WO2023186099A1 - 信息反馈方法、装置及设备 - Google Patents

信息反馈方法、装置及设备 Download PDF

Info

Publication number
WO2023186099A1
WO2023186099A1 PCT/CN2023/085491 CN2023085491W WO2023186099A1 WO 2023186099 A1 WO2023186099 A1 WO 2023186099A1 CN 2023085491 W CN2023085491 W CN 2023085491W WO 2023186099 A1 WO2023186099 A1 WO 2023186099A1
Authority
WO
WIPO (PCT)
Prior art keywords
information
demand
requirement
feedback
indicate
Prior art date
Application number
PCT/CN2023/085491
Other languages
English (en)
French (fr)
Inventor
周通
Original Assignee
维沃移动通信有限公司
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 维沃移动通信有限公司 filed Critical 维沃移动通信有限公司
Publication of WO2023186099A1 publication Critical patent/WO2023186099A1/zh

Links

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04W—WIRELESS COMMUNICATION NETWORKS
    • H04W28/00—Network traffic management; Network resource management
    • H04W28/16—Central resource management; Negotiation of resources or communication parameters, e.g. negotiating bandwidth or QoS [Quality of Service]
    • 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

  • This application belongs to the field of communication technology, and specifically relates to an information feedback method, device and equipment.
  • Wireless mobile communications combined with artificial intelligence can better improve communication quality, such as AI-based channel quality compression, AI-based beam management, and AI-based positioning.
  • communication transceivers such as base stations and terminals
  • communication transceivers are configured with multiple analog beams.
  • the channel quality measured in different transmitting and receiving analog beams changes.
  • How to quickly and accurately select the transceiver beam group with the highest channel quality from all possible transceiver simulation beam combinations is the key to affecting transmission quality.
  • the terminal can effectively predict the simulated transceiver beam with the highest channel quality based on the AI neural network model and report it to the network side, thereby achieving better transmission quality.
  • the first device receives the AI service requirement sent by the second device (such as core network element) and implements the AI service requirement, so that the second device can realize the split AI service requirement through the first device To implement complex AI services.
  • the first device implements the AI service requirements, it may affect the normal operation of the device itself, making the first device unable to work normally.
  • Embodiments of the present application provide an information feedback method, device and equipment, which can solve the problem that when the first device implements AI service requirements, it may affect the normal operation of the device itself, making the first device unable to work normally.
  • the first aspect provides an information feedback method, including:
  • the first device receives demand information sent by the second device, where the demand information is used to indicate the demand corresponding to the artificial intelligence AI service;
  • the first device sends feedback information on the requirement information to the second device, where the feedback information is used to indicate whether the requirement corresponding to the requirement information is agreed to.
  • the second aspect provides an information feedback method, including:
  • the second device sends demand information to the first device, where the demand information is used to indicate the demand corresponding to the artificial intelligence AI service;
  • the second device receives feedback information on the requirement information sent by the first device, where the feedback information is used to indicate whether to agree to the requirement corresponding to the requirement information.
  • an information feedback device in a third aspect, includes the information feedback device, including:
  • a receiving module configured to receive demand information sent by the second device, where the demand information is used to indicate the demand corresponding to the artificial intelligence AI service;
  • a sending module configured to send feedback information on the demand information to the second device, where the feedback information is used to indicate whether the demand corresponding to the demand information is agreed to.
  • an information feedback device includes the information feedback device, including:
  • a sending module configured to send demand information to the first device, where the demand information is used to indicate the demand corresponding to the artificial intelligence AI service;
  • a receiving module configured to receive feedback information on the demand information sent by the first device, where the feedback information is used to indicate whether to agree to the demand corresponding to the demand information.
  • a first device in a fifth aspect, includes a processor and a memory.
  • the memory stores a program or instructions executable on the processor. The program or instructions are executed by the processor. When implementing the steps of the method described in the first aspect.
  • a first device including a processor and a communication interface, wherein the communication interface is used to receive demand information sent by the second device, and the demand information is used to indicate requirements corresponding to artificial intelligence AI services. ; The communication interface is also used to send feedback information on the demand information to the second device, where the feedback information is used to indicate whether the demand corresponding to the demand information is agreed to.
  • a second device in a seventh aspect, includes a processor and a memory.
  • the memory stores programs or instructions executable on the processor. The programs or instructions are executed by the processor. When implementing the steps of the method described in the second aspect.
  • a second device including a processor and a communication interface, wherein the communication interface is used to send demand information to the first device, and the demand information is used to indicate requirements corresponding to artificial intelligence AI services;
  • the communication interface is further configured to receive feedback information on the demand information sent by the first device, where the feedback information is used to indicate whether to agree to the demand corresponding to the demand information.
  • an information feedback system including: a first device and a second device.
  • the first device can be used to perform the steps of the information feedback method as described in the first aspect.
  • the second device can be used to The steps of the information feedback method described in the second aspect are performed.
  • a readable storage medium In a tenth aspect, a readable storage medium is provided. Programs or instructions are stored on the readable storage medium. When the programs or instructions are executed by a processor, the steps of the method described in the first aspect are implemented, or the steps of the method are implemented as described in the first aspect. The steps of the method described in the second aspect.
  • a chip in an eleventh aspect, includes a processor and a communication interface.
  • the communication interface is coupled to the processor.
  • the processor is used to run programs or instructions to implement the method described in the first aspect. method, or implement a method as described in the second aspect.
  • a computer program/program product is provided, the computer program/program product is stored in a storage medium, and the computer program/program product is executed by at least one processor to implement as described in the first aspect
  • the steps of the information feedback method, or the computer program/program product is executed by at least one processor to implement the steps of the information feedback method as described in the second aspect.
  • the first device receives demand information sent by the second device, and the demand information is used to indicate the demand corresponding to the artificial intelligence AI service; the first device sends the demand information to the second device.
  • the feedback information is used to indicate whether the requirement corresponding to the requirement information is agreed to. In this way, after receiving the demand information sent by the second device, the first device sends feedback information to the second device to avoid directly implementing the AI service demand and affecting the normal operation of the device itself.
  • Figure 1 is a block diagram of a wireless communication system applicable to the embodiment of the present application.
  • Figure 2 is one of the flow charts of an information feedback method provided by an embodiment of the present application.
  • Figure 3 is the second flow chart of an information feedback method provided by an embodiment of the present application.
  • Figure 4 is one of the structural diagrams of an information feedback device provided by an embodiment of the present application.
  • Figure 5 is the second structural diagram of an information feedback device provided by an embodiment of the present application.
  • Figure 6 is a structural diagram of a communication device provided by an embodiment of the present application.
  • Figure 7 is a schematic structural diagram of a terminal provided by an embodiment of the present application.
  • Figure 8 is a schematic structural diagram of a network side device provided by an embodiment of the present application.
  • first, second, etc. in the description and claims of this application are used to distinguish similar objects and are not used to describe a specific order or sequence. It is to be understood that the terms so used are interchangeable under appropriate circumstances so that the embodiments of the present application can be practiced in sequences other than those illustrated or described herein, and that "first" and “second” are distinguished objects It is usually one type, and the number of objects is not limited.
  • the first object can be one or multiple.
  • “and/or” in the description and claims indicates at least one of the connected objects, and the character “/" generally indicates that the related objects are in an "or” relationship.
  • LTE Long Term Evolution
  • LTE-Advanced, LTE-A Long Term Evolution
  • LTE-A Long Term Evolution
  • CDMA Code Division Multiple Access
  • TDMA Time Division Multiple Access
  • FDMA Frequency Division Multiple Access
  • OFDMA Orthogonal Frequency Division Multiple Access
  • SC-FDMA single-carrier frequency division multiple access
  • NR New Radio
  • FIG. 1 shows a block diagram of a wireless communication system to which embodiments of the present application are applicable.
  • the wireless communication system includes a terminal 11 and a network side device 12.
  • the terminal 11 may be a mobile phone, a tablet computer (Tablet Personal Computer), a laptop computer (Laptop Computer), or a notebook computer, a personal digital assistant (Personal Digital Assistant, PDA), a palmtop computer, a netbook, or a super mobile personal computer.
  • Tablet Personal Computer Tablet Personal Computer
  • laptop computer laptop computer
  • PDA Personal Digital Assistant
  • PDA Personal Digital Assistant
  • UMPC ultra-mobile personal computer
  • UMPC mobile Internet device
  • MID mobile Internet Device
  • AR augmented reality
  • VR virtual reality
  • robots wearable devices
  • WUE Vehicle User Equipment
  • PUE Pedestrian User Equipment
  • smart home home equipment with wireless communication functions, such as refrigerators, TVs, washing machines or furniture, etc.
  • game consoles personal computers (personal computer, PC), teller machine or self-service machine and other terminal-side devices.
  • Wearable devices include: smart watches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart bracelets, smart rings, smart necklaces, smart anklets) bracelets, smart anklets, etc.), smart wristbands, smart clothing, etc.
  • the network side device 12 may include an access network device or a core network device, where the access network device may also be called a radio access network device, a radio access network (Radio Access Network, RAN), a radio access network function or a wireless access network unit.
  • Access network equipment may include a base station, a Wireless Local Area Network (WLAN) access point or a WiFi node, etc.
  • WLAN Wireless Local Area Network
  • the base station may be called a Node B, an Evolved Node B (eNB), an access point, a base transceiver station ( Base Transceiver Station (BTS), radio base station, radio transceiver, Basic Service Set (BSS), Extended Service Set (ESS), home B-node, home evolved B-node, transmitting and receiving point ( Transmitting Receiving Point (TRP) or some other appropriate terminology in the field, as long as the same technical effect is achieved, the base station is not limited to specific technical terms. It should be noted that in the embodiment of this application, only in the NR system The base station is introduced as an example, and the specific type of base station is not limited.
  • Core network equipment may include but is not limited to at least one of the following: core network nodes, core network functions, mobility management entities (Mobility Management Entity, MME), access mobility management functions (Access and Mobility Management Function, AMF), session management functions (Session Management Function, SMF), User Plane Function (UPF), Policy Control Function (PCF), Policy and Charging Rules Function (PCRF), Edge Application Service Discovery function (Edge Application Server Discovery Function, EASDF), Unified Data Management (UDM), Unified Data Repository (UDR), Home Subscriber Server (HSS), Centralized network configuration ( Centralized network configuration (CNC), Network Repository Function (NRF), Network Exposure Function (NEF), Local NEF (Local NEF, or L-NEF), Binding Support Function (Binding Support Function, BSF), application function (Application Function, AF), etc.
  • MME mobility management entities
  • AMF Access and Mobility Management Function
  • SMF Session Management Function
  • UPF User Plane Function
  • PCF Policy Control Function
  • Figure 2 is a flow chart of an information feedback method provided by an embodiment of the present application. As shown in Figure 2, the information feedback method includes the following steps:
  • Step 101 The first device receives demand information sent by the second device, where the demand information is used to indicate the demand corresponding to the artificial intelligence AI service.
  • AI services can include services related to AI neural network models such as model training, model fine-tuning, model inference, and model verification. It should be noted that for AI neural network models trained based on simulation data or sample data collected from other cells, due to generalization issues, direct inference in the cell where the terminal is located often does not achieve good performance.
  • the terminal can retrain the AI neural network model based on the data of the community where the terminal is located or use AI services such as fine-tuning to improve the performance of the AI neural network model.
  • the AI service may include at least one of model training, fine-tuning, model inference, and model verification.
  • the first device may be a terminal, a base station, a core network element, etc.
  • the second device can be a base station, or a self-organized network (Self-Organized Networks, SON), or a network management system (Operation Administration and Maintenance, OAM), or a core network element (for example, Network Data Analytics Function, NWDAF)), or network management, etc.
  • SON Self-Organized Networks
  • OAM Operaation Administration and Maintenance
  • NWDAF Network Data Analytics Function
  • Step 102 The first device sends feedback information on the demand information to the second device, where the feedback information is used to indicate whether to agree to the demand corresponding to the demand information.
  • the feedback information may be capability feedback information.
  • the feedback information may be smart capability feedback information, and the smart capability feedback information is used to indicate the smart capability of the first device.
  • the first device can be used for AI service execution and intelligent capability reporting
  • the second device can be used for AI service orchestration and deployment.
  • AI service orchestration and deployment AI services can be split into three dimensions: AI computing power, AI algorithms, and AI data.
  • the first device receives the demand information corresponding to the AI service, makes an evaluation based on its own capabilities, and reports acceptance, rejection, or demand differences. Therefore, the AI service execution end can execute the AI service orchestration and deploy the AI services allocated by the node within its own capabilities, so that the first device can not only provide high-quality AI services, but also ensure the normal operation of its other processes.
  • the first device receives demand information sent by the second device, and the demand information includes demands for AI computing power, and/or AI algorithms, and/or AI data.
  • the first device may send feedback information on the requirement information to the second device.
  • the feedback information may include acceptance, rejection, or capability difference information.
  • the capability difference information may be used to indicate the difference between the device capability and the requirement information.
  • the first device may send feedback information on the demand information to the second device according to its own decision.
  • the first device may be a terminal. If the current status of the terminal does not meet expectations, the terminal accepts the assigned AI services may affect the normal operation of other processes on the terminal, and it cannot guarantee the high-quality completion of AI services by the terminal.
  • the terminal receives demand information sent by the AI service orchestration and deployment device, and the demand information is used to indicate the demand corresponding to the artificial intelligence AI service; and sends feedback information to the demand information to the AI service orchestration and deployment device. Therefore, before accepting the AI service sent by the AI service orchestration and deployment device, the terminal can comprehensively consider the needs of the AI service and the intelligent capabilities of the device itself, and provide feedback. The problem of intelligent capability feedback can be solved from the AI service deployment process.
  • the demand information includes AI data requirements
  • the feedback information can be used to feedback differences in AI data requirements.
  • the feedback information can include information items in the demand information that cannot be satisfied in the AI data requirements, and/or demand information.
  • the current device capabilities of the first device corresponding to the information items that cannot be satisfied in the AI data requirements.
  • the demand information includes AI computing power requirements
  • the feedback information can be used to feedback differences in AI computing power requirements.
  • the feedback information can include information items in the demand information that cannot be satisfied in the AI computing power requirements, and/or , the current device capability of the first device corresponding to the information entry in the demand information that cannot be satisfied in the AI computing power demand.
  • the demand information includes AI algorithm requirements
  • the feedback information can be used to feedback differences in AI algorithm requirements.
  • the feedback information can include information items in the demand information that cannot be satisfied in the AI algorithm requirements, and/or demand information.
  • the current device capabilities of the first device corresponding to the information items that cannot be satisfied in the AI algorithm requirements.
  • the first device can combine the device's own computing power, algorithm, data and other aspects of device capabilities to determine whether its own device capabilities meet the demand information, and whether to accept the demand corresponding to the demand information, and determine the feedback information. For example, if the device capability of the first device meets the demand information and is willing to accept the demand corresponding to the demand information, the feedback information can be used to indicate acceptance of the demand; and/or if the first device is unwilling to accept the demand corresponding to the demand information. demand, the feedback information can be used to indicate rejection of the demand; and/or, if the equipment capability of the first device does not meet the demand information, but is willing to accept the demand corresponding to the demand information, the feedback information can be used to indicate the capability difference information. .
  • the first device receives demand information sent by the second device, and the demand information is used to indicate the demand corresponding to the artificial intelligence AI service; the first device sends the demand information to the second device.
  • the feedback information is used to indicate whether the requirement corresponding to the requirement information is agreed to. In this way, after receiving the demand information sent by the second device, the first device sends feedback information to the second device to avoid directly implementing the AI service demand and affecting the normal operation of the device itself.
  • the demand information includes: demand items, or demand items and demand values corresponding to the demand items.
  • the requirement item may be used to indicate the requirements corresponding to the AI service, which may include, for example, AI computing power requirements, AI algorithm requirements, AI data requirements, etc.
  • the first device receives demand information sent by the second device.
  • the demand information includes demand items in terms of AI computing power, and/or AI algorithms, and/or AI data, and requirements corresponding to the demand items. value.
  • the first device may send feedback information on the requirement information to the second device, and the feedback information may be used to indicate acceptance, rejection, or capability difference information.
  • the first device receives demand information sent by the second device, and the demand information includes demand items in terms of AI computing power, and/or AI algorithms, and/or AI data.
  • the first device may send to the second device a request for the Feedback information of demand information, which can be used to indicate rejection or equipment capabilities corresponding to the demand item.
  • the demand item can be expressed in the form of demand items.
  • the demand information includes an AI computing power demand item
  • the feedback information can include an AI computing power demand item value.
  • the AI computing power demand item value is assigned a value corresponding to the AI computing power demand item.
  • the requirement information includes an AI algorithm requirement entry
  • the feedback information may include an AI algorithm requirement entry value
  • the AI algorithm requirement entry value is assigned the equipment capability corresponding to the AI algorithm requirement entry
  • the The demand information includes an AI data demand item
  • the feedback information may include an AI data demand item value
  • the AI data demand item value is assigned a device capability corresponding to the AI data demand item.
  • the first device can determine whether it is willing to accept the demand corresponding to the demand item and determine the feedback information based on the device's own status. For example, if the first device is unwilling to accept the requirement corresponding to the requirement item, the feedback information may indicate rejection; if the first device is willing to accept the requirement corresponding to the requirement item, the feedback information may indicate the current device capability corresponding to the requirement item.
  • the second device can split the AI service from the three dimensions of AI computing power, AI algorithm, and AI data to obtain demand items in these three dimensions.
  • Demand items in the AI computing power dimension may include at least one of the following: computing power; storage; total calculation amount; execution time limit; and power consumption.
  • Requirements in the AI algorithm dimension may include at least one of the following: AI task classification; AI learning framework; AI network development environment; and AI basic model library.
  • Requirements on the AI data dimension may include at least one of the following: label data; number of label dimensions; label dimensions; order of label dimensions; number of data in label dimensions; label data interval; AI model input data; AI model input dimensions Number; AI model input dimensions; order of AI model input dimensions; number of data in AI model input dimensions; AI model input data collection interval; AI model label delay.
  • the demand information includes a demand item, so that the demand item corresponding to the AI service can be indicated, and the first device can feed back the device capability corresponding to the demand item to the second device through the demand item; or,
  • the demand information includes a demand item and a demand value corresponding to the demand item, so that the demand item and the demand value can indicate the demand corresponding to the AI service, and the second device can feed back the demand item correspondence to the second device through the demand item and the demand value. Whether the equipment capacity can meet the demand value.
  • the feedback information includes any one of the following:
  • the first indication information is used to indicate that the requirement item is agreed to and the device capability meets the requirement value
