项：
所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息，用户面传输模型的指示信息、分析标识、所述ML模型的通知目标地址；
其中，所述分析标识用于指示所述ML模型用于所述分析标识对应的数据分析任务；
所述ML模型的通知目标地址为空，或者，所述ML模型的通知目标地址为用户面会话对应所述终端的IP地址信息，或者，所述通知目标地址为所述接入移动管理功能的地址信息。
可选的，所述响应消息还包括：
第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
可选的，所述获取ML模型对应的目标网元的信息，包括：
从网络存储功能获取所述ML模型对应的目标网元的信息。
可选的，网络接口1502还用于：
接收所述终端发送的第二消息，所述第二消息用于上报或者更新所述终端的模型能力信息。
可选的，所述终端的模型能力信息包括如下至少一项：
所述终端支持的模型的标识信息；
所述终端支持的模型的功能信息；
所述终端支持的模型的类型信息；
所述终端支持模型接收的指示信息；
所述终端支持模型发送的指示信息；
所述终端通过用户面传输模型的能力信息；
所述终端的模型存储空间信息。
可选的，所述向终端发送所述目标网元的信息，包括：
向所述终端发送第三消息，所述第三消息包括所述ML模型对应的目标网元的信息，以及还包括如下至少一项：
第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输；
用户面传输模型能力的获取指示信息。
可选的，网络接口1502还用于：
接收所述终端发送的第四消息，所述第四消息包括用户面会话对应所述终端的IP地址信息。
可选的，所述第四消息还包括：
所述终端的用户面传输模型的能力信息；
网络接口1502还用于：
向模型训练逻辑网元或者所述目标网元发送第五消息，所述第五消息包括如下至少一项：
所述用户面会话对应所述终端的IP地址信息；
所述终端的用户面传输模型的能力信息。
在上述网络侧设备为目标网元的实施例中：
网络接口1502用于：向接入移动管理功能或者终端发送ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能。
可选的，所述目标网元的信息包括如下至少一项：
所述目标网元的全量域名FQDN；
所述目标网元的地址信息；
所述目标网元存储所述ML模型的统一资源定位符URL；
所述目标网元传输所述ML模型的数据网络名称DNN；
所述目标网元传输所述ML模型的单一网络切片选择辅助信息S-NSSAI；
所述目标网元传输所述ML模型的无线接入技术RAT类型；
所述目标网元传输所述ML模型的接入类型；
所述目标网元传输所述ML模型的用户面隧道相关的安全信息。
可选的，网络接口1502还用于：
接收来自所述终端的第一消息，所述第一消息用于请求获取所述ML模型，或者，所述第一消息用于触发上传所述ML模型。
可选的，所述第一消息包括如下至少一项：
用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
可选的，网络接口1502还用于：
接收所述接入移动管理功能发送的指示消息，所述指示信息用于指示所述ML模型；
所述向接入移动管理功能发送ML模型对应的目标网元的信息，包括：
向所述接入移动管理功能发送响应消息，所述响应消息包括所述ML模型对应的目标 网元的信息。
可选的，所述指示信息包括如下至少一项：
所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息，用户面传输模型的指示信息、分析标识、所述ML模型的通知目标地址；
其中，所述分析标识用于指示所述ML模型用于所述分析标识对应的数据分析任务；
所述ML模型的通知目标地址为空，或者，所述ML模型的通知目标地址为用户面会话对应所述终端的IP地址信息，或者，所述通知目标地址为所述接入移动管理功能的地址信息。
可选的，网络接口1502还用于：
根据所述第一消息或所述指示信息确定所述目标网元的信息。
可选的，所述根据所述第一消息或所述指示信息确定所述目标网元的信息，包括：
向网络存储功能发送用于确定所述ML模型的对应的目标网元的发现消息，所述发现消息包括如下至少一项：所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、用户面传输模型的指示信息；
所述模型训练逻辑网元接收所述网络存储功能发送的所述目标网元的信息。
可选的，所述根据所述第一消息或所述指示信息确定所述目标网元的信息，包括：
