WO2024061287A1 - Artificial intelligence (ai) model transmission method and apparatus, and terminal and medium - Google Patents
Artificial intelligence (ai) model transmission method and apparatus, and terminal and medium Download PDFInfo
- Publication number
- WO2024061287A1 WO2024061287A1 PCT/CN2023/120138 CN2023120138W WO2024061287A1 WO 2024061287 A1 WO2024061287 A1 WO 2024061287A1 CN 2023120138 W CN2023120138 W CN 2023120138W WO 2024061287 A1 WO2024061287 A1 WO 2024061287A1
- Authority
- WO
- WIPO (PCT)
- Prior art keywords
- model
- target
- transmission mode
- information
- transmission
- Prior art date
Links
- 230000005540 biological transmission Effects 0.000 title claims abstract description 891
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 861
- 238000000034 method Methods 0.000 title claims abstract description 126
- 238000012545 processing Methods 0.000 claims description 25
- 230000001629 suppression Effects 0.000 claims description 12
- 230000011664 signaling Effects 0.000 claims description 10
- 230000008054 signal transmission Effects 0.000 claims description 8
- 238000004891 communication Methods 0.000 abstract description 42
- 230000006870 function Effects 0.000 description 18
- 230000000694 effects Effects 0.000 description 14
- 238000010586 diagram Methods 0.000 description 12
- 230000001360 synchronised effect Effects 0.000 description 10
- 238000003062 neural network model Methods 0.000 description 7
- 238000004590 computer program Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 6
- 210000002569 neuron Anatomy 0.000 description 5
- 238000001514 detection method Methods 0.000 description 4
- 239000011159 matrix material Substances 0.000 description 4
- 230000003068 static effect Effects 0.000 description 4
- 239000002699 waste material Substances 0.000 description 3
- 241000699670 Mus sp. Species 0.000 description 2
- 230000004913 activation Effects 0.000 description 2
- 230000003190 augmentative effect Effects 0.000 description 2
- 238000003066 decision tree Methods 0.000 description 2
- 238000007599 discharging Methods 0.000 description 2
- 239000004973 liquid crystal related substance Substances 0.000 description 2
- 238000011084 recovery Methods 0.000 description 2
- 238000012706 support-vector machine Methods 0.000 description 2
- 208000003028 Stuttering Diseases 0.000 description 1
- 239000000654 additive Substances 0.000 description 1
- 230000000996 additive effect Effects 0.000 description 1
- 238000004422 calculation algorithm Methods 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 238000001914 filtration Methods 0.000 description 1
- 230000007774 longterm Effects 0.000 description 1
- 238000005259 measurement Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000002441 reversible effect Effects 0.000 description 1
- 239000004984 smart glass Substances 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000005406 washing Methods 0.000 description 1
Classifications
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W72/00—Local resource management
- H04W72/04—Wireless resource allocation
- H04W72/044—Wireless resource allocation based on the type of the allocated resource
- H04W72/046—Wireless resource allocation based on the type of the allocated resource the resource being in the space domain, e.g. beams
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
- H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
- H04B7/06—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
- H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
- H04B7/06—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
- H04B7/0613—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission
- H04B7/0615—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal
- H04B7/0619—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal using feedback from receiving side
- H04B7/0621—Feedback content
- H04B7/0626—Channel coefficients, e.g. channel state information [CSI]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
- H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
- H04B7/06—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
- H04B7/0613—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission
- H04B7/0615—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal
- H04B7/0619—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal using feedback from receiving side
- H04B7/0621—Feedback content
- H04B7/0632—Channel quality parameters, e.g. channel quality indicator [CQI]
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04B—TRANSMISSION
- H04B7/00—Radio transmission systems, i.e. using radiation field
- H04B7/02—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas
- H04B7/04—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas
- H04B7/06—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station
- H04B7/0613—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission
- H04B7/0615—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal
- H04B7/0619—Diversity systems; Multi-antenna system, i.e. transmission or reception using multiple antennas using two or more spaced independent antennas at the transmitting station using simultaneous transmission of weighted versions of same signal using feedback from receiving side
- H04B7/0636—Feedback format
- H04B7/0639—Using selective indices, e.g. of a codebook, e.g. pre-distortion matrix index [PMI] or for beam selection
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L41/00—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks
- H04L41/16—Arrangements for maintenance, administration or management of data switching networks, e.g. of packet switching networks using machine learning or artificial intelligence
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L5/00—Arrangements affording multiple use of the transmission path
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
- H04L5/00—Arrangements affording multiple use of the transmission path
- H04L5/003—Arrangements for allocating sub-channels of the transmission path
- H04L5/0048—Allocation of pilot signals, i.e. of signals known to the receiver
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W16/00—Network planning, e.g. coverage or traffic planning tools; Network deployment, e.g. resource partitioning or cells structures
- H04W16/24—Cell structures
- H04W16/28—Cell structures using beam steering
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W72/00—Local resource management
- H04W72/04—Wireless resource allocation
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W72/00—Local resource management
- H04W72/04—Wireless resource allocation
- H04W72/044—Wireless resource allocation based on the type of the allocated resource
-
- H—ELECTRICITY
- H04—ELECTRIC COMMUNICATION TECHNIQUE
- H04W—WIRELESS COMMUNICATION NETWORKS
- H04W72/00—Local resource management
- H04W72/50—Allocation or scheduling criteria for wireless resources
- H04W72/56—Allocation or scheduling criteria for wireless resources based on priority criteria
Definitions
- the present application belongs to the field of communication technology, and specifically relates to an AI model transmission method, device, terminal and medium.