  • Second indication information the second indication information is used to indicate agreement with the requirement item, and the second indication information carries device capabilities or capability difference information corresponding to the requirement item, and the capability difference information is used to indicate The difference between equipment capabilities and stated demand values;
  • the third indication information is used to indicate rejection of the requirement item.
  • the first device sends feedback information on the demand information to the second device, which may include:
  • the third indication information is sent to the second device.
  • the first device sends feedback information on the demand information to the second device, which may include:
  • the first device When the device capability of the first device meets the demand value, and the first device can maintain the normal operation of the first device while meeting the demand value, send a message to the second device. the first indication information; and/or
  • the second device sends the second indication information
  • the first device When the device capability of the first device does not meet the demand value and the first device can only maintain the normal operation of the first device, send the third indication information to the second device .
  • the first device after the first device sends the first instruction information to the second device, it can directly execute the AI service corresponding to the requirement item; or, the first device can receive an instruction sent by the second device to determine execution. After receiving the instruction information, execute the AI service corresponding to the requirement item.
  • the first device after the first device sends the second instruction information to the second device, it can directly execute the AI service corresponding to the requirement item; or, the first device can determine execution after receiving the instruction sent by the second device. The first device can then execute the AI service corresponding to the requirement item after receiving the instruction information sent by the second device for instructing to modify the requirement value before executing the AI service corresponding to the requirement item; etc. , this embodiment is not limited to this.
  • the second device may send instruction information to modify the demand value to the first device, and modify the demand value to a demand value that matches the device capability.
  • the first device sends third indication information to the second device, that is, the first device refuses to execute the AI service corresponding to the requirement item.
  • the first device can feed back to the second device whether to agree or reject the requirement item corresponding to the AI service, and feed back the requirement item to the second device.
  • Whether the corresponding device capability can meet the demand value allows the first device to perform the assigned AI service within its own capabilities.
  • the number of the requirement items is at least two
  • the second indication information carries the first requirement item and the device capability or capability difference information corresponding to the first requirement item
  • the first requirement item is The demand item among the at least two demand items that does not match the equipment capability.
  • the capability difference information corresponding to the first requirement item may be used to indicate the difference between the equipment capability and the requirement value corresponding to the first requirement item.
  • the AI computing power demand includes computing power
  • the computing power value corresponding to the AI computing power demand is 20T floating point operations per second (FLOPs).
  • the first device determines that the actual computing power of the current device is 15T FLOPs
  • the capability difference information corresponding to the first requirement item is 5T FLOPs
  • the device capability corresponding to the first requirement item is 15T FLOPs
  • the second indication information can carry the capability difference Information 5T FLOPs, or device capability 15T FLOPs.
  • the first demand item does not match the equipment capability, it may be considered that the demand value corresponding to the first demand item does not match the equipment capability. For example, it may be that the demand value corresponding to the first demand item is greater than the equipment capability.
  • the first device can, after receiving the requirement corresponding to the AI service, The first demand item whose own equipment capability cannot meet the demand value is fed back to the second equipment, and the capability difference information corresponding to the first demand item is fed back.
  • the feedback information includes any of the following:
  • the fourth instruction information is used to indicate rejection of the requirement item
  • the fifth indication information is used to indicate agreement with the requirement item, and the fifth indication information carries the device capability corresponding to the requirement item.
  • the device capability corresponding to the requirement item may be the current device capability corresponding to the requirement item. Taking the requirement item including computing power as an example, the device capability corresponding to the requirement item may include the current computing power of the device.
  • the first device sends feedback information on the demand information to the second device, which may include:
  • the fifth indication information is sent to the second device.
  • the first device sends feedback information on the demand information to the second device, which may include:
  • the fifth indication information is sent to the second device.
  • the first device sends fourth instruction information to the second device, that is, the first device refuses to execute the AI service corresponding to the requirement item.
  • the first device may receive an instruction sent by the second device to execute the AI service corresponding to the requirement item according to the device capability of the first device, and perform the instruction according to the first device.
  • the current device capability of the device executes the AI service corresponding to the requirement item.
  • the second device can adjust the task allocation of the AI service based on the device capabilities reported by the first device.
  • the first device can feed back to the second device whether to agree or reject the requirement item corresponding to the AI service, and feed back to the second device the device corresponding to the requirement item. ability.
  • the requirements include at least one of the following:
  • AI computing power requirements AI algorithm requirements
  • AI data requirements AI data requirements
  • the first device receives the demand information sent by the second device, including the first device receiving the AI computing power demand sent by the second device, and/or the AI computing power demand and the demand corresponding to the AI computing power demand. value;
  • the first device sends feedback information on the demand information to the second device, including the first device sending feedback information on the AI computing power demand to the second device, where the feedback information is used to indicate whether to agree to the AI computing power.
  • demand and the feedback information also includes device capabilities corresponding to the AI computing power demand, or capability difference information corresponding to the AI computing power demand. This enables device capability feedback on the AI computing power requirements requested by the second device from the first device.
  • the first device receives the demand information sent by the second device, including: the first device receives the AI algorithm demand sent by the second device, and/or the AI algorithm demand and the demand value corresponding to the AI algorithm demand;
  • the first device sends feedback information on the requirement information to the second device, including: the first device sends feedback information on the AI algorithm requirement to the second device, where the feedback information is used to indicate whether the AI algorithm requirement is agreed to, And the feedback information also includes device capabilities corresponding to the AI algorithm requirements, or capability difference information corresponding to the AI algorithm requirements. This enables device capability feedback on the AI algorithm requirements requested by the second device from the first device.
  • the first device receives the demand information sent by the second device, including the first device receiving the AI data demand sent by the second device, and/or the AI data demand and the demand value corresponding to the AI data demand;
  • the first device sends feedback information on the demand information to the second device, including: the first device sends feedback information on the AI data demand to the second device, where the feedback information is used to indicate whether to agree to the AI data demand, And the feedback information also includes device capabilities corresponding to the AI data requirements, or capability difference information corresponding to the AI data requirements. This enables device capability feedback on the AI data requirements requested by the second device from the first device.
  • the first device receives at least one of the AI computing power requirements, AI algorithm requirements, and AI data requirements sent by the second device, and sends feedback information corresponding to the requirement items to the second device, which is beneficial to the first device.
  • the second device provides higher quality AI services.
  • the AI computing power requirements include at least one of the following:
  • the computing power in the AI computing power requirements can be used to indicate the computing power required by the AI service, and the computing power can be expressed in the number of multiplications/second.
  • the storage in AI computing power requirements can be used to indicate the amount of storage required by the AI service.
  • the total amount of calculations in the AI computing power requirements can be used to indicate the total amount of calculations required to execute the AI service.
  • the total amount of calculations can be calculated from the computing power and the inference time of the AI model.
  • the execution time limit in the AI computing power requirement can be used to indicate the time limit for executing the AI service.
  • the power consumption in AI computing power requirements can be used to indicate the power consumption required by AI services.
  • the first device receives demand information sent by the second device, where the demand information includes computing power, storage At least one computing power requirement among storage, total calculation amount, execution time limit, and power consumption, or at least one computing power requirement and its corresponding demand value;
  • the first device sends feedback information on the demand information to the second device according to device capabilities.
  • the feedback information can be used to indicate whether to agree with the at least one computing power requirement, and the feedback information can carry The equipment capacity corresponding to the computing power demand or the capacity difference information between the equipment capacity and the demand value.
  • the first device receives demand information sent by the second device.
  • the demand information includes AI algorithm requirements and AI computing power requirements.
  • the AI algorithm requirements include model training, model reasoning, model verification, model monitoring, and model deployment.
  • At least one AI task in the AI computing power requirement includes at least one computing power requirement or at least one of the computing power required to execute the at least one AI task, storage, total calculation amount, execution time limit and power consumption. Item computing power requirements and corresponding demand values;
  • the first device sending feedback information on the demand information to the second device may include the first device judging whether it can meet the AI computing power requirements required to execute the at least one AI task based on device capabilities. If it is judged that the AI computing power required to perform the at least one AI task can be met, the first instruction information is sent to the second device; if it is judged that only part of the AI computing power required to perform the at least one AI task can be met If necessary, the second instruction information is sent to the second device; if it is judged that only the normal operation of the first device can be maintained, the third instruction information is sent to the second device.
  • the first device receives demand information sent by the second device.
  • the demand information includes AI algorithm requirements and AI computing power requirements.
  • the AI algorithm requirements include model training, model reasoning, model verification, model monitoring, and model deployment.
  • At least one AI task in the AI computing power requirement includes at least one computing power requirement among computing power, storage, total calculation amount, execution time limit and power consumption required to execute the at least one AI task;
  • the first device sending feedback information on the demand information to the second device may include the first device judging whether the at least one AI task can be performed while maintaining normal operation of the device based on device capabilities, When it is determined that the at least one AI task can be executed while maintaining normal operation of the device, fifth instruction information is sent to the second device, the fifth instruction information carrying the at least one computing power requirement corresponding to Device capability; when it is determined that the at least one AI task cannot be performed while maintaining normal operation of the device, send fourth instruction information to the second device.
  • the first requirement can be facilitated through at least one of the AI computing power requirements including computing power, storage, total calculation amount, execution time limit, and power consumption, or the requirement term and the requirement value corresponding to the requirement term.
  • the device determines whether its own device capabilities can meet the AI computing power requirements.
  • the AI algorithm requirements include at least one of the following:
  • the AI task classification can be used to indicate the task type corresponding to the AI algorithm.
  • AI learning frameworks can be used to instruct The learning framework of AI algorithms specifically includes learning methods and training methods.
  • the AI network development environment can be used to indicate the network development environment of the AI algorithm.
  • the AI basic model library can be used to indicate the basic model library of the AI algorithm.
  • the demand information may include at least one demand item in the AI learning framework, AI network development environment, and AI basic model library required to perform the AI task, and the demand value corresponding to the demand item.
  • the first device sends feedback information on the demand information to the second device. It may be that the first device determines whether it agrees with the demand corresponding to the demand information based on device capabilities, and sends feedback information to the second device.
  • the feedback information is used to indicate whether the requirement corresponding to the requirement information is agreed to. For example, taking the requirement information including an AI learning framework as an example, when the AI algorithm of the first device is the AI learning framework, the requirements corresponding to the requirement information can be agreed to.
  • the device determines whether its own device capabilities can meet the needs of the AI algorithm.
  • the AI task classification includes at least one of the following:
  • the first device receives demand information sent by the second device.
  • the demand information includes at least one AI task among model training, model inference, model verification, model monitoring, and model deployment.
  • the first device When the device capability of the first device can only maintain the normal operation of the first device, send feedback information agreeing to the demand corresponding to the demand information to the second device; when the device capability of the first device can maintain the When the first device is operating normally and can perform the AI task, feedback information agreeing with the requirement corresponding to the requirement information is sent to the second device.
  • the first device receives demand information sent by the second device.
  • the demand information includes at least one AI task in model training, model inference, model verification, model monitoring, and model deployment, and the requirements for executing the AI task.
  • the required AI algorithm requirements and/or AI data requirements when the equipment capability of the first device can meet the AI algorithm requirements and/or AI data requirements, the first device agrees to perform the at least one AI task and submits the request to the third device.
  • the second device sends feedback information to the demand information, and the feedback information indicates that it agrees with the demand corresponding to the demand information; in the case that the equipment capabilities of the first device can partially meet the AI algorithm requirements and/or AI data requirements, the second device A device agrees to perform the at least one AI task, sends feedback information on the demand information to the second device, the feedback information indicates agreeing to the demand corresponding to the demand information, and feeds back capability difference information; in the first device If the device capability cannot meet the AI algorithm requirements and/or AI data requirements, the first device refuses to perform the at least one AI task.
  • AI tasks such as model training, model inference, model verification, model monitoring, and model deployment issued by the second device, it can improve the performance of the first device in model training, model inference, model verification, model monitoring, and so on.
  • the quality of AI tasks such as model deployment.
  • the AI learning framework includes at least one of the following:
  • the demand information may include an AI learning framework required to perform AI tasks.
  • the AI learning framework is at least one of supervised deep learning, unsupervised deep learning, meta-learning, transfer learning, reinforcement learning and federated learning.
  • One item you can agree to the demand corresponding to the demand information when the AI algorithm of the first device is the AI learning framework indicated in the demand information; when the AI algorithm of the first device is not the AI learning framework indicated in the demand information In the case of , reject the demand corresponding to the demand information.
  • the AI data requirements include at least one of the following:
  • the number of data input dimensions of the AI model is the number of data input dimensions of the AI model
  • the tag data may include the reference signal received power (RSRP) of the beam channel, the reference signal received power RSRP, the signal-to-noise and interference plus noise ratio (Signal-to-noise and Interference Ratio, SINR), etc., and the first Device communication-related data.
  • the number of label dimensions can be used to indicate the number of dimensions of label data that need to be collected.
  • Tag dimensions can be used to indicate the dimensions of tag data that need to be collected.
  • the order of tag dimensions can be used to indicate the order of each dimension of tag data that needs to be collected.
  • the number of data in the label dimension can be used to indicate the number of data in each dimension of the label data that needs to be collected.
  • Tag data interval can be used to indicate the data interval at which tag data needs to be collected.
  • AI model input data may be used to indicate input data for the AI model.
  • the number of AI model input dimensions can be used to indicate the number of dimensions of the AI model input data that needs to be collected.
  • AI model input dimensions can be used to indicate the dimensions of AI model input data that need to be collected.
  • the order of AI model input dimensions can be used to indicate the need The order of each dimension of the AI model input data to be collected.
  • the number of data in the input dimensions of the AI model can be used to indicate the number of data in each dimension of the AI model input data that needs to be collected.
  • AI model input data collection interval is used to indicate the data interval at which AI model input data needs to be collected.
  • AI model label latency can be used to indicate the delay between obtaining label data and the end of inference or receiving AI model input data.
  • the demand information may include label data required to perform AI tasks, the number of label dimensions, label dimensions, the order of label dimensions, the number of data in label dimensions, label data interval, and AI model label delay.
  • At least one piece of tag data information in the first device can determine the tag data required to perform the AI task through the at least one piece of tag data information.
  • the label data can be label data used for model training.
  • the first device determines whether it can obtain the tag data required to perform the AI task based on the device capabilities. If it can obtain the tag data required to perform the AI task, it agrees to the requirements corresponding to the demand information; if it cannot obtain the tag data required to perform the AI task, it agrees to the demand information. If the label data required for the AI task is rejected, the demand corresponding to the demand information will be rejected.
  • the first device can collect tag data according to its own device capabilities.
  • the demand information may include the AI model input data required to perform the AI task, the number of AI model input dimensions, the AI model input dimensions, the order of the AI model input dimensions, and the number of data in the AI model input dimensions. , at least one piece of AI model input data information in the AI model input data collection interval, and the first device can determine the AI model input data required to perform the AI task through the at least one piece of AI model input data information.
  • the AI model input data can be input data used for model inference.
  • the first device determines whether it can obtain the AI model input data required to perform the AI task based on the device capabilities.
  • the first device can obtain the AI model input data according to its own device capabilities.
  • the tag data is used to indicate at least one of the following:
  • RSRP Reference Signal Received Power
  • RSRQ Reference Signal Received Quality
  • SINR Signal-to-noise and interference ratio
  • RSSI Received Signal Strength Indication
  • Precoding matrix indicator (Precoding matrix indicator, PMI);
  • CQI Channel quality indicator
  • the tag data interval is used to indicate at least one of the following:
  • the AI model label delay is used to indicate at least one of the following:
  • the delay between the end of inference and the acquisition of tag data may be the time delay between the end of inference by the first device and the acquisition of tag data by the first device.
  • the delay between receiving the AI model input data and obtaining the label data may be the delay between the first device receiving the AI model input data and the first device obtaining the label data.
  • the first device can be a terminal
  • the second device can be a core network element
  • the AI service can be model training.
  • the terminal receives the demand information sent by the core network element.
  • the demand information includes AI data demand, and the AI data demand in the demand information specifies that the number of input dimensions of the AI model is 2 dimensions, which are the transmit beam dimension and the time dimension, where , the number of transmit beam dimensions is 64.
  • the terminal can measure that the current transmit beam dimension is 8.
  • the terminal is willing to accept the demand information, but needs to feedback items that are different between the current data collection situation and the demand information: the number of transmit beam dimensions, and the terminal's current transmit beam dimension number: 8.
  • the terminal sends feedback information to the core network element.
  • the feedback information includes the item of the difference: the number of transmit beam dimensions, and the terminal's current number of transmit beam dimensions: 8.
  • the terminal reports capability difference information with the AI data requirements to the core network element.
  • the first device can be a terminal
  • the second device can be a core network element
  • the AI service can be model training.
  • the terminal receives the demand information sent by the core network element.
  • the demand information includes the AI computing power demand, and the AI computing power demand in the demand information defines the computing power as 20T FLOPs.
  • the terminal evaluates the actual computing power to be 15T FLOPs based on its own capabilities. At this time, the terminal is willing to accept the demand information, but cannot support the demand information due to its own capabilities.
  • the terminal sends feedback information to the core network element.
  • the feedback information includes items that are different from the demand information: computing power, and the current actual computing power of 15T FLOPs.
  • the terminal reports capability difference information with the AI computing power requirements to the core network element.
  • the first device can be a terminal
  • the second device can be a core network element
  • the AI service can be model training.
  • the terminal receives the demand information sent by the core network element.