向模型存储网元发送信息请求消息，所述信息请求消息用于请求所述模型存储网元反馈所述ML模型对应的目标网元的信息；
接收所述模型存储网元发送所述ML模型对应的目标网元的信息，所述目标网元的信息包括所述模型存储网元存储所述ML模型的URL。
可选的，所述响应消息还包括：
第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
可选的，网络接口1502还用于如下至少一项：
确定使用用户面方式传输所述ML模型；
获取所述目标网元的信息；
接收所述接入移动管理功能向发送的第五消息，所述第五消息包括如下至少一项：
用户面会话对应所述终端的IP地址信息；
所述终端的用户面传输模型的能力信息。
本申请实施例还提供了一种可读存储介质，所述可读存储介质上存储程序或指令，所述程序或指令被处理器执行时实现如本申请实施例提供的终端侧的信息传输方法的步骤，或者实现如本申请实施例提供的目标网元侧的信息传输方法的步骤，或者实现如本申请实施例提供的接入移动管理功能侧的信息传输方法的步骤，或者实现如本申请实施例提供的模型训练逻辑网元侧的信息传输方法的步骤。
其中，所述处理器为上述实施例中所述的终端中的处理器。所述可读存储介质，包括计算机可读存储介质，如计算机只读存储器ROM、随机存取存储器RAM、磁碟或者光盘等。
本申请实施例另提供了一种芯片，所述芯片包括处理器和通信接口，所述通信接口和所述处理器耦合，所述处理器用于运行程序或指令，实现如本申请实施例提供的终端侧的信息传输方法，或实现如本申请实施例提供的目标网元侧的信息传输方法，或实现如本申请实施例提供的接入移动管理功能侧的信息传输方法，或实现如本申请实施例提供的模型训练逻辑网元侧的信息传输方法。
应理解，本申请实施例提到的芯片还可以称为系统级芯片，系统芯片，芯片系统或片上系统芯片等。
本申请实施例另提供了一种计算机程序/程序产品，所述计算机程序/程序产品被存储在存储介质中，所述计算机程序/程序产品被至少一个处理器执行以实现如本申请实施例提供的终端侧的信息传输方法的步骤，或所述计算机程序/程序产品被至少一个处理器执行以实现如本申请实施例提供的目标网元侧的信息传输方法的步骤，或所述计算机程序/程序产品被至少一个处理器执行以实现如本申请实施例提供的接入移动管理功能侧的信息传输方法的步骤，或所述计算机程序/程序产品被至少一个处理器执行以实现如本申请实施例提供的模型训练逻辑网元侧的信息传输方法的步骤。
本申请实施例还提供了一种信息传输系统，包括：终端、目标网元、接入移动管理功能和模型训练逻辑网元，所述终端可用于执行如本申请实施例提供的终端侧的信息传输方法的步骤，所述目标网元可用于执行如本申请实施例提供的目标网元侧的信息传输方法的步骤，所述接入移动管理功能可用于执行如本申请实施例提供的接入移动管理功能侧的信息传输方法的步骤，所述模型训练逻辑网元可用于执行如本申请实施例提供的模型训练逻辑网元侧的信息传输方法的步骤。
需要说明的是，在本文中，术语“包括”、“包含”或者其任何其他变体意在涵盖非排他性的包含，从而使得包括一系列要素的过程、方法、物品或者装置不仅包括那些要素，而且还包括没有明确列出的其他要素，或者是还包括为这种过程、方法、物品或者装置所固有的要素。在没有更多限制的情况下，由语句“包括一个……”限定的要素，并不排除在包括该要素的过程、方法、物品或者装置中还存在另外的相同要素。此外，需要指出的是，本申请实施方式中的方法和装置的范围不限按示出或讨论的顺序来执行功能，还可包括根据所涉及的功能按基本同时的方式或按相反的顺序来执行功能，例如，可以按不同于所描述的次序来执行所描述的方法，并且还可以添加、省去、或组合各种步骤。另外，参照某些示例所描述的特征可在其他示例中被组合。
通过以上的实施方式的描述，本领域的技术人员可以清楚地了解到上述实施例方法可借助软件加必需的通用硬件平台的方式来实现，当然也可以通过硬件，但很多情况下前者是更佳的实施方式。基于这样的理解，本申请的技术方案本质上或者说对现有技术做出贡 献的部分可以以计算机软件产品的形式体现出来，该计算机软件产品存储在一个存储介质(如ROM/RAM、磁碟、光盘)中，包括若干指令用以使得一台终端(可以是手机，计算机，服务器，空调器，或者网络设备等)执行本申请各个实施例所述的方法。
上面结合附图对本申请的实施例进行了描述，但是本申请并不局限于上述的具体实施方式，上述的具体实施方式仅仅是示意性的，而不是限制性的，本领域的普通技术人员在本申请的启示下，在不脱离本申请宗旨和权利要求所保护的范围情况下，还可做出很多形式，均属于本申请的保护之内。