  • the demand information includes the AI algorithm demand, and the AI algorithm demand definition basic model in the demand information is a convolutional neural network model.
  • the model currently stored in the terminal is only the fully connected model.
  • the terminal is willing to accept the demand information, but cannot support the demand information due to its own capabilities.
  • the terminal sends feedback information to the core network element.
  • the feedback information includes the algorithm difference entry: basic model, and the current value: full connection model.
  • the terminal reports the capability difference information required by the AI algorithm to the core network element.
  • Figure 3 is a flow chart of an information feedback method provided by an embodiment of the present application. As shown in Figure 3, the information feedback method includes the following steps:
  • Step 201 The second device sends demand information to the first device, where the demand information is used to indicate the corresponding demand for the artificial intelligence AI service;
  • Step 202 The second device receives feedback information on the demand information sent by the first device, where the feedback information is used to indicate whether it agrees with the demand corresponding to the demand information.
  • the demand information includes: demand items, or demand items and demand values corresponding to the demand items.
  • the feedback information includes any one of the following:
  • the first indication information is used to indicate that the requirement item is agreed to and the device capability meets the requirement value
  • Second indication information the second indication information is used to indicate agreement with the requirement item, and the second indication information carries device capabilities or capability difference information corresponding to the requirement item, and the capability difference information is used to indicate The difference between equipment capabilities and stated demand values;
  • the third indication information is used to indicate rejection of the requirement item.
  • the number of the requirement items is at least two
  • the second indication information carries the first requirement item and the device capability or capability difference information corresponding to the first requirement item
  • the first requirement item is The demand item among the at least two demand items that does not match the equipment capability.
  • the feedback information includes any of the following:
  • the fourth instruction information is used to indicate rejection of the requirement item
  • the fifth indication information is used to indicate agreement with the requirement item, and the fifth indication information carries the device capability corresponding to the requirement item.
  • the requirements include at least one of the following:
  • AI computing power requirements AI algorithm requirements
  • AI data requirements AI data requirements
  • this embodiment is an implementation of the second device corresponding to the embodiment shown in Figure 2.
  • the first device directly implements the AI service requirement and affects the normal operation of the device itself.
  • the execution subject may be an information feedback device.
  • an information feedback device performing an information feedback method is used as an example to illustrate the information feedback device provided by the embodiment of the present application.
  • Figure 4 is a structural diagram of an information feedback device provided by an embodiment of the present application.
  • the first device includes the information feedback device.
  • the information feedback device 300 includes:
  • the receiving module 301 is used to receive demand information sent by the second device, where the demand information is used to indicate the demand corresponding to the artificial intelligence AI service;
  • the sending module 302 is configured to send feedback information on the demand information to the second device, where the feedback information is used to indicate whether the demand corresponding to the demand information is agreed to.
  • the demand information includes: demand items, or demand items and demand values corresponding to the demand items.
  • the feedback information includes any one of the following:
  • the first indication information is used to indicate that the requirement item is agreed to and the device capability meets the requirement value
  • Second indication information the second indication information is used to indicate agreement with the requirement item, and the second indication information carries device capabilities or capability difference information corresponding to the requirement item, and the capability difference information is used to indicate The difference between equipment capabilities and stated demand values;
  • the third indication information is used to indicate rejection of the requirement item.
  • the number of the requirement items is at least two
  • the second indication information carries the first requirement item and the device capability or capability difference information corresponding to the first requirement item
  • the first requirement item is The demand item among the at least two demand items that does not match the equipment capability.
  • the feedback information includes any of the following:
  • the fourth instruction information is used to indicate rejection of the requirement item
  • the fifth indication information is used to indicate agreement with the requirement item, and the fifth indication information carries the device capability corresponding to the requirement item.
  • the requirements include at least one of the following:
  • AI computing power requirements AI algorithm requirements
  • AI data requirements AI data requirements
  • the AI computing power requirements include at least one of the following:
  • the AI algorithm requirements include at least one of the following:
  • the AI task classification includes at least one of the following:
  • the AI learning framework includes at least one of the following: supervised deep learning;
  • the AI data requirements include at least one of the following: label data;
  • the number of data input dimensions of the AI model is the number of data input dimensions of the AI model
  • the tag data is used to indicate at least one of the following:
  • SINR Signal to interference plus noise ratio
  • the precoding matrix indicates PMI
  • the tag data interval is used to indicate at least one of the following:
  • the AI model label delay is used to indicate at least one of the following:
  • the information feedback device in the embodiment of the present application after receiving the demand information sent by the second device, sends feedback information on the demand information to the second device, which can avoid directly realizing the AI service demand and affecting the normal operation of the device itself.
  • the information feedback device in the embodiment of the present application may be an electronic device, such as an electronic device with an operating system, or may be a component in the electronic device, such as an integrated circuit or chip.
  • the electronic device may be a terminal or other devices other than the terminal.
  • terminals may include but are not limited to the types of terminals 11 listed above, and other devices may be servers, network attached storage (Network Attached Storage, NAS), etc., which are not specifically limited in the embodiment of this application.
  • NAS Network Attached Storage
  • the information feedback device provided by the embodiment of the present application can implement each process implemented by the method embodiment in Figure 2 and achieve the same technical effect. To avoid duplication, it will not be described again here.
  • Figure 5 is a structural diagram of an information feedback device provided by an embodiment of the present application.
  • the second device includes the information feedback device.
  • the information feedback device 400 includes:
  • the sending module 401 is used to send demand information to the first device, where the demand information is used to indicate the demand corresponding to the artificial intelligence AI service;
  • the receiving module 402 is configured to receive feedback information on the demand information sent by the first device, where the feedback information is used to indicate whether to agree to the demand corresponding to the demand information.
  • the demand information includes: demand items, or demand items and demand values corresponding to the demand items.
  • the feedback information includes any one of the following:
  • the first indication information is used to indicate that the requirement item is agreed to and the device capability meets the requirement value
  • the second indication information is used to indicate agreement with the requirement item, and the second indication information carries device capabilities or capability difference information corresponding to the requirement item, and the capability difference information is used to indicate Equipment capabilities and The difference in said demand value;
  • the third indication information is used to indicate rejection of the requirement item.
  • the number of the requirement items is at least two
  • the second indication information carries the first requirement item and the device capability or capability difference information corresponding to the first requirement item
  • the first requirement item is The demand item among the at least two demand items that does not match the equipment capability.
  • the feedback information includes any of the following:
  • the fourth instruction information is used to indicate rejection of the requirement item
  • the fifth indication information is used to indicate agreement with the requirement item, and the fifth indication information carries the device capability corresponding to the requirement item.
  • the requirements include at least one of the following:
  • AI computing power requirements AI algorithm requirements
  • AI data requirements AI data requirements
  • the information feedback device in the embodiment of the present application can prevent the first device from directly realizing the AI service requirement and affecting the normal operation of the device itself.
  • the information feedback device in the embodiment of the present application may be an electronic device, such as an electronic device with an operating system, or may be a component in the electronic device, such as an integrated circuit or chip.
  • the electronic device may be a terminal or other devices other than the terminal.
  • terminals may include but are not limited to the types of terminals 11 listed above, and other devices may be servers, network attached storage (Network Attached Storage, NAS), etc., which are not specifically limited in the embodiment of this application.
  • NAS Network Attached Storage
  • the information feedback device provided by the embodiment of the present application can implement each process implemented by the method embodiment in Figure 3 and achieve the same technical effect. To avoid duplication, it will not be described again here.
  • this embodiment of the present application also provides a communication device 500, including a processor 501 and a memory 502.
  • the memory 502 stores programs or instructions that can be run on the processor 501, for example.
  • the communication device 500 is the first device
  • the program or instruction is executed by the processor 501
  • each step of the above embodiment of the information feedback method applied to the first device is implemented, and the same technical effect can be achieved.
  • the communication device 500 is a second device, when the program or instruction is executed by the processor 501, each step of the above embodiment of the information feedback method applied to the second device is implemented, and the same technical effect can be achieved. To avoid duplication, here No longer.
  • Embodiments of the present application also provide a terminal.
  • the terminal may be a first device, including a processor and a communication interface.
  • the communication interface is used to receive demand information sent by the second device.
  • the demand information is used to instruct artificial intelligence AI services.
  • the communication interface is also used to send feedback information on the requirement information to the second device, where the feedback information is used to indicate whether to agree to the requirements corresponding to the requirement information.
  • This terminal embodiment corresponds to the above-mentioned first device-side method embodiment.
  • Each implementation process and implementation manner of the above-mentioned method embodiment can be applied to this terminal embodiment, and can achieve the same technical effect.
  • FIG. 7 is a schematic diagram of the hardware structure of a terminal that implements an embodiment of the present application.
  • the terminal 600 includes but is not limited to: a radio frequency unit 601, a network module 602, an audio output unit 603, an input unit 604, a sensor 605, a display unit 606, a user input unit 607, an interface unit 608, a memory 609, a processor 610, etc. At least some parts.
  • the terminal 600 may also include a power supply (such as a battery) that supplies power to various components.
  • the power supply may be logically connected to the processor 610 through a power management system, thereby managing charging, discharging, and power consumption through the power management system. Management and other functions.
  • the terminal structure shown in FIG. 7 does not constitute a limitation on the terminal.
  • the terminal may include more or fewer components than shown in the figure, or some components may be combined or arranged differently, which will not be described again here.
  • the input unit 604 may include a graphics processing unit (Graphics Processing Unit, GPU) 6041 and a microphone 6042.
  • the graphics processor 6041 is responsible for the image capture device (GPU) in the video capture mode or the image capture mode. Process the image data of still pictures or videos obtained by cameras (such as cameras).
  • the display unit 606 may include a display panel 6061, which may be configured in the form of a liquid crystal display, an organic light emitting diode, or the like.
  • the user input unit 607 includes a touch panel 6071 and at least one of other input devices 6072 .
  • Touch panel 6 071 also known as touch screen.
  • the touch panel 6071 may include two parts: a touch detection device and a touch controller.
  • Other input devices 6072 may include but are not limited to physical keyboards, function keys (such as volume control keys, switch keys, etc.), trackballs, mice, and joysticks, which will not be described again here.
  • the radio frequency unit 601 after receiving downlink data from the network side device, can transmit it to the processor 610 for processing; in addition, the radio frequency unit 601 can send uplink data to the network side device.
  • the radio frequency unit 601 includes, but is not limited to, an antenna, an amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, etc.
  • Memory 609 may be used to store software programs or instructions as well as various data.
  • the memory 609 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 instructions required for at least one function (such as a sound playback function, Image playback function, etc.) etc.
  • memory 609 may include volatile memory or non-volatile memory, or memory 609 may include both volatile and non-volatile memory.
  • non-volatile memory can be read-only memory (Read-Only Memory, ROM), programmable read-only memory (Programmable ROM, PROM), erasable programmable read-only memory (Erasable PROM, EPROM), electrically removable memory. Erase programmable read-only memory (Electrically EPROM, EEPROM) or flash memory.
  • Volatile memory can be random access memory (Random Access Memory, RAM), static random access memory (Static RAM, SRAM), dynamic random access memory (Dynamic RAM, DRAM), synchronous dynamic random access memory (Synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (Double Data Rate SDRAM, DDRSDRAM), enhanced synchronous dynamic random access memory (Enhanced SDRAM, ESDRAM), synchronous link dynamic random access memory (Synch link DRAM) , SLDRAM) and direct memory bus random access memory (Direct Rambus RAM, DRRAM).
  • RAM Random Access Memory
  • SRAM static random access memory
  • DRAM dynamic random access memory
  • DRAM synchronous dynamic random access memory
  • SDRAM double data rate synchronous dynamic random access memory
  • Double Data Rate SDRAM Double Data Rate SDRAM
  • DDRSDRAM double data rate synchronous dynamic random access memory
  • Enhanced SDRAM, ESDRAM enhanced synchronous dynamic random access memory
  • Synch link DRAM synchronous link dynamic random access memory
  • SLDRAM direct memory bus
  • the processor 610 may include one or more processing units; optionally, the processor 610 integrates an application processor and a modem processor, where the application processor mainly handles operations related to the operating system, user interface, application programs, etc., Modem processors mainly process wireless communication signals, such as baseband processors. It can be understood that the above modem processor may not be integrated into the processor 610.
  • the terminal may be the first device:
  • the radio frequency unit 601 is configured to: receive demand information sent by the second device, where the demand information is used to indicate artificial intelligence Needs corresponding to AI services;
  • the radio frequency unit 601 is also configured to send feedback information on the demand information to the second device, where the feedback information is used to indicate whether the demand corresponding to the demand information is agreed to.
  • the demand information includes: demand items, or demand items and demand values corresponding to the demand items.
  • the feedback information includes any one of the following:
  • the first indication information is used to indicate that the requirement item is agreed to and the device capability meets the requirement value
  • Second indication information the second indication information is used to indicate agreement with the requirement item, and the second indication information carries device capabilities or capability difference information corresponding to the requirement item, and the capability difference information is used to indicate The difference between equipment capabilities and stated demand values;
  • the third indication information is used to indicate rejection of the requirement item.
  • the number of the requirement items is at least two
  • the second indication information carries the first requirement item and the device capability or capability difference information corresponding to the first requirement item
  • the first requirement item is The demand item among the at least two demand items that does not match the equipment capability.
  • the feedback information includes any of the following:
  • the fourth instruction information is used to indicate rejection of the requirement item
  • the fifth indication information is used to indicate agreement with the requirement item, and the fifth indication information carries the device capability corresponding to the requirement item.
  • the requirements include at least one of the following:
  • AI computing power requirements AI algorithm requirements
  • AI data requirements AI data requirements
  • the AI computing power requirements include at least one of the following:
  • the AI algorithm requirements include at least one of the following:
  • the AI task classification includes at least one of the following:
  • the AI learning framework includes at least one of the following:
  • the AI data requirements include at least one of the following:
  • the number of data input dimensions of the AI model is the number of data input dimensions of the AI model
  • the tag data is used to indicate at least one of the following:
  • SINR Signal to interference plus noise ratio
  • the precoding matrix indicates PMI
  • the tag data interval is used to indicate at least one of the following:
  • the AI model label delay is used to indicate at least one of the following:
  • the terminal after receiving the demand information sent by the second device, the terminal sends feedback information to the second device to avoid directly implementing the AI service demand and affecting the normal operation of the device itself.
  • Embodiments of the present application also provide a network-side device, including a processor and a communication interface.
  • the communication interface is used to: receive demand information sent by the second device, where the demand information is used to indicate requirements corresponding to artificial intelligence AI services. Send feedback information on the demand information to the second device, where the feedback information is used to indicate whether to agree to the demand corresponding to the demand information; or, the communication interface is used to: send demand information to the first device, The demand information is used to indicate the demand corresponding to the artificial intelligence AI service, and feedback information on the demand information sent by the first device is received, and the feedback information is used to indicate whether to agree to the demand corresponding to the demand information.
  • This network-side device embodiment corresponds to the above-mentioned first device-side method embodiment or the second device-side method embodiment.
  • Each implementation process and implementation manner of the above-mentioned method embodiment can be applied to this network-side device embodiment, And can achieve the same technical effect.
  • the embodiment of the present application also provides a network side device.
  • the network side device 700 includes: an antenna 701 , a radio frequency device 702 , a baseband device 703 , a processor 704 and a memory 705 .
  • the antenna 701 is connected to the radio frequency device 702 .
  • the radio frequency device 702 receives information through the antenna 701 and sends the received information to the baseband device 703 for processing.
  • the baseband device 703 processes the information to be sent and sends it to the radio frequency device 702.
  • the radio frequency device 702 processes the received information and then sends it out through the antenna 701.
  • the method performed by the network side device in the above embodiment can be implemented in the baseband device 703, which includes a baseband processor.
  • the baseband device 703 may include, for example, at least one baseband board on which multiple chips are disposed, as shown in FIG. Program to perform the network device operations shown in the above method embodiments.
  • the network side device may also include a network interface 706, which is, for example, a common public radio interface (CPRI).
  • a network interface 706, which is, for example, a common public radio interface (CPRI).
  • CPRI common public radio interface
  • the network side device 700 in this embodiment of the present invention also includes: instructions or programs stored in the memory 705 and executable on the processor 704.
  • the processor 704 calls the instructions or programs in the memory 705 to execute Figure 4 or Figure 5 Place It shows the execution method of each module and achieves the same technical effect. To avoid duplication, it will not be repeated here.
  • Embodiments of the present application also provide a readable storage medium.
  • Programs or instructions are stored on the readable storage medium.
  • the program or instructions are executed by a processor, each process of the above information feedback method embodiment is implemented, and the same can be achieved. The technical effects will not be repeated here to avoid repetition.
  • the processor is the processor in the terminal described in the above embodiment.
  • the readable storage medium includes computer readable storage media, such as computer read-only memory ROM, random access memory RAM, magnetic disk or optical disk, etc.
  • An embodiment of the present application further provides a chip.
  • the chip includes a processor and a communication interface.
  • the communication interface is coupled to the processor.
  • the processor is used to run programs or instructions to implement the above information feedback method embodiment. Each process can achieve the same technical effect. To avoid duplication, it will not be described again here.
  • chips mentioned in the embodiments of this application may also be called system-on-chip, system-on-a-chip, system-on-chip or system-on-chip, etc.
  • Embodiments of the present application further provide a computer program/program product.
  • the computer program/program product is stored in a storage medium.
  • the computer program/program product is executed by at least one processor to implement each of the above information feedback methods.
  • the process can achieve the same technical effect. To avoid repetition, it will not be described again here.
  • Embodiments of the present application also provide an information feedback system, including: a first device and a second device.
  • the first device can be used to perform the steps of the information feedback method on the first device side as described above.
  • the second device It may be used to perform the steps of the information receiving method on the second device side as described above.
  • the methods of the above embodiments can be implemented by means of software plus the necessary general hardware platform. Of course, it can also be implemented by hardware, but in many cases the former is better. implementation.
  • the technical solution of the present application can be embodied in the form of a computer software product that is essentially or contributes to the existing technology.
  • the computer software product is stored in a storage medium (such as ROM/RAM, disk , CD), including several instructions to cause a terminal (which can be a mobile phone, computer, server, air conditioner, or network device, etc.) to execute the methods described in various embodiments of this application.

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Quality & Reliability (AREA)
  • Databases & Information Systems (AREA)
  • Mobile Radio Communication Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

本申请公开了一种信息反馈方法、装置及设备，属于通信技术领域，本申请实施例的信息反馈方法包括：第一设备接收第二设备发送的需求信息，所述需求信息用于指示人工智能AI服务对应的需求；所述第一设备向所述第二设备发送对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。

Description

信息反馈方法、装置及设备
相关申请的交叉引用
本申请主张在2022年4月2日在中国提交的中国专利申请No.202210350469.2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，具体涉及一种信息反馈方法、装置及设备。
背景技术
无线移动通信结合人工智能(Artificial Intelligence，AI)能够较好地提高通信质量，例如，基于AI信道质量压缩、基于AI的波束管理、基于AI的定位。以波束管理为例，在毫米波无线通信中，通信收发端(如基站和终端)都配置了多个模拟波束，对于同一个终端，在不同的发送和接收模拟波束测量到信道质量是变化的。如何快速并准确地从所有可能的收发模拟波束组合中选择出信道质量最高的收发波束组，是影响传输质量的关键。在引入AI神经网络模型后，终端可以基于AI神经网络模型有效地预测信道质量最高的收发模拟波束，并上报给网络侧，从而能够获得更好的传输质量。
第一设备(例如终端、基站等)接收第二设备(例如核心网网元)发送的AI服务需求，并实现该AI服务需求，从而第二设备能够通过第一设备实现拆分的AI服务需求以实现复杂的AI服务。然而，第一设备实现AI服务需求时可能会影响设备自身的正常运行，使得第一设备无法正常工作。
发明内容
本申请实施例提供一种信息反馈方法、装置及设备，能够解决第一设备实现AI服务需求时可能会影响设备自身的正常运行，使得第一设备无法正常工作的问题。
第一方面，提供了一种信息反馈方法，包括：
第一设备接收第二设备发送的需求信息，所述需求信息用于指示人工智能AI服务对应的需求；
所述第一设备向所述第二设备发送对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。
第二方面，提供了一种信息反馈方法，包括：
第二设备向第一设备发送需求信息，所述需求信息用于指示人工智能AI服务对应的需求；
所述第二设备接收所述第一设备发送的对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。
第三方面，提供了一种信息反馈装置，第一设备包括所述信息反馈装置，包括：
接收模块，用于接收第二设备发送的需求信息，所述需求信息用于指示人工智能AI服务对应的需求；
发送模块，用于向所述第二设备发送对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。
第四方面，提供了一种信息反馈装置，第二设备包括所述信息反馈装置，包括：
发送模块，用于向第一设备发送需求信息，所述需求信息用于指示人工智能AI服务对应的需求；
接收模块，用于接收所述第一设备发送的对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。
第五方面，提供了一种第一设备，该第一设备包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如第一方面所述的方法的步骤。
第六方面，提供了一种第一设备，包括处理器及通信接口，其中，所述通信接口用于接收第二设备发送的需求信息，所述需求信息用于指示人工智能AI服务对应的需求；所述通信接口还用于向所述第二设备发送对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。
第七方面，提供了一种第二设备，该第二设备包括处理器和存储器，所述存储器存储可在所述处理器上运行的程序或指令，所述程序或指令被所述处理器执行时实现如第二方面所述的方法的步骤。
第八方面，提供了一种第二设备，包括处理器及通信接口，其中，所述通信接口用于向第一设备发送需求信息，所述需求信息用于指示人工智能AI服务对应的需求；所述通信接口还用于接收所述第一设备发送的对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。
第九方面，提供了一种信息反馈系统，包括：第一设备及第二设备，所述第一设备可用于执行如第一方面所述的信息反馈方法的步骤，所述第二设备可用于执行如第二方面所述的信息反馈方法的步骤。
第十方面，提供了一种可读存储介质，所述可读存储介质上存储程序或指令，所述程序或指令被处理器执行时实现如第一方面所述的方法的步骤，或者实现如第二方面所述的方法的步骤。
第十一方面，提供了一种芯片，所述芯片包括处理器和通信接口，所述通信接口和所述处理器耦合，所述处理器用于运行程序或指令，实现如第一方面所述的方法，或实现如第二方面所述的方法。
第十二方面，提供了一种计算机程序/程序产品，所述计算机程序/程序产品被存储在存储介质中，所述计算机程序/程序产品被至少一个处理器执行以实现如第一方面所述的信息反馈方法的步骤，或者，所述计算机程序/程序产品被至少一个处理器执行以实现如第二方面所述的信息反馈方法的步骤。
本申请实施例中，第一设备接收第二设备发送的需求信息，所述需求信息用于指示人工智能AI服务对应的需求；所述第一设备向所述第二设备发送对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。这样，第一设备接收第二设备发送的需求信息后，向第二设备发送对所述需求信息的反馈信息，能够避免直接实现该AI服务需求导致影响设备自身的正常运行。
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信息反馈方法的流程图之一；
图3是本申请实施例提供的一种信息反馈方法的流程图之二；
图4是本申请实施例提供的一种信息反馈装置的结构图之一；
图5是本申请实施例提供的一种信息反馈装置的结构图之二；
图6是本申请实施例提供的一种通信设备的结构图；
图7是本申请实施例提供的一种终端的结构示意图；
图8是本申请实施例提供的一种网络侧设备的结构示意图。
具体实施方式
下面将结合本申请实施例中的附图，对本申请实施例中的技术方案进行清楚描述，显然，所描述的实施例是本申请一部分实施例，而不是全部的实施例。基于本申请中的实施例，本领域普通技术人员所获得的所有其他实施例，都属于本申请保护的范围。
本申请的说明书和权利要求书中的术语“第一”、“第二”等是用于区别类似的对象，而不用于描述特定的顺序或先后次序。应该理解这样使用的术语在适当情况下可以互换，以便本申请的实施例能够以除了在这里图示或描述的那些以外的顺序实施，且“第一”、“第二”所区别的对象通常为一类，并不限定对象的个数，例如第一对象可以是一个，也可以是多个。此外，说明书以及权利要求中“和/或”表示所连接对象的至少其中之一，字符“/”一般表示前后关联对象是一种“或”的关系。
值得指出的是，本申请实施例所描述的技术不限于长期演进型(Long Term Evolution，LTE)/LTE的演进(LTE-Advanced，LTE-A)系统，还可用于其他无线通信系统，诸如码分多址(Code Division Multiple Access，CDMA)、时分多址(Time Division Multiple Access，TDMA)、频分多址(Frequency Division Multiple Access，FDMA)、正交频分多址(Orthogonal Frequency Division Multiple Access，OFDMA)、单载波频分多址(Single-carrier Frequency  Division Multiple Access，SC-FDMA)和其他系统。本申请实施例中的术语“系统”和“网络”常被可互换地使用，所描述的技术既可用于以上提及的系统和无线电技术，也可用于其他系统和无线电技术。以下描述出于示例目的描述了新空口(New Radio，NR)系统，并且在以下大部分描述中使用NR术语，但是这些技术也可应用于NR系统应用以外的应用，如第6代(6th Generation，6G)通信系统。
图1示出本申请实施例可应用的一种无线通信系统的框图。无线通信系统包括终端11和网络侧设备12。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(Pedestrian 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。
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信息反馈方法、装置及设备进行详细地说明。
参见图2，图2是本申请实施例提供的一种信息反馈方法的流程图，如图2所示，信息反馈方法包括以下步骤：
步骤101、第一设备接收第二设备发送的需求信息，所述需求信息用于指示人工智能AI服务对应的需求。
其中，AI服务可以包括模型训练、模型微调(fine-tuning)、模型推理及模型验证等与AI神经网络模型相关的服务。需要说明的是，对于基于仿真数据或者其他小区采集的样本数据训练得到的AI神经网络模型，由于泛化性的问题，如果直接在终端所在的小区推理往往得不到很好的性能。可以通过终端基于终端所在的小区数据对AI神经网络模型再进行训练或者fine-tuning等AI服务提高AI神经网络模型的性能。
一种实施方式中，AI服务可以包括模型训练、fine-tuning、模型推理及模型验证中的至少一项。
另外，第一设备可以为终端，或者基站，或者核心网网元等。第二设备可以为基站，或者自组织网络(Self-Organized Networks，SON)，或者网管系统(Operation Administration and Maintenance，OAM)，或者核心网网元(例如，网络数据分析功能(Network Data Analytics Function，NWDAF))，或者网管等。
步骤102、所述第一设备向所述第二设备发送对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。
其中，该反馈信息可以是能力反馈信息，示例地，该反馈信息可以是智能能力反馈信息，该智能能力反馈信息用于指示第一设备的智能能力。
需要说明的是，第一设备可以用于AI服务执行及智能能力上报，第二设备可以用于AI服务编排部署。通过AI服务编排部署可以将AI服务拆分为AI算力、AI算法、AI数据三个维度。本申请实施例中，第一设备接收到AI服务对应的需求信息，根据自身能力做出评估，并上报接受、或拒绝、或需求差异。从而在AI服务执行端可以在自身能力范围内执行AI服务编排部署节点分配的AI服务，从而第一设备既可以提供高质量的AI服务，又可以确保自身其他流程的正常运行。
一种实施方式中，第一设备接收第二设备发送的需求信息，该需求信息包括AI算力，和/或，AI算法，和/或，AI数据方面的需求。第一设备可以向所述第二设备发送对所述需求信息的反馈信息，该反馈信息可以包括接受、拒绝或者能力差异信息，该能力差异信息可以用于指示设备能力与所述需求信息的差异。示例地，第一设备可以根据自身决策向所述第二设备发送对所述需求信息的反馈信息。
需要说明的是，第一设备可以为终端，若终端当前状态不符合预期，终端接受分配的 AI服务，可能会影响终端其他流程的正常运行，并且，也不能保障终端对AI服务的高质量完成。本申请实施例中，终端接收AI服务编排部署设备发送的需求信息，所述需求信息用于指示人工智能AI服务对应的需求；并向AI服务编排部署设备发送对所述需求信息的反馈信息。从而，终端在接受AI服务编排部署设备发送的AI服务前，能够综合考虑AI服务的需求和设备自身的智能能力，并进行反馈，能够从AI服务部署的流程来解决智能能力反馈的问题。
一种实施方式中，需求信息包括AI数据需求，反馈信息可以用于反馈AI数据需求差异，示例地，反馈信息可以包括需求信息中AI数据需求中无法满足的信息条目，和/或，需求信息中AI数据需求中无法满足的信息条目对应的第一设备的当前设备能力。
一种实施方式中，需求信息包括AI算力需求，反馈信息可以用于反馈AI算力需求差异，示例地，反馈信息可以包括需求信息中AI算力需求中无法满足的信息条目，和/或，需求信息中AI算力需求中无法满足的信息条目对应的第一设备的当前设备能力。
一种实施方式中，需求信息包括AI算法需求，反馈信息可以用于反馈AI算法需求差异，示例地，反馈信息可以包括需求信息中AI算法需求中无法满足的信息条目，和/或，需求信息中AI算法需求中无法满足的信息条目对应的第一设备的当前设备能力。
需要说明的是，第一设备可以结合设备自身算力、算法、数据等各方面的设备能力，判断自身设备能力是否满足需求信息，以及是否接受该需求信息对应的需求，并确定反馈信息。示例地，若第一设备的设备能力满足需求信息，并愿意接受该需求信息对应的需求，则反馈信息可以用于指示接受该需求；和/或，若第一设备不愿意接受该需求信息对应的需求，则反馈信息可以用于指示拒绝该需求；和/或，若第一设备的设备能力不满足需求信息，但愿意接受该需求信息对应的需求，则反馈信息可以用于指示能力差异信息。
本申请实施例中，第一设备接收第二设备发送的需求信息，所述需求信息用于指示人工智能AI服务对应的需求；所述第一设备向所述第二设备发送对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。这样，第一设备接收第二设备发送的需求信息后，向第二设备发送对所述需求信息的反馈信息，能够避免直接实现该AI服务需求导致影响设备自身的正常运行。
可选地，所述需求信息包括：需求项，或者，需求项及所述需求项对应的需求值。
其中，所述需求项可以用于指示AI服务对应的需求，示例地，可以包括AI算力需求、AI算法需求及AI数据需求等等。
一种实施方式中，第一设备接收第二设备发送的需求信息，该需求信息包括AI算力，和/或，AI算法，和/或，AI数据方面的需求项，及需求项对应的需求值。第一设备可以向所述第二设备发送对所述需求信息的反馈信息，该反馈信息可以用于指示接受、拒绝或者能力差异信息。
一种实施方式中，第一设备接收第二设备发送的需求信息，该需求信息包括AI算力，和/或，AI算法，和/或，AI数据方面的需求项。第一设备可以向所述第二设备发送对所述 需求信息的反馈信息，该反馈信息可以用于指示拒绝或者与需求项对应的设备能力。需求项可以采用需求条目的方式表示，示例地，该需求信息包括AI算力需求条目，反馈信息可以包括AI算力需求条目值，该AI算力需求条目值赋值为与AI算力需求条目对应的设备能力；和/或，该需求信息包括AI算法需求条目，反馈信息可以包括AI算法需求条目值，该AI算法需求条目值赋值为与AI算法需求条目对应的设备能力；和/或，该需求信息包括AI数据需求条目，反馈信息可以包括AI数据需求条目值，该AI数据需求条目值赋值为与AI数据需求条目对应的设备能力。
另外，在所述需求信息包括需求项的情况下，第一设备可以结合设备自身状态，判断是否愿意接受该需求项对应的需求，确定反馈信息。示例地，若第一设备不愿意接受该需求项对应的需求，则反馈信息可以指示拒绝；若第一设备愿意接受该需求项对应的需求，则反馈信息可以指示需求项对应的当前设备能力。
需要说明的是，第二设备可以对AI服务从AI算力、AI算法及AI数据三个维度进行拆分，得到该三个维度上的需求项。AI算力维度上的需求项可以包括如下至少一项：算力；存储；总计算量；执行时间限制；耗电量。AI算法维度上的需求项可以包括如下至少一项：AI任务分类；AI学习框架；AI网络开发环境；AI基础模型库。AI数据维度上的需求项可以包括如下至少一项：标签数据；标签维度个数；标签维度；标签维度的顺序；标签维度的数据个数；标签数据间隔；AI模型输入数据；AI模型输入维度个数；AI模型输入维度；AI模型输入维度的顺序；AI模型输入维度的数据个数；AI模型输入数据收集间隔；AI模型标签时延。
该实施方式中，所述需求信息包括需求项，从而能够通过需求项指示AI服务对应的需求，第一设备能够通过该需求项向第二设备反馈该需求项对应的设备能力；或者，所述需求信息包括需求项及所述需求项对应的需求值，从而能够通过需求项及需求值指示AI服务对应的需求，第二设备能够通过该需求项及需求值向第二设备反馈该需求项对应的设备能力能否满足该需求值。
可选地，在所述需求信息包括需求项及所述需求项对应的需求值的情况下，所述反馈信息包括如下任意一项：
第一指示信息，所述第一指示信息用于指示同意所述需求项，且设备能力满足所述需求值；
第二指示信息，所述第二指示信息用于指示同意所述需求项，且所述第二指示信息携带与所述需求项对应的设备能力或能力差异信息，所述能力差异信息用于指示设备能力与所述需求值的差异；
第三指示信息，所述第三指示信息用于指示拒绝所述需求项。
一种实施方式中，所述第一设备向所述第二设备发送对所述需求信息的反馈信息，可以包括：
在所述第一设备的设备能力满足所述需求值，且所述第一设备同意所述需求项的情况 下，向所述第二设备发送所述第一指示信息；和/或
在所述第一设备的设备能力不满足所述需求值，且所述第一设备同意满足所述需求项对应的部分需求值的情况下，向所述第二设备发送所述第二指示信息；和/或
在所述第一设备的设备能力不满足所述需求值，且所述第一设备不同意所述需求项的情况下，向所述第二设备发送所述第三指示信息。
一种实施方式中，所述第一设备向所述第二设备发送对所述需求信息的反馈信息，可以包括：
在所述第一设备的设备能力满足所述需求值，且所述第一设备在满足所述需求值的同时可维持所述第一设备的正常运行的情况下，向所述第二设备发送所述第一指示信息；和/或
在所述第一设备的设备能力不满足所述需求值，且所述第一设备在满足所述需求项对应的部分需求值的同时可维持所述第一设备的正常运行的情况下，向所述第二设备发送所述第二指示信息；和/或
在所述第一设备的设备能力不满足所述需求值，且所述第一设备仅能维持所述第一设备的正常运行的情况下，向所述第二设备发送所述第三指示信息。
一种实施方式中，第一设备向第二设备发送第一指示信息后，可以直接执行该需求项对应的AI服务；或者，第一设备可以在接收到第二设备发送的用于指示确定执行的指示信息后再执行该需求项对应的AI服务。