Claims (57)

  1. 一种信息传输方法，包括：
    终端获取机器学习ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；
    所述终端利用用户面会话或者隧道与所述目标网元之间进行模型传输，其中，所述隧道为基于用户面会话建立的所述终端和所述目标网元之间的隧道，所述模型传输包括如下至少一项：
    从所述目标网元接收所述ML模型的信息；
    向所述目标网元发送所述ML模型的信息。
  2. 如权利要求1所述的方法，其中，所述ML模型的信息包括如下至少一项：
    所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、所述ML模型的模型结构信息、所述ML模型的模型参数信息。
  3. 如权利要求2所述的方法，其中，所述ML模型的要求信息包括如下至少一项：
    模型传输时延要求信息、模型大小要求信息、用于指示模型需支持共享的共享指示信息、用于限定模型提供网元身份的身份限定信息、模型表述方式限定信息、模型性能要求信息、模型使用范围要求信息。
  4. 如权利要求1所述的方法，其中，所述目标网元的信息包括如下至少一项：
    所述目标网元的全量域名FQDN；
    所述目标网元的地址信息；
    所述目标网元存储所述ML模型的统一资源定位符URL；
    所述目标网元传输所述ML模型的数据网络名称DNN；
    所述目标网元传输所述ML模型的单一网络切片选择辅助信息S-NSSAI；
    所述目标网元传输所述ML模型的无线接入技术RAT类型；
    所述目标网元传输所述ML模型的接入类型；
    所述目标网元传输所述ML模型的用户面隧道相关的安全信息。
  5. 如权利要求1所述的方法，其中，所述方法还包括：
    所述终端向接入移动管理功能或者模型训练逻辑网元发送第一消息，所述第一消息用于请求获取所述ML模型，或者，所述第一消息用于请求上传所述ML模型。
  6. 如权利要求5所述的方法，其中，所述第一消息包括如下至少一项：
    所述用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
  7. 如权利要求1至6中任一项所述的方法，其中，所述方法还包括：
    所述终端向接入移动管理功能或者模型训练逻辑网元发送第二消息，所述第二消息用于上报或者更新所述终端的模型能力信息。
  8. 如权利要求7所述的方法，其中，所述终端的模型能力信息包括如下至少一项：
    所述终端支持的模型的标识信息；
    所述终端支持的模型的功能信息；
    所述终端支持的模型的类型信息；
    所述终端支持模型接收的指示信息；
    所述终端支持模型发送的指示信息；
    所述终端通过用户面传输模型的能力信息；
    所述终端的模型存储空间信息。
  9. 如权利要求1至6中任一项所述的方法，其中，所述终端获取机器学习ML模型对应的目标网元的信息，包括：
    所述终端接收接入移动管理功能或者模型训练逻辑网元发送的第三消息，所述第三消息包括所述ML模型对应的目标网元的信息。
  10. 如权利要求9所述的方法，其中，所述第三消息还包括如下至少一项：
    第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输；
    用户面传输模型能力的获取指示信息。
  11. 如权利要求9所述的方法，其中，所述方法还包括：
    所述终端向所述接入移动管理功能或者模型训练逻辑网元发送第四消息，所述第四消息包括所述用户面会话对应所述终端的IP地址信息。
  12. 如权利要求11所述的方法，其中，所述第四消息还包括：
    所述终端的用户面传输模型的能力信息。
  13. 如权利要求1至6中任一项所述的方法，其中，所述终端利用用户面会话或者隧道与所述目标网元之间进行模型传输，包括：
    所述终端根据所述目标网元的信息，确定传输所述ML模型对应的对端目的地址；
    所述终端利用所述用户面会话或者隧道与所述对端目的地址之间进行模型传输。
  14. 如权利要求13所述的方法，其中，所述终端根据所述目标网元的信息，确定传输所述ML模型对应的对端目的地址，包括以下至少一项：
    所述终端根据所述目标网元的FQDN，确定所述目标网元的IP地址，所述终端将所述目标网元的IP地址作为所述对端目的地址；
    所述终端将所述目标网元的地址作为所述对端目的地址；
    所述终端根据所述目标网元存储所述ML模型的URL，确定所述对端目的地址。
  15. 一种信息传输方法，包括：
    目标网元利用用户面会话或者隧道与终端之间进行模型传输，其中，所述目标网元支持用户面传输模型的功能，所述隧道为基于用户面会话建立的所述目标网元和终端之间的隧道，所述模型传输包括如下至少一项：
    向所述终端发送ML模型的信息；
    接收所述终端发送的所述ML模型的信息。
  16. 如权利要求15所述的方法，其中，所述方法还包括：