一种实施方式中，第一设备向第二设备发送第二指示信息后，可以直接执行该需求项对应的AI服务；或者，第一设备可以在接收到第二设备发送的用于指示确定执行的指示信息后再执行该需求项对应的AI服务；或者，第一设备可以在接收到第二设备发送的用于指示修改需求值的指示信息后再执行该需求项对应的AI服务；等等，本实施方式对此不进行限定。示例地，第二设备在接收到第一设备发送的第二指示信息后，可以向第一设备发送修改需求值的指示信息，将需求值修改为与设备能力匹配的需求值。
一种实施方式中，第一设备向第二设备发送第三指示信息，即第一设备拒绝执行该需求项对应的AI服务。
该实施方式中，通过第一指示信息、第二指示信息或第三指示信息，第一设备能够向第二设备反馈是否同意或拒绝AI服务对应的需求项，并向第二设备反馈该需求项对应的设备能力能否满足该需求值，便于第一设备在自身能力范围内执行分配的AI服务。
可选地，所述需求项的数量为至少两个，所述第二指示信息携带第一需求项及与所述第一需求项对应的设备能力或能力差异信息，所述第一需求项为所述至少两个需求项中与所述设备能力不匹配的需求项。
其中，第一需求项对应的能力差异信息可以用于指示设备能力与所述第一需求项对应的需求值的差异。以第一需求项为AI算力需求，AI算力需求包括算力，且AI算力需求对应的算力值为20T每秒浮点计算次数(Floating point operations per second，FLOPs)为 例，第一设备确定当前设备的实际算力为15T FLOPs，则第一需求项对应的能力差异信息为5T FLOPs，第一需求项对应的设备能力为15T FLOPs，第二指示信息可以携带能力差异信息5T FLOPs，或者设备能力15T FLOPs。第一需求项与设备能力不匹配，可以认为是，第一需求项对应的需求值与设备能力不匹配，示例地，可以是第一需求项对应的需求值大于设备能力。
该实施方式中，通过所述第二指示信息携带的第一需求项及与所述第一需求项对应的设备能力或能力差异信息，从而第一设备在接收到AI服务对应的需求后，能够将自身设备能力不能满足需求值的第一需求项反馈至第二设备，并反馈第一需求项对应的能力差异信息。
可选地，在所述需求信息包括需求项的情况下，所述反馈信息包括如下任意一项：
第四指示信息，所述第四指示信息用于指示拒绝所述需求项；
第五指示信息，所述第五指示信息用于指示同意所述需求项，且所述第五指示信息携带与所述需求项对应的设备能力。
其中，所述需求项对应的设备能力可以为所述需求项对应的当前设备能力。以需求项包括算力为例，所述需求项对应的设备能力可以包括设备当前算力。
一种实施方式中，所述第一设备向所述第二设备发送对所述需求信息的反馈信息，可以包括：
在所述第一设备不同意所述需求项的情况下，向所述第二设备发送所述第四指示信息；和/或
在所述第一设备同意所述需求项的情况下，向所述第二设备发送所述第五指示信息。
一种实施方式中，所述第一设备向所述第二设备发送对所述需求信息的反馈信息，可以包括：
在所述第一设备的设备能力仅可维持所述第一设备的正常运行的情况下，向所述第二设备发送所述第四指示信息；
和/或
在所述第一设备的设备能力可维持所述第一设备的正常运行，且可满足所述需求项的情况下，向所述第二设备发送所述第五指示信息。
一种实施方式中，第一设备向第二设备发送第四指示信息，即第一设备拒绝执行该需求项对应的AI服务。
一种实施方式中，第一设备向第二设备发送第五指示信息后，可以接收第二设备发送的按第一设备的设备能力执行所述需求项对应的AI服务的指令，并按照第一设备的当前设备能力执行所述需求项对应的AI服务。第二设备可以根据第一设备上报的设备能力调整AI服务的任务分配。
该实施方式中，通过第四指示信息或第五指示信息，第一设备能够向第二设备反馈是否同意或拒绝AI服务对应的需求项，并向第二设备反馈与所述需求项对应的设备能力。
可选地，所述需求项包括如下至少一项：
AI算力需求；AI算法需求；AI数据需求。
一种实施方式中，第一设备接收第二设备发送的需求信息，包括，第一设备接收第二设备发送的AI算力需求，和/或，AI算力需求及AI算力需求对应的需求值；
所述第一设备向所述第二设备发送对所述需求信息的反馈信息，包括，第一设备向第二设备发送AI算力需求的反馈信息，该反馈信息用于指示是否同意AI算力需求，且该反馈信息还包括与所述AI算力需求对应的设备能力，或者，与所述AI算力需求对应的能力差异信息。从而能够实现对第二设备向第一设备请求的AI算力需求的设备能力反馈。
一种实施方式中，第一设备接收第二设备发送的需求信息，包括，第一设备接收第二设备发送的AI算法需求，和/或，AI算法需求及AI算法需求对应的需求值；
所述第一设备向所述第二设备发送对所述需求信息的反馈信息，包括，第一设备向第二设备发送AI算法需求的反馈信息，该反馈信息用于指示是否同意AI算法需求，且该反馈信息还包括与所述AI算法需求对应的设备能力，或者，与所述AI算法需求对应的能力差异信息。从而能够实现对第二设备向第一设备请求的AI算法需求的设备能力反馈。
一种实施方式中，第一设备接收第二设备发送的需求信息，包括，第一设备接收第二设备发送的AI数据需求，和/或，AI数据需求及AI数据需求对应的需求值；
所述第一设备向所述第二设备发送对所述需求信息的反馈信息，包括，第一设备向第二设备发送AI数据需求的反馈信息，该反馈信息用于指示是否同意AI数据需求，且该反馈信息还包括与所述AI数据需求对应的设备能力，或者，与所述AI数据需求对应的能力差异信息。从而能够实现对第二设备向第一设备请求的AI数据需求的设备能力反馈。
该实施方式中，第一设备接收第二设备发送的AI算力需求、AI算法需求、AI数据需求中的至少一项，向第二设备发送需求项对应的反馈信息，有利于第一设备为第二设备提供质量较高的AI服务。
可选地，所述AI算力需求包括如下至少一项：
算力；
存储；
总计算量；
执行时间限制；
耗电量。
其中，AI算力需求中的算力可以用于指示AI服务所需要消耗的算力，算力可以用乘加次数/秒表示。AI算力需求中的存储可以用于指示AI服务所需要消耗的存储量。AI算力需求中的总计算量可以用于指示执行AI服务的总计算量，该总计算量可以通过算力和AI模型的推理时间计算得到。AI算力需求中的执行时间限制可以用于指示执行AI服务的时间限制。AI算力需求中的耗电量可以用于指示AI服务所需要消耗的电量。
一种实施方式中，第一设备接收第二设备发送的需求信息，该需求信息包括算力、存 储、总计算量、执行时间限制、耗电量中的至少一项算力需求，或者至少一项算力需求及对应的需求值；
所述第一设备根据设备能力向所述第二设备发送对所述需求信息的反馈信息，该反馈信息可以用于指示是否同意所述至少一项算力需求，且所述反馈信息可以携带与所述算力需求对应的设备能力或者设备能力与需求值之间的能力差异信息。
一种实施方式中，第一设备接收第二设备发送的需求信息，该需求信息包括AI算法需求及AI算力需求，该AI算法需求包括模型训练、模型推理、模型验证、模型监视、模型部署中的至少一项AI任务，AI算力需求包括执行该至少一项AI任务所需的算力、存储、总计算量、执行时间限制及耗电量中的至少一项算力需求或者至少一项算力需求及对应的需求值；
所述第一设备向所述第二设备发送对所述需求信息的反馈信息，可以包括，第一设备根据设备能力判断是否能够满足执行该至少一项AI任务所需的AI算力需求，在判断能够满足执行该至少一项AI任务所需的AI算力需求的情况下，向第二设备发送第一指示信息；在判断仅能够满足执行该至少一项AI任务所需的部分AI算力需求的情况下，向第二设备发送第二指示信息；在判断仅能够维持第一设备的正常运行的情况下，向第二设备发送第三指示信息。
一种实施方式中，第一设备接收第二设备发送的需求信息，该需求信息包括AI算法需求及AI算力需求，该AI算法需求包括模型训练、模型推理、模型验证、模型监视、模型部署中的至少一项AI任务，AI算力需求包括执行该至少一项AI任务所需的算力、存储、总计算量、执行时间限制及耗电量中的至少一项算力需求；
所述第一设备向所述第二设备发送对所述需求信息的反馈信息，可以包括，第一设备根据设备能力判断在维持设备的正常运行的情况下是否能够执行该至少一项AI任务，在判断在维持设备的正常运行的情况下能够执行该至少一项AI任务的情况下，向第二设备发送第五指示信息，该第五指示信息携带与所述至少一项算力需求对应的设备能力；在判断在维持设备的正常运行的情况下不能够执行该至少一项AI任务的情况下，向第二设备发送第四指示信息。
该实施方式中，通过AI算力需求中的算力、存储、总计算量、执行时间限制、耗电量中的至少一个需求项，或需求项及需求项对应的需求值，能够便于第一设备判断自身设备能力能否满足AI算力需求。
可选地，所述AI算法需求包括如下至少一项：
AI任务分类；
AI学习框架；
AI网络开发环境；
AI基础模型库。
其中，AI任务分类可以用于指示AI算法对应的任务类型。AI学习框架可以用于指示 AI算法的学习框架，具体包括学习方法和训练方法。AI网络开发环境可以用于指示AI算法的网络开发环境。AI基础模型库可以用于指示AI算法的基础模型库。
一种实施方式中，所述需求信息可以包括执行AI任务所需的AI学习框架、AI网络开发环境、AI基础模型库中的至少一项需求项及需求项对应的需求值。所述第一设备向所述第二设备发送对所述需求信息的反馈信息，可以是，第一设备根据设备能力判断是否同意所述需求信息对应的需求，向第二设备发送反馈信息，该反馈信息用于指示是否同意所述需求信息对应的需求。示例地，以需求信息包括AI学习框架为例，可以在第一设备的AI算法为该AI学习框架时，同意所述需求信息对应的需求。
该实施方式中，通过AI算法需求中的AI任务分类、AI学习框架、AI网络开发环境、AI基础模型库中的至少一个需求项，或需求项及需求项对应的需求值，能够便于第一设备判断自身设备能力能否满足AI算法需求。
可选地，所述AI任务分类包括如下至少一项：
模型训练；
模型推理；
模型验证；
模型监视；
模型部署。
一种实施方式中，第一设备接收第二设备发送的需求信息，该需求信息包括模型训练、模型推理、模型验证、模型监视、模型部署中的至少一项AI任务，在所述第一设备的设备能力仅可维持所述第一设备的正常运行的情况下，向所述第二设备发送同意所述需求信息对应的需求的反馈信息；在所述第一设备的设备能力可维持所述第一设备的正常运行，且可进行所述AI任务的情况下，向所述第二设备发送同意所述需求信息对应的需求的反馈信息。
一种实施方式中，第一设备接收第二设备发送的需求信息，该需求信息包括模型训练、模型推理、模型验证、模型监视、模型部署中的至少一项AI任务，及执行该AI任务所需的AI算法需求和/或AI数据需求；在第一设备的设备能力能够满足该AI算法需求和/或AI数据需求的情况下，第一设备同意执行所述至少一项AI任务，向第二设备发送对所述需求信息的反馈信息，该反馈信息指示同意所述需求信息对应的需求；在第一设备的设备能力能够部分满足该AI算法需求和/或AI数据需求的情况下，第一设备同意执行所述至少一项AI任务，向第二设备发送对所述需求信息的反馈信息，该反馈信息指示同意所述需求信息对应的需求，并反馈能力差异信息；在第一设备的设备能力不能够满足该AI算法需求和/或AI数据需求的情况下，第一设备拒绝执行所述至少一项AI任务。
这样，通过对第二设备下发的模型训练、模型推理、模型验证、模型监视、模型部署等AI任务的需求的反馈，能够提高第一设备执行模型训练、模型推理、模型验证、模型监视、模型部署等AI任务的质量。
可选地，所述AI学习框架包括如下至少一项：
有监督深度学习；
无监督深度学习；
元学习；
迁移学习；
强化学习；
联邦学习。
需要说明的是，所述需求信息可以包括执行AI任务所需的AI学习框架，该AI学习框架为有监督深度学习、无监督深度学习、元学习、迁移学习、强化学习及联邦学习中的至少一项，可以在第一设备的AI算法为需求信息中指示的AI学习框架的情况下，同意所述需求信息对应的需求；在第一设备的AI算法不为需求信息中指示的AI学习框架的情况下，拒绝所述需求信息对应的需求。
可选地，所述AI数据需求包括如下至少一项：
标签数据；
标签维度个数；
标签维度；
标签维度的顺序；
标签维度的数据个数；
标签数据间隔；
AI模型输入数据；
AI模型输入维度个数；
AI模型输入维度；
AI模型输入维度的顺序；
AI模型输入维度的数据个数；
AI模型输入数据收集间隔；
AI模型标签时延。
其中，标签数据可以包括波束信道的参考信号接收功率(Reference Signal Received Power，RSRP)、参考信号接收功率RSRP、信号与干扰加噪声比(Signal-to-noise and Interference Ratio，SINR)等与第一设备的通信相关的数据。标签维度个数可以用于指示需要收集的标签数据的维度个数。标签维度可以用于指示需要收集的标签数据的维度。标签维度的顺序可以用于指示需要收集的标签数据各维度的顺序。标签维度的数据个数可以用于指示需要收集的标签数据的各个维度上的数据个数。标签数据间隔可以用于指示需要收集的标签数据的数据间隔。AI模型输入数据可以用于指示AI模型的输入数据。AI模型输入维度个数可以用于指示需要收集的AI模型输入数据的维度个数。AI模型输入维度可以用于指示需要收集的AI模型输入数据的维度。AI模型输入维度的顺序可以用于指示需 要收集的AI模型输入数据的各个维度的顺序。AI模型输入维度的数据个数可以用于指示需要收集的AI模型输入数据的各个维度的数据个数。AI模型输入数据收集间隔用于指示需要收集的AI模型输入数据的数据间隔。AI模型标签时延可以用于指示获得标签数据与推理结束或接收AI模型输入数据之间的时延。通过上述AI数据需求，能够便于第一设备判断自身设备能力能否满足AI数据需求。
一种实施方式中，所述需求信息可以包括执行AI任务所需的标签数据、标签维度个数、标签维度、标签维度的顺序、标签维度的数据个数、标签数据间隔、AI模型标签时延中的至少一项标签数据信息，第一设备可以通过该至少一项标签数据信息确定执行AI任务所需的标签数据。以AI任务为模型训练为例，该标签数据可以为用于模型训练的标签数据。第一设备根据设备能力判断能否获取执行该AI任务所需的标签数据，若能够获取到执行该AI任务所需的标签数据，则同意所述需求信息对应的需求；若不能够获取到执行该AI任务所需的标签数据，则拒绝所述需求信息对应的需求。从而第一设备能够根据自身设备能力收集标签数据。
一种实施方式中，所述需求信息可以包括执行AI任务所需的AI模型输入数据、AI模型输入维度个数、AI模型输入维度、AI模型输入维度的顺序、AI模型输入维度的数据个数、AI模型输入数据收集间隔中的至少一项AI模型输入数据信息，第一设备可以通过该至少一项AI模型输入数据信息确定执行AI任务所需的AI模型输入数据。以AI任务为模型推理为例，该AI模型输入数据可以为用于模型推理的输入数据。第一设备根据设备能力判断能否获取执行该AI任务所需的AI模型输入数据，若能够获取到执行该AI任务所需的AI模型输入数据，则同意所述需求信息对应的需求；若不能够获取到执行该AI任务所需的AI模型输入数据，则拒绝所述需求信息对应的需求。从而第一设备能够根据自身设备能力获取AI模型输入数据。
可选地，所述标签数据用于指示如下至少一项：
波束信道的参考信号接收功率(Reference Signal Received Power，RSRP)；
波束信道的参考信号接收质量(Reference Signal Received Quality，RSRQ)；
波束信道的信号与干扰加噪声比(Signal-to-noise and interference ratio，SINR)；
小区信道的RSRP；
小区信道的RSRQ；
小区信道的SINR；
小区信道的接收信号强度指示(Received Signal Strength Indication，RSSI)；
小区信道冲激响应；
预编码矩阵指示(Precoding matrix indicator，PMI)；
秩指示(Rank indicator，RI)；
信道质量指示(Channel quality indicator，CQI)。
可选地，所述标签数据间隔用于指示如下至少一项：
时间间隔；
频率间隔；
时延间隔；
相位间隔；
多普勒间隔；
波束间隔。
可选地，所述AI模型标签时延用于指示如下至少一项：
推理结束与获得标签数据之间的时延；
接收AI模型输入数据与获得标签数据之间的时延。
其中，所述推理结束与获得标签数据之间的时延，可以是，第一设备推理结束与第一设备获得标签数据之间的时延。所述接收AI模型输入数据与获得标签数据之间的时延，可以是，第一设备接收AI模型输入数据与第一设备获得标签数据之间的时延。
以下通过三个具体的实施例对本申请实施例的信息反馈方法进行说明：
实施例1：
第一设备可以为终端，第二设备可以为核心网网元，AI服务可以为模型训练。
终端接收核心网网元发送的需求信息，该需求信息包括AI数据需求，且该需求信息中的AI数据需求指定AI模型输入维度个数为2个维度，分别为发送波束维度和时间维度，其中，发送波束维度个数为64。
终端可以测量到当前发送波束维度为8，终端愿意接受该需求信息，但需要反馈目前数据收集情况与需求信息中有差异的条目：发送波束维度个数，以及终端当前的发送波束维度个数：8。终端向核心网网元发送反馈信息，该反馈信息包括该差异的条目：发送波束维度个数，以及终端当前的发送波束维度个数：8。
在该实施例中，终端向核心网网元上报与AI数据需求的能力差异信息。
实施例2：
第一设备可以为终端，第二设备可以为核心网网元，AI服务可以为模型训练。
终端接收核心网网元发送的需求信息，该需求信息包括AI算力需求，且该需求信息中的AI算力需求定义的算力为20T FLOPs。
终端根据自身能力评估实际算力为15T FLOPs，此时，终端愿意接受该需求信息，但由于自身能力无法支持该需求信息。终端向核心网网元发送反馈信息，该反馈信息包括：与需求信息中有差异的条目：算力，和当前的实际算力15T FLOPs。
在该实施例中，终端向核心网网元上报与AI算力需求的能力差异信息。
实施例3：
第一设备可以为终端，第二设备可以为核心网网元，AI服务可以为模型训练。
终端接收核心网网元发送的需求信息，该需求信息包括AI算法需求，且该需求信息中的AI算法需求定义基础模型为卷积神经网络模型。
终端当前存储的模型只有全连接模型，终端愿意接受该需求信息，但由于自身能力无法支持该需求信息。终端向核心网网元发送反馈信息，该反馈信息包括算法差异条目：基础模型，及当前值：全连接模型。
在该实施例中，终端向核心网网元上报与AI算法需求的能力差异信息。
参见图3，图3是本申请实施例提供的一种信息反馈方法的流程图，如图3所示，信息反馈方法包括以下步骤：
步骤201、第二设备向第一设备发送需求信息，所述需求信息用于指示人工智能AI服务对应的需求；
步骤202、所述第二设备接收所述第一设备发送的对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。
可选地，所述需求信息包括：需求项，或者，需求项及所述需求项对应的需求值。
可选地，在所述需求信息包括需求项及所述需求项对应的需求值的情况下，所述反馈信息包括如下任意一项：
第一指示信息，所述第一指示信息用于指示同意所述需求项，且设备能力满足所述需求值；
第二指示信息，所述第二指示信息用于指示同意所述需求项，且所述第二指示信息携带与所述需求项对应的设备能力或能力差异信息，所述能力差异信息用于指示设备能力与所述需求值的差异；
第三指示信息，所述第三指示信息用于指示拒绝所述需求项。
可选地，所述需求项的数量为至少两个，所述第二指示信息携带第一需求项及与所述第一需求项对应的设备能力或能力差异信息，所述第一需求项为所述至少两个需求项中与所述设备能力不匹配的需求项。
可选地，在所述需求信息包括需求项的情况下，所述反馈信息包括如下任意一项：
第四指示信息，所述第四指示信息用于指示拒绝所述需求项；
第五指示信息，所述第五指示信息用于指示同意所述需求项，且所述第五指示信息携带与所述需求项对应的设备能力。
可选地，所述需求项包括如下至少一项：
AI算力需求；AI算法需求；AI数据需求。
需要说明的是，本实施例作为与图2所示的实施例中对应的第二设备的实施方式，其具体的实施方式可以参见图2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。这样，能够避免第一设备直接实现该AI服务需求导致影响设备自身的正常运行。
本申请实施例提供的信息反馈方法，执行主体可以为信息反馈装置。本申请实施例中以信息反馈装置执行信息反馈方法为例，说明本申请实施例提供的信息反馈的装置。
请参见图4，图4是本申请实施例提供的一种信息反馈装置的结构图，第一设备包括所述信息反馈装置，如图4所示，信息反馈装置300包括：
接收模块301，用于接收第二设备发送的需求信息，所述需求信息用于指示人工智能AI服务对应的需求；
发送模块302，用于向所述第二设备发送对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。
可选地，所述需求信息包括：需求项，或者，需求项及所述需求项对应的需求值。
可选地，在所述需求信息包括需求项及所述需求项对应的需求值的情况下，所述反馈信息包括如下任意一项：
第一指示信息，所述第一指示信息用于指示同意所述需求项，且设备能力满足所述需求值；
第二指示信息，所述第二指示信息用于指示同意所述需求项，且所述第二指示信息携带与所述需求项对应的设备能力或能力差异信息，所述能力差异信息用于指示设备能力与所述需求值的差异；
第三指示信息，所述第三指示信息用于指示拒绝所述需求项。
可选地，所述需求项的数量为至少两个，所述第二指示信息携带第一需求项及与所述第一需求项对应的设备能力或能力差异信息，所述第一需求项为所述至少两个需求项中与所述设备能力不匹配的需求项。
可选地，在所述需求信息包括需求项的情况下，所述反馈信息包括如下任意一项：
第四指示信息，所述第四指示信息用于指示拒绝所述需求项；
第五指示信息，所述第五指示信息用于指示同意所述需求项，且所述第五指示信息携带与所述需求项对应的设备能力。
可选地，所述需求项包括如下至少一项：
AI算力需求；AI算法需求；AI数据需求。
可选地，所述AI算力需求包括如下至少一项：
算力；
存储；
总计算量；
执行时间限制；
耗电量。
可选地，所述AI算法需求包括如下至少一项：
AI任务分类；
AI学习框架；
AI网络开发环境；
AI基础模型库。
可选地，所述AI任务分类包括如下至少一项：
模型训练；
模型推理；
模型验证；
模型监视；
模型部署。
可选地，所述AI学习框架包括如下至少一项：有监督深度学习；
无监督深度学习；
元学习；
迁移学习；
强化学习；
联邦学习。
可选地，所述AI数据需求包括如下至少一项：标签数据；
标签维度个数；
标签维度；
标签维度的顺序；
标签维度的数据个数；
标签数据间隔；
AI模型输入数据；
AI模型输入维度个数；
AI模型输入维度；
AI模型输入维度的顺序；
AI模型输入维度的数据个数；
AI模型输入数据收集间隔；
AI模型标签时延。
可选地，所述标签数据用于指示如下至少一项：
波束信道的参考信号接收功率RSRP；
波束信道的参考信号接收质量RSRQ；
波束信道的信号与干扰加噪声比SINR；
小区信道的RSRP；
小区信道的RSRQ；
小区信道的SINR；
小区信道的接收信号强度指示RSSI；
小区信道冲激响应；
预编码矩阵指示PMI；
秩指示RI；
信道质量指示CQI。
可选地，所述标签数据间隔用于指示如下至少一项：
时间间隔；
频率间隔；
时延间隔；
相位间隔；
多普勒间隔；
波束间隔。
可选地，所述AI模型标签时延用于指示如下至少一项：
推理结束与获得标签数据之间的时延；
接收AI模型输入数据与获得标签数据之间的时延。
本申请实施例中的信息反馈装置，接收第二设备发送的需求信息后，向第二设备发送对所述需求信息的反馈信息，能够避免直接实现该AI服务需求导致影响设备自身的正常运行。
本申请实施例中的信息反馈装置可以是电子设备，例如具有操作系统的电子设备，也可以是电子设备中的部件，例如集成电路或芯片。该电子设备可以是终端，也可以为除终端之外的其他设备。示例性的，终端可以包括但不限于上述所列举的终端11的类型，其他设备可以为服务器、网络附属存储器(Network Attached Storage，NAS)等，本申请实施例不作具体限定。
本申请实施例提供的信息反馈装置能够实现图2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5，图5是本申请实施例提供的一种信息反馈装置的结构图，第二设备包括所述信息反馈装置，如图5所示，信息反馈装置400包括：
发送模块401，用于向第一设备发送需求信息，所述需求信息用于指示人工智能AI服务对应的需求；
接收模块402，用于接收所述第一设备发送的对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。
可选地，所述需求信息包括：需求项，或者，需求项及所述需求项对应的需求值。
可选地，在所述需求信息包括需求项及所述需求项对应的需求值的情况下，所述反馈信息包括如下任意一项：
第一指示信息，所述第一指示信息用于指示同意所述需求项，且设备能力满足所述需求值；
第二指示信息，所述第二指示信息用于指示同意所述需求项，且所述第二指示信息携带与所述需求项对应的设备能力或能力差异信息，所述能力差异信息用于指示设备能力与 所述需求值的差异；
第三指示信息，所述第三指示信息用于指示拒绝所述需求项。
可选地，所述需求项的数量为至少两个，所述第二指示信息携带第一需求项及与所述第一需求项对应的设备能力或能力差异信息，所述第一需求项为所述至少两个需求项中与所述设备能力不匹配的需求项。
可选地，在所述需求信息包括需求项的情况下，所述反馈信息包括如下任意一项：
第四指示信息，所述第四指示信息用于指示拒绝所述需求项；
第五指示信息，所述第五指示信息用于指示同意所述需求项，且所述第五指示信息携带与所述需求项对应的设备能力。
可选地，所述需求项包括如下至少一项：
AI算力需求；AI算法需求；AI数据需求。
本申请实施例中的信息反馈装置，能够避免第一设备直接实现该AI服务需求导致影响设备自身的正常运行。
本申请实施例中的信息反馈装置可以是电子设备，例如具有操作系统的电子设备，也可以是电子设备中的部件，例如集成电路或芯片。该电子设备可以是终端，也可以为除终端之外的其他设备。示例性的，终端可以包括但不限于上述所列举的终端11的类型，其他设备可以为服务器、网络附属存储器(Network Attached Storage，NAS)等，本申请实施例不作具体限定。
本申请实施例提供的信息反馈装置能够实现图3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。
可选地，如图6所示，本申请实施例还提供一种通信设备500，包括处理器501和存储器502，存储器502上存储有可在所述处理器501上运行的程序或指令，例如，该通信设备500为第一设备时，该程序或指令被处理器501执行时实现上述应用于第一设备的信息反馈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500为第二设备时，该程序或指令被处理器501执行时实现上述应用于第二设备的信息反馈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终端，所述终端可以为第一设备，包括处理器和通信接口，通信接口用于接收第二设备发送的需求信息，所述需求信息用于指示人工智能AI服务对应的需求；通信接口还用于向所述第二设备发送对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。该终端实施例与上述第一设备侧方法实施例对应，上述方法实施例的各个实施过程和实现方式均可适用于该终端实施例中，且能达到相同的技术效果。具体地，图7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600包括但不限于：射频单元601、网络模块602、音频输出单元603、输入单元604、传感器605、显示单元606、用户输入单元607、接口单元608、存储器609以及处理器610等中的至少部分部件。
本领域技术人员可以理解，终端600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610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7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604可以包括图形处理单元(Graphics Processing Unit，GPU)6041和麦克风6042，图形处理器6041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606可包括显示面板6061，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6061。用户输入单元607包括触控面板6071以及其他输入设备6072中的至少一种。触控面板6 071，也称为触摸屏。触控面板6071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6072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601接收来自网络侧设备的下行数据后，可以传输给处理器610进行处理；另外，射频单元601可以向网络侧设备发送上行数据。通常，射频单元601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609可用于存储软件程序或指令以及各种数据。存储器609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609可以包括易失性存储器或非易失性存储器，或者，存储器609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609包括但不限于这些和任意其它适合类型的存储器。
处理器610可包括一个或多个处理单元；可选的，处理器610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610中。
其中，所述终端可以为第一设备：
射频单元601用于：接收第二设备发送的需求信息，所述需求信息用于指示人工智能 AI服务对应的需求；
射频单元601还用于：向所述第二设备发送对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。
可选地，所述需求信息包括：需求项，或者，需求项及所述需求项对应的需求值。
可选地，在所述需求信息包括需求项及所述需求项对应的需求值的情况下，所述反馈信息包括如下任意一项：
第一指示信息，所述第一指示信息用于指示同意所述需求项，且设备能力满足所述需求值；
第二指示信息，所述第二指示信息用于指示同意所述需求项，且所述第二指示信息携带与所述需求项对应的设备能力或能力差异信息，所述能力差异信息用于指示设备能力与所述需求值的差异；
第三指示信息，所述第三指示信息用于指示拒绝所述需求项。
可选地，所述需求项的数量为至少两个，所述第二指示信息携带第一需求项及与所述第一需求项对应的设备能力或能力差异信息，所述第一需求项为所述至少两个需求项中与所述设备能力不匹配的需求项。
可选地，在所述需求信息包括需求项的情况下，所述反馈信息包括如下任意一项：
第四指示信息，所述第四指示信息用于指示拒绝所述需求项；
第五指示信息，所述第五指示信息用于指示同意所述需求项，且所述第五指示信息携带与所述需求项对应的设备能力。
可选地，所述需求项包括如下至少一项：
AI算力需求；AI算法需求；AI数据需求。
可选地，所述AI算力需求包括如下至少一项：
算力；
存储；
总计算量；
执行时间限制；
耗电量。
可选地，所述AI算法需求包括如下至少一项：
AI任务分类；
AI学习框架；
AI网络开发环境；
AI基础模型库。
可选地，所述AI任务分类包括如下至少一项：
模型训练；
模型推理；
模型验证；
模型监视；
模型部署。
可选地，所述AI学习框架包括如下至少一项：
有监督深度学习；
无监督深度学习；
元学习；
迁移学习；
强化学习；
联邦学习。
可选地，所述AI数据需求包括如下至少一项：
标签数据；
标签维度个数；
标签维度；
标签维度的顺序；
标签维度的数据个数；
标签数据间隔；
AI模型输入数据；
AI模型输入维度个数；
AI模型输入维度；
AI模型输入维度的顺序；
AI模型输入维度的数据个数；
AI模型输入数据收集间隔；
AI模型标签时延。
可选地，所述标签数据用于指示如下至少一项：
波束信道的参考信号接收功率RSRP；
波束信道的参考信号接收质量RSRQ；
波束信道的信号与干扰加噪声比SINR；
小区信道的RSRP；
小区信道的RSRQ；
小区信道的SINR；
小区信道的接收信号强度指示RSSI；
小区信道冲激响应；
预编码矩阵指示PMI；
秩指示RI；
信道质量指示CQI。
可选地，所述标签数据间隔用于指示如下至少一项：
时间间隔；
频率间隔；
时延间隔；
相位间隔；
多普勒间隔；
波束间隔。
可选地，所述AI模型标签时延用于指示如下至少一项：
推理结束与获得标签数据之间的时延；
接收AI模型输入数据与获得标签数据之间的时延。
在该实施方式中，终端接收第二设备发送的需求信息后，向第二设备发送对所述需求信息的反馈信息，能够避免直接实现该AI服务需求导致影响设备自身的正常运行。
本申请实施例还提供一种网络侧设备，包括处理器和通信接口，所述通信接口用于：接收第二设备发送的需求信息，所述需求信息用于指示人工智能AI服务对应的需求，向所述第二设备发送对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求；或者，所述通信接口用于：向第一设备发送需求信息，所述需求信息用于指示人工智能AI服务对应的需求，接收所述第一设备发送的对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。该网络侧设备实施例与上述第一设备侧的方法实施例或者第二设备侧的方法实施例对应，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网络侧设备实施例中，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8所示，该网络侧设备700包括：天线701、射频装置702、基带装置703、处理器704和存储器705。天线701与射频装置702连接。在上行方向上，射频装置702通过天线701接收信息，将接收的信息发送给基带装置703进行处理。在下行方向上，基带装置703对要发送的信息进行处理，并发送给射频装置702，射频装置702对收到的信息进行处理后经过天线701发送出去。
以上实施例中网络侧设备执行的方法可以在基带装置703中实现，该基带装置703包括基带处理器。
基带装置703例如可以包括至少一个基带板，该基带板上设置有多个芯片，如图8所示，其中一个芯片例如为基带处理器，通过总线接口与存储器705连接，以调用存储器705中的程序，执行以上方法实施例中所示的网络设备操作。
该网络侧设备还可以包括网络接口706，该接口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700还包括：存储在存储器705上并可在处理器704上运行的指令或程序，处理器704调用存储器705中的指令或程序执行图4或图5所 示各模块执行的方法，并达到相同的技术效果，为避免重复，故不在此赘述。
本申请实施例还提供一种可读存储介质，所述可读存储介质上存储有程序或指令，该程序或指令被处理器执行时实现上述信息反馈方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
其中，所述处理器为上述实施例中所述的终端中的处理器。所述可读存储介质，包括计算机可读存储介质，如计算机只读存储器ROM、随机存取存储器RAM、磁碟或者光盘等。
本申请实施例另提供了一种芯片，所述芯片包括处理器和通信接口，所述通信接口和所述处理器耦合，所述处理器用于运行程序或指令，实现上述信息反馈方法实施例的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
应理解，本申请实施例提到的芯片还可以称为系统级芯片，系统芯片，芯片系统或片上系统芯片等。
本申请实施例另提供了一种计算机程序/程序产品，所述计算机程序/程序产品被存储在存储介质中，所述计算机程序/程序产品被至少一个处理器执行以实现上述信息反馈方法的各个过程，且能达到相同的技术效果，为避免重复，这里不再赘述。
本申请实施例还提供了一种信息反馈系统，包括：第一设备及第二设备，所述第一设备可用于执行如上所述第一设备侧的信息反馈方法的步骤，所述第二设备可用于执行如上所述第二设备侧的信息接收方法的步骤。
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 (25)