    所述目标网元获取如下至少一项：
    所述用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息；其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
  17. 如权利要求16所述的方法，其中，所述目标网元接收接入移动管理功能或者模型训练逻辑网元发送的如下至少一项：
    所述用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息；其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
  18. 如权利要求16所述的方法，其中，所述目标网元利用用户面会话或者隧道与终端之间进行模型传输，包括：
    所述目标网元根据所述终端的IP地址信息，利用用户面会话或者隧道与终端之间进行模型传输。
  19. 如权利要求15所述的方法，其中，所述ML模型的信息包括如下至少一项：
    所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、所述ML模型的模型结构信息、所述ML模型的模型参数信息。
  20. 如权利要求19所述的方法，其中，所述ML模型的要求信息包括如下至少一项：
    模型传输时延要求信息、模型大小要求信息、用于指示模型需支持共享的共享指示信息、用于限定模型提供网元身份的身份限定信息、模型表述方式限定信息、模型性能要求信息、模型使用范围要求信息。
  21. 一种信息传输方法，包括：
    接入移动管理功能获取机器学习ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；
    所述接入移动管理功能向终端发送所述目标网元的信息。
  22. 如权利要求21所述的方法，其中，所述目标网元的信息包括如下至少一项：
    所述目标网元的全量域名FQDN；
    所述目标网元的地址信息；
    所述目标网元存储所述ML模型的统一资源定位符URL；
    所述目标网元传输所述ML模型的数据网络名称DNN；
    所述目标网元传输所述ML模型的单一网络切片选择辅助信息S-NSSAI；
    所述目标网元传输所述ML模型的无线接入技术RAT类型；
    所述目标网元传输所述ML模型的接入类型；
    所述目标网元传输所述ML模型的用户面隧道相关的安全信息。
  23. 如权利要求21所述的方法，其中，所述方法还包括：
    所述接入移动管理功能接收所述终端发送的第一消息，所述第一消息用于请求获取所述ML模型，或者，所述第一消息用于请求上传所述ML模型。
  24. 如权利要求23所述的方法，其中，所述第一消息包括如下至少一项：
    所述用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
  25. 如权利要求23所述的方法，其中，所述接入移动管理功能获取ML模型对应的目标网元的信息，包括：
    所述接入移动管理功能根据所述第一消息，获取所述ML模型对应的目标网元的信息。
  26. 如权利要求25所述的方法，其中，所述接入移动管理功能根据所述第一消息获取所述ML模型对应的目标网元的信息，包括：
    所述接入移动管理功能向网络存储功能发送用于确定所述ML模型的对应的目标网元的发现消息，所述发现消息包括所述第一消息携带的如下至少一项：所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、用户面传输模型的指示信息；
    所述接入移动管理功能接收所述网络存储功能发送的所述目标网元的信息。
  27. 如权利要求23所述的方法，其中，所述方法还包括：
    所述接入移动管理功能根据所述第一消息，确定所述ML模型的模型提供网元。
  28. 如权利要求27所述的方法，其中，所述接入移动管理功能根据所述第一消息，确定所述ML模型的模型提供网元，包括：
    所述接入移动管理功能向网络存储功能发送用于确定所述ML模型的模型提供网元的发现消息，所述发现消息包括所述第一消息携带的如下至少一项：
    所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、用户面传输模型的指示信息；
    所述接入移动管理功能接收所述网络存储功能发送的所述模型提供网元的信息。
  29. 如权利要求28所述的方法，其中，所述模型提供网元包括：模型训练逻辑网元或模型训练应用功能，所述方法还包括：
    所述接入移动管理功能向所述模型训练逻辑网元或模型训练应用功能发送指示消息，所述指示信息用于指示所述ML模型；
    所述接入移动管理功能获取ML模型对应的目标网元的信息，包括：
    所述接入移动管理功能接收所述模型训练逻辑网元或模型训练应用功能发送响应消息，所述响应消息包括所述ML模型对应的目标网元的信息。
  30. 如权利要求29所述的方法，其中，所述指示信息包括如下至少一项：
    所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息，用户面传输模型的指示信息、分析标识、所述ML模型的通知目标地址；
    其中，所述分析标识用于指示所述ML模型用于所述分析标识对应的数据分析任务；
    所述ML模型的通知目标地址为空，或者，所述ML模型的通知目标地址为用户面会话对应所述终端的IP地址信息，或者，所述通知目标地址为所述接入移动管理功能的地址信息。
  31. 如权利要求29所述的方法，其中，所述响应消息还包括：
    第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
  32. 如权利要求23所述的方法，其中，所述接入移动管理功能获取ML模型对应的目标网元的信息，包括：
    所述接入移动管理功能从网络存储功能获取所述ML模型对应的目标网元的信息。
  33. 如权利要求21至32中任一项所述的方法，其中，所述方法还包括：
    所述接入移动管理功能接收所述终端发送的第二消息，所述第二消息用于上报或者更新所述终端的模型能力信息。
  34. 如权利要求33所述的方法，其中，所述终端的模型能力信息包括如下至少一项：
    所述终端支持的模型的标识信息；
    所述终端支持的模型的功能信息；
    所述终端支持的模型的类型信息；
    所述终端支持模型接收的指示信息；
    所述终端支持模型发送的指示信息；
    所述终端通过用户面传输模型的能力信息；
    所述终端的模型存储空间信息。
  35. 如权利要求21至32中任一项所述的方法，其中，所述接入移动管理功能向终端发送所述目标网元的信息，包括：
    所述接入移动管理功能向所述终端发送第三消息，所述第三消息包括所述ML模型对应的目标网元的信息，以及还包括如下至少一项：
    第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话 进行传输；
    用户面传输模型能力的获取指示信息。
  36. 如权利要求35所述的方法，其中，所述方法还包括：
    所述接入移动管理功能接收所述终端发送的第四消息，所述第四消息包括用户面会话对应所述终端的IP地址信息。
  37. 如权利要求36所述的方法，其中，所述第四消息还包括：
    所述终端的用户面传输模型的能力信息；
    所述方法还包括：
    所述接入移动管理功向模型训练逻辑网元或者所述目标网元发送第五消息，所述第五消息包括如下至少一项：
    所述用户面会话对应所述终端的IP地址信息；
    所述终端的用户面传输模型的能力信息。
  38. 一种信息传输方法，包括：
    模型训练逻辑网元向接入移动管理功能或者终端发送ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能。
  39. 如权利要求38所述的方法，其中，所述目标网元的信息包括如下至少一项：
    所述目标网元的全量域名FQDN；
    所述目标网元的地址信息；
    所述目标网元存储所述ML模型的统一资源定位符URL；
    所述目标网元传输所述ML模型的数据网络名称DNN；
    所述目标网元传输所述ML模型的单一网络切片选择辅助信息S-NSSAI；
    所述目标网元传输所述ML模型的无线接入技术RAT类型；
    所述目标网元传输所述ML模型的接入类型；
    所述目标网元传输所述ML模型的用户面隧道相关的安全信息。
  40. 如权利要求38所述的方法，其中，所述方法还包括：
    所述模型训练逻辑网元接收来自所述终端的第一消息，所述第一消息用于请求获取所述ML模型，或者，所述第一消息用于触发上传所述ML模型。
  41. 如权利要求40所述的方法，其中，所述第一消息包括如下至少一项：
    用户面会话对应所述终端的IP地址信息、所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息、第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
  42. 如权利要求38所述的方法，其中，所述方法还包括：
    所述模型训练逻辑网元接收所述接入移动管理功能发送的指示消息，所述指示信息用于指示所述ML模型；
    所述模型训练逻辑网元向接入移动管理功能发送ML模型对应的目标网元的信息，包括：
    所述模型训练逻辑网元向所述接入移动管理功能发送响应消息，所述响应消息包括所述ML模型对应的目标网元的信息。
  43. 如权利要求42所述的方法，其中，所述指示信息包括如下至少一项：
    所述终端的模型能力信息、所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、所述ML模型的要求信息，用户面传输模型的指示信息、分析标识、所述ML模型的通知目标地址；