  1. 一种信息反馈方法，包括：
    第一设备接收第二设备发送的需求信息，所述需求信息用于指示人工智能AI服务对应的需求；
    所述第一设备向所述第二设备发送对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。
  2. 根据权利要求1所述的方法，其中，所述需求信息包括：需求项，或者，需求项及所述需求项对应的需求值。
  3. 根据权利要求2所述的方法，其中，在所述需求信息包括需求项及所述需求项对应的需求值的情况下，所述反馈信息包括如下任意一项：
    第一指示信息，所述第一指示信息用于指示同意所述需求项，且设备能力满足所述需求值；
    第二指示信息，所述第二指示信息用于指示同意所述需求项，且所述第二指示信息携带与所述需求项对应的设备能力或能力差异信息，所述能力差异信息用于指示设备能力与所述需求值的差异；
    第三指示信息，所述第三指示信息用于指示拒绝所述需求项。
  4. 根据权利要求3所述的方法，其中，所述需求项的数量为至少两个，所述第二指示信息携带第一需求项及与所述第一需求项对应的设备能力或能力差异信息，所述第一需求项为所述至少两个需求项中与所述设备能力不匹配的需求项。
  5. 根据权利要求2所述的方法，其中，在所述需求信息包括需求项的情况下，所述反馈信息包括如下任意一项：
    第四指示信息，所述第四指示信息用于指示拒绝所述需求项；
    第五指示信息，所述第五指示信息用于指示同意所述需求项，且所述第五指示信息携带与所述需求项对应的设备能力。
  6. 根据权利要求2-5中任一项所述的方法，其中，所述需求项包括如下至少一项：
    AI算力需求；AI算法需求；AI数据需求。
  7. 根据权利要求6所述的方法，其中，所述AI算力需求包括如下至少一项：
    算力；
    存储；
    总计算量；
    执行时间限制；
    耗电量。
  8. 根据权利要求6所述的方法，其中，所述AI算法需求包括如下至少一项：
    AI任务分类；
    AI学习框架；
    AI网络开发环境；
    AI基础模型库。
  9. 根据权利要求8所述的方法，其中，所述AI任务分类包括如下至少一项：
    模型训练；
    模型推理；
    模型验证；
    模型监视；
    模型部署。
  10. 根据权利要求8所述的方法，其中，所述AI学习框架包括如下至少一项：
    有监督深度学习；
    无监督深度学习；
    元学习；
    迁移学习；
    强化学习；
    联邦学习。
  11. 根据权利要求8所述的方法，其中，所述AI数据需求包括如下至少一项：
    标签数据；
    标签维度个数；
    标签维度；
    标签维度的顺序；
    标签维度的数据个数；
    标签数据间隔；
    AI模型输入数据；
    AI模型输入维度个数；
    AI模型输入维度；
    AI模型输入维度的顺序；
    AI模型输入维度的数据个数；
    AI模型输入数据收集间隔；
    AI模型标签时延。
  12. 根据权利要求11所述的方法，其中，所述标签数据用于指示如下至少一项：
    波束信道的参考信号接收功率RSRP；
    波束信道的参考信号接收质量RSRQ；
    波束信道的信号与干扰加噪声比SINR；
    小区信道的RSRP；
    小区信道的RSRQ；
    小区信道的SINR；
    小区信道的接收信号强度指示RSSI；
    小区信道冲激响应；
    预编码矩阵指示PMI；
    秩指示RI；
    信道质量指示CQI。
  13. 根据权利要求11所述的方法，其中，所述标签数据间隔用于指示如下至少一项：
    时间间隔；
    频率间隔；
    时延间隔；
    相位间隔；
    多普勒间隔；
    波束间隔。
  14. 根据权利要求11所述的方法，其中，所述AI模型标签时延用于指示如下至少一项：
    推理结束与获得标签数据之间的时延；
    接收AI模型输入数据与获得标签数据之间的时延。
  15. 一种信息反馈方法，包括：
    第二设备向第一设备发送需求信息，所述需求信息用于指示人工智能AI服务对应的需求；
    所述第二设备接收所述第一设备发送的对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。
  16. 根据权利要求15所述的方法，其中，所述需求信息包括：需求项，或者，需求项及所述需求项对应的需求值。
  17. 根据权利要求16所述的方法，其中，在所述需求信息包括需求项及所述需求项对应的需求值的情况下，所述反馈信息包括如下任意一项：
    第一指示信息，所述第一指示信息用于指示同意所述需求项，且设备能力满足所述需求值；
    第二指示信息，所述第二指示信息用于指示同意所述需求项，且所述第二指示信息携带与所述需求项对应的设备能力或能力差异信息，所述能力差异信息用于指示设备能力与所述需求值的差异；
    第三指示信息，所述第三指示信息用于指示拒绝所述需求项。
  18. 根据权利要求17所述的方法，其中，所述需求项的数量为至少两个，所述第二指示信息携带第一需求项及与所述第一需求项对应的设备能力或能力差异信息，所述第一 需求项为所述至少两个需求项中与所述设备能力不匹配的需求项。
  19. 根据权利要求16所述的方法，其中，在所述需求信息包括需求项的情况下，所述反馈信息包括如下任意一项：
    第四指示信息，所述第四指示信息用于指示拒绝所述需求项；
    第五指示信息，所述第五指示信息用于指示同意所述需求项，且所述第五指示信息携带与所述需求项对应的设备能力。
  20. 根据权利要求16-19中任一项所述的方法，其中，所述需求项包括如下至少一项：
    AI算力需求；AI算法需求；AI数据需求。
  21. 一种信息反馈装置，第一设备包括所述信息反馈装置，包括：
    接收模块，用于接收第二设备发送的需求信息，所述需求信息用于指示人工智能AI服务对应的需求；
    发送模块，用于向所述第二设备发送对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。
  22. 一种信息反馈装置，第二设备包括所述信息反馈装置，包括：
    发送模块，用于向第一设备发送需求信息，所述需求信息用于指示人工智能AI服务对应的需求；
    接收模块，用于接收所述第一设备发送的对所述需求信息的反馈信息，所述反馈信息用于指示是否同意所述需求信息对应的需求。
  23. 一种第一设备，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，其中，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任一项所述的信息反馈方法的步骤。
  24. 一种第二设备，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，其中，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任一项所述的信息反馈方法的步骤。
  25. 一种可读存储介质，所述可读存储介质上存储程序或指令，其中，所述程序或指令被处理器执行时实现如权利要求1至14任一项所述的信息反馈方法的步骤，或者实现如权利要求15至20任一项所述的信息反馈方法的步骤。
PCT/CN2023/085491 2022-04-02 2023-03-31 信息反馈方法、装置及设备 WO2023186099A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202210350469.2 2022-04-02
CN202210350469.2A CN116939713A (zh) 2022-04-02 2022-04-02 信息反馈方法、装置及设备

Publications (1)

Publication Number Publication Date
WO2023186099A1 true WO2023186099A1 (zh) 2023-10-05

Family

ID=88199504

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/085491 WO2023186099A1 (zh) 2022-04-02 2023-03-31 信息反馈方法、装置及设备

Country Status (2)

Country Link
CN (1) CN116939713A (zh)
WO (1) WO2023186099A1 (zh)

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN111768006A (zh) * 2020-06-24 2020-10-13 北京金山云网络技术有限公司 一种人工智能模型的训练方法、装置、设备及存储介质
WO2021134459A1 (en) * 2019-12-31 2021-07-08 Asiainfo Technologies (China) , Inc. Ai intelligentialization based on signaling interaction
CN113419846A (zh) * 2021-02-22 2021-09-21 阿里巴巴集团控股有限公司 资源配置方法和装置、电子设备及计算机可读存储介质
WO2022000188A1 (zh) * 2020-06-29 2022-01-06 北京小米移动软件有限公司 用户设备辅助信息的上报方法及装置、用户设备、存储介质
WO2022011946A1 (zh) * 2020-10-23 2022-01-20 平安科技（深圳）有限公司 一种数据预测方法、装置、计算机设备及存储介质

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2021134459A1 (en) * 2019-12-31 2021-07-08 Asiainfo Technologies (China) , Inc. Ai intelligentialization based on signaling interaction
CN111768006A (zh) * 2020-06-24 2020-10-13 北京金山云网络技术有限公司 一种人工智能模型的训练方法、装置、设备及存储介质
WO2022000188A1 (zh) * 2020-06-29 2022-01-06 北京小米移动软件有限公司 用户设备辅助信息的上报方法及装置、用户设备、存储介质
WO2022011946A1 (zh) * 2020-10-23 2022-01-20 平安科技（深圳）有限公司 一种数据预测方法、装置、计算机设备及存储介质
CN113419846A (zh) * 2021-02-22 2021-09-21 阿里巴巴集团控股有限公司 资源配置方法和装置、电子设备及计算机可读存储介质

Also Published As

Publication number Publication date
CN116939713A (zh) 2023-10-24

Similar Documents

Publication Publication Date Title
WO2021179176A1 (en) Federated learning in telecom communication system
WO2024012235A1 (zh) 人工智能请求分析方法、装置及设备
WO2023186099A1 (zh) 信息反馈方法、装置及设备
WO2023040887A1 (zh) 信息上报方法、装置、终端及可读存储介质
WO2024067281A1 (zh) Ai模型的处理方法、装置及通信设备
WO2023186091A1 (zh) 样本确定方法、装置及设备
WO2024067280A1 (zh) 更新ai模型参数的方法、装置及通信设备
WO2023134653A1 (zh) 通信网络预测方法、终端及网络侧设备
WO2024125525A1 (zh) Ai算力上报方法、终端及网络侧设备
WO2023169402A1 (zh) 模型的准确度确定方法、装置及网络侧设备
WO2023186090A1 (zh) 验证方法、装置及设备
WO2024032695A1 (zh) Csi预测处理方法、装置、通信设备及可读存储介质
WO2023098535A1 (zh) 信息交互方法、装置及通信设备
WO2024067438A1 (zh) Ai模型推理方法、设备及可读存储介质
WO2023098586A1 (zh) 信息交互方法、装置及通信设备
WO2024120447A1 (zh) 模型监督触发方法、装置、ue、网络侧设备、可读存储介质及通信系统
WO2023186014A1 (zh) 信号发送方法、信号接收方法及通信设备
WO2024140422A1 (zh) Ai单元的性能监测方法、终端及网络侧设备
WO2023040885A1 (zh) 参数选择方法、参数配置方法、终端及网络侧设备
WO2024032694A1 (zh) Csi预测处理方法、装置、通信设备及可读存储介质
WO2023151585A1 (zh) 终端目标面能力上报、获取方法、终端及网络设备
WO2023241514A1 (zh) 参考信号接收、发送方法、终端及网络侧设备
WO2024022267A1 (zh) 算力任务迁移方法及通信设备
WO2024008154A1 (zh) 联邦学习方法、装置、通信设备及可读存储介质
WO2024093713A1 (zh) 资源配置方法、装置、通信设备及可读存储介质

Legal Events

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

Ref document number: 23778463

Country of ref document: EP

Kind code of ref document: A1