    其中，所述分析标识用于指示所述ML模型用于所述分析标识对应的数据分析任务；
    所述ML模型的通知目标地址为空，或者，所述ML模型的通知目标地址为用户面会话对应所述终端的IP地址信息，或者，所述通知目标地址为所述接入移动管理功能的地址信息。
  44. 如权利要求40或42所述的方法，其中，所述方法还包括：
    所述模型训练逻辑网元根据第一消息或指示信息确定所述目标网元的信息。
  45. 如权利要求44所述的方法，其中，所述模型训练逻辑网元根据所述第一消息或所述指示信息确定所述目标网元的信息，包括：
    所述模型训练逻辑网元向网络存储功能发送用于确定所述ML模型的对应的目标网元的发现消息，所述发现消息包括如下至少一项：所述ML模型的标识信息、所述ML模型的功能信息、所述ML模型的类型信息、用户面传输模型的指示信息；
    所述模型训练逻辑网元接收所述网络存储功能发送的所述目标网元的信息。
  46. 如权利要求44所述的方法，其中，所述模型训练逻辑网元根据所述第一消息或所述指示信息确定所述目标网元的信息，包括：
    所述模型训练逻辑网元向模型存储网元发送信息请求消息，所述信息请求消息用于请求所述模型存储网元反馈所述ML模型对应的目标网元的信息；
    所述模型训练逻辑网元接收所述模型存储网元发送所述ML模型对应的目标网元的信息，所述目标网元的信息包括所述模型存储网元存储所述ML模型的URL。
  47. 如权利要求42所述的方法，其中，所述响应消息还包括：
    第一指示信息，其中，所述第一指示信息用于指示所述ML模型需要利用用户面会话进行传输。
  48. 如权利要求38至43中任一项所述的方法，其中，所述方法还包括如下至少一项：
    所述模型训练逻辑网元确定使用用户面方式传输所述ML模型；
    所述模型训练逻辑网元获取所述目标网元的信息；
    所述模型训练逻辑网元接收所述接入移动管理功能向发送的第五消息，所述第五消息包括如下至少一项：
    用户面会话对应所述终端的IP地址信息；
    所述终端的用户面传输模型的能力信息。
  49. 一种信息传输装置，包括：
    获取模块，用于获取机器学习ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；
    传输模块，用于利用用户面会话或者隧道与所述目标网元之间进行模型传输，其中，所述隧道为基于用户面会话建立的终端和所述目标网元之间的隧道，所述模型传输包括如下至少一项：
    从所述目标网元接收所述ML模型的信息；
    向所述目标网元发送所述ML模型的信息。
  50. 一种信息传输装置，包括：
    传输模块，用于利用用户面会话或者隧道与终端之间进行模型传输，其中，所述装置对应的目标网元支持用户面传输模型的功能，所述隧道为基于用户面会话建立的所述目标网元和终端之间的隧道，所述模型传输包括如下至少一项：
    向所述终端发送ML模型的信息；
    接收所述终端发送的所述ML模型的信息。
  51. 一种信息传输装置，包括：
    获取模块，用于获取机器学习ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能；
    第一发送模块，用于向终端发送所述目标网元的信息。
  52. 一种信息传输装置，包括：
    发送模块，用于向接入移动管理功能或者终端发送ML模型对应的目标网元的信息，所述目标网元支持用户面传输模型的功能。
  53. 一种终端，包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如权利要求1至14任一项所述的信息传输方法的步骤。
  54. 一种网络侧设备，所述网络侧设备为目标网元，包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如权利要求15至20任一项所述的信息传输方法的步骤。
  55. 一种网络侧设备，所述网络侧设备为接入移动管理功能，包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如权利要求21至37任一项所述的信息传输方法的步骤。
  56. 一种网络侧设备，所述网络侧设备为模型训练逻辑网元，包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如权利要求38至48任一项所述的信息传输方法的步骤。
  57. 一种可读存储介质，所述可读存储介质上存储程序或指令，所述程序或指令被处 理器执行时实现如权利要求1至14任一项所述的信息传输方法的步骤，或者实现如权利要求15至20任一项所述的信息传输方法的步骤，或者实现如权利要求21至37任一项所述的信息传输方法的步骤，或者实现如权利要求38至48任一项所述的信息传输方法的步骤。
PCT/CN2023/135261 2022-12-07 2023-11-30 信息传输方法、装置、终端及网络侧设备 Ceased WO2024120285A1 (zh)

Priority Applications (2)

Application Number Priority Date Filing Date Title
EP23899864.5A EP4633213A4 (en) 2022-12-07 2023-11-30 METHOD AND APPARATUS FOR TRANSMITTING INFORMATION, AS WELL AS TERMINAL AND NETWORK-SIDE DEVICE
US19/230,149 US20250300906A1 (en) 2022-12-07 2025-06-06 Information transmission method and apparatus, and terminal and network-side device

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211567268.4A CN118158656A (zh) 2022-12-07 2022-12-07 信息传输方法、装置、终端及网络侧设备
CN202211567268.4 2022-12-07

Related Child Applications (1)

Application Number Title Priority Date Filing Date
US19/230,149 Continuation US20250300906A1 (en) 2022-12-07 2025-06-06 Information transmission method and apparatus, and terminal and network-side device

Publications (1)

Publication Number Publication Date
WO2024120285A1 true WO2024120285A1 (zh) 2024-06-13

Family

ID=91297365

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/135261 Ceased WO2024120285A1 (zh) 2022-12-07 2023-11-30 信息传输方法、装置、终端及网络侧设备

Country Status (4)

Country Link
US (1) US20250300906A1 (zh)
EP (1) EP4633213A4 (zh)
CN (1) CN118158656A (zh)
WO (1) WO2024120285A1 (zh)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2026079771A1 (ko) * 2024-10-08 2026-04-16 엘지전자 주식회사 인공 지능에 관련된 통신

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022021421A1 (zh) * 2020-07-31 2022-02-03 Oppo广东移动通信有限公司 模型管理方法、系统、装置、通信设备及存储介质
US20220182263A1 (en) * 2020-12-03 2022-06-09 Qualcomm Incorporated Model discovery and selection for cooperative machine learning in cellular networks
CN115250485A (zh) * 2021-04-27 2022-10-28 华为技术有限公司 模型分发的方法和装置

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11443132B2 (en) * 2019-03-06 2022-09-13 International Business Machines Corporation Continuously improve recognition or prediction accuracy using a machine learning model to train and manage an edge application
WO2022041285A1 (zh) * 2020-08-31 2022-03-03 华为技术有限公司 一种模型数据传输方法及通信装置

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2022021421A1 (zh) * 2020-07-31 2022-02-03 Oppo广东移动通信有限公司 模型管理方法、系统、装置、通信设备及存储介质
US20220182263A1 (en) * 2020-12-03 2022-06-09 Qualcomm Incorporated Model discovery and selection for cooperative machine learning in cellular networks
CN115250485A (zh) * 2021-04-27 2022-10-28 华为技术有限公司 模型分发的方法和装置

Non-Patent Citations (5)

* Cited by examiner, † Cited by third party
Title
CHADI KHIRALLAH, SAMSUNG ELECTRONICS NORDIC AB: "AI/ML Model Management", 3GPP DRAFT; R2-2211990; TYPE DISCUSSION, 3RD GENERATION PARTNERSHIP PROJECT (3GPP), MOBILE COMPETENCE CENTRE ; 650, ROUTE DES LUCIOLES ; F-06921 SOPHIA-ANTIPOLIS CEDEX ; FRANCE, vol. 3GPP RAN 2, no. Toulouse, FR; 20221114 - 20221118, 4 November 2022 (2022-11-04), Mobile Competence Centre ; 650, route des Lucioles ; F-06921 Sophia-Antipolis Cedex ; France, XP052216078 *
CHINA TELECOM: "Discussion on general aspects of AI/ML for NR air interface", 3GPP DRAFT; R1-2203656, 3RD GENERATION PARTNERSHIP PROJECT (3GPP), MOBILE COMPETENCE CENTRE ; 650, ROUTE DES LUCIOLES ; F-06921 SOPHIA-ANTIPOLIS CEDEX ; FRANCE, vol. RAN WG1, no. e-Meeting; 20220509 - 20220520, 29 April 2022 (2022-04-29), Mobile Competence Centre ; 650, route des Lucioles ; F-06921 Sophia-Antipolis Cedex ; France, XP052153096 *
JIANGSHENG FAN, OPPO: "AIML Methods Discussion in General", 3GPP DRAFT; R2-2211192; TYPE DISCUSSION; FS_NR_AIML_AIR, 3RD GENERATION PARTNERSHIP PROJECT (3GPP), MOBILE COMPETENCE CENTRE ; 650, ROUTE DES LUCIOLES ; F-06921 SOPHIA-ANTIPOLIS CEDEX ; FRANCE, vol. 3GPP RAN 2, no. Toulouse, FR; 20221114 - 20221118, 4 November 2022 (2022-11-04), Mobile Competence Centre ; 650, route des Lucioles ; F-06921 Sophia-Antipolis Cedex ; France, XP052215304 *
See also references of EP4633213A4 *
XIANG PAN, VIVO: "Further discussion on AIML methods", 3GPP DRAFT; R2-2211234; TYPE DISCUSSION; FS_NR_AIML_AIR, 3RD GENERATION PARTNERSHIP PROJECT (3GPP), MOBILE COMPETENCE CENTRE ; 650, ROUTE DES LUCIOLES ; F-06921 SOPHIA-ANTIPOLIS CEDEX ; FRANCE, vol. 3GPP RAN 2, no. Toulouse, FR; 20221114 - 20221118, 4 November 2022 (2022-11-04), Mobile Competence Centre ; 650, route des Lucioles ; F-06921 Sophia-Antipolis Cedex ; France, XP052215346 *

Also Published As

Publication number Publication date
US20250300906A1 (en) 2025-09-25
EP4633213A4 (en) 2026-04-22
EP4633213A1 (en) 2025-10-15
CN118158656A (zh) 2024-06-07

Similar Documents

Publication Publication Date Title
WO2023246584A1 (zh) 算力处理方法、装置及通信设备
CN116156667B (zh) 物联网设备的会话建立方法及装置
WO2024120282A1 (zh) 信息发送、信息接收方法、装置、终端及网络侧设备
US20250300906A1 (en) Information transmission method and apparatus, and terminal and network-side device
WO2024120352A1 (zh) 通信方法、终端及核心网功能
WO2025026192A1 (zh) Ai业务的处理方法、装置、设备及可读存储介质
WO2024061143A1 (zh) 应用程序确定方法、装置、终端及服务器
WO2023143423A1 (zh) 信息获取与存储、上报方法、装置、终端及网络功能
WO2024078400A1 (zh) 模型请求方法、装置、通信设备及可读存储介质
WO2024067436A1 (zh) 信息传输方法、装置及设备
WO2023078268A1 (zh) Dns消息的发送方法、终端及网络侧设备
CN116847330A (zh) 终端不可用时期的协商方法、终端及网络侧设备
CN117177229A (zh) 数据传输方法、装置、通信设备及网元
CN116828546A (zh) 非3gpp设备的会话通道建立方法、装置及设备
WO2024120286A1 (zh) 传输方法、装置、终端及网络侧设备
US20260065169A1 (en) Information transmission method and apparatus, model determining method and apparatus, and related device
WO2025195234A1 (zh) 终端能力注册方法、获取方法、装置、终端及设备
CN120238435A (zh) 信息传输方法、装置、终端及网络侧设备
WO2024140712A1 (zh) 模型提供、模型获取、设备查询方法、装置和通信设备
CN117528676A (zh) 托管网络的选网信息的获取方法、终端及网络侧设备
CN117692986A (zh) 路由选择策略执行结果处理方法、装置及设备
WO2025026260A1 (zh) 会话修改方法、会话修改装置和通信设备
CN118714616A (zh) 通信、会话或规则配置或更新的控制、信息传输方法
WO2026067378A1 (zh) 通信方法、装置及设备
WO2025026247A1 (zh) 会话修改方法、信息转发方法、装置和通信设备

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: 23899864

Country of ref document: EP

Kind code of ref document: A1

WWE Wipo information: entry into national phase

Ref document number: 2023899864

Country of ref document: EP

NENP Non-entry into the national phase

Ref country code: DE

ENP Entry into the national phase

Ref document number: 2023899864

Country of ref document: EP

Effective date: 20250707

WWP Wipo information: published in national office

Ref document number: 2023899864

Country of ref document: EP