- terminal devices can receive trained artificial intelligence (AI) models from other devices and use the trained AI models to communicate, thereby improving communication throughput and reducing communication costs. time delay.
- AI artificial intelligence
- the terminal device cannot apply the trained AI model received from other devices.
- the terminal device may need to perform multiple transmissions with other devices before it can obtain the trained AI model that the terminal device can apply. model to communicate using the trained AI model.
- Embodiments of the present application provide an AI model transmission method, device, terminal and medium, which can solve the problem of waste of transmission resources.
- an AI model transmission method is provided, which is applied to a first device.
- the method includes: the first device receives target model information of the target AI model from a second device; the first device obtains the target AI based on the target model information. model; the target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first device; the target application time of the target AI model is determined based on the target transmission mode ; the target AI model triggers the first device to send a target feedback event to the second device, and the target feedback event is used to feed back the application status of the target AI model.
- the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
- an AI model transmission device is provided.
- the AI model transmission device is a first AI model transmission device.
- the first AI model transmission device includes: a receiving module. Among them, the receiving module is used to receive the target model information of the target AI model from the second AI model transmission device; the first AI model transmission device obtains the target AI model according to the target model information; the target AI model satisfies at least one of the following: target The target transmission mode of the AI model is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device; the target application time of the target AI model is determined based on the target transmission mode; the target AI model triggers the first AI model The transmission device sends a target feedback event to the second AI model transmission device, where the target feedback event is used to feed back the application status of the target AI model.
- the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
- an AI model transmission method is provided, applied to a second device.
- the method includes: the second device sends target model information of the target AI model to the first device; the target model information is used by the first device to obtain the target AI model; the target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first device; the target application time of the target AI model is determined based on the target transmission mode Determined; the target AI model triggers the first device to send a target feedback event to the second device, and the target feedback event is used to feed back the application status of the target AI model.
- the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
- an AI model transmission device is provided, and the AI model transmission device is a second AI model transmission device,
- the second AI model transmission device includes: a sending module.
- the sending module is used to send the target model information of the target AI model to the first AI model transmission device; the target model information is used by the first device to obtain the target AI model; the target AI model satisfies at least one of the following: target AI model
- the target transmission mode is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device; the target application time of the target AI model is determined based on the target transmission mode; the target AI model triggers the first AI model transmission device
- a target feedback event is sent to the second AI model transmission device, where the target feedback event is used to feed back the application status of the target AI model.
- the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
- a terminal which includes a processor and a memory, wherein the memory stores a program or instruction that can be run on the processor, and when the program or instruction is executed by the processor, the steps of the method described in the first aspect are implemented, or the steps of the method described in the third aspect are implemented.
- a terminal including a processor and a communication interface, wherein the communication interface is used to receive target model information of the target AI model from the second device, and the processor is used to obtain the target AI model based on the target model information. ; Or, the communication structure is used to send target model information of the target AI model to the first device, and the target AI model is used by the first device to obtain the target AI model.
- the target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device; the target application time of the target AI model is determined based on the target The transmission mode is determined; the target AI model triggers the first AI model transmission device to send a target feedback event to the second AI model transmission device, and the target feedback event is used to feed back the application status of the target AI model.
- the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
- a network side device in a seventh aspect, includes a processor and a memory.
- the memory stores programs or instructions that can be run on the processor.
- the program or instructions are executed by the processor.
- a network side device including a processor and a communication interface, wherein the communication interface is used to receive target model information of the target AI model from the second device, and the processor is used to obtain the target based on the target model information.
- AI model or, the communication structure is used to send target model information of the target AI model to the first device, and the target AI model is used by the first device to obtain the target AI model.
- the target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device; the target application time of the target AI model is determined based on the target The transmission mode is determined; the target AI model triggers the first AI model transmission device to send a target feedback event to the second AI model transmission device, and the target feedback event is used to feed back the application status of the target AI model.
- the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
- a ninth aspect provides an AI model transmission system, including: a first terminal and a second terminal.
- the first terminal can be used to perform the steps of the method described in the first aspect
- the second terminal can be used to perform the steps of the method described in the first aspect.
- a readable storage medium on which a program or instruction is stored.
- the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented, or the steps of the method described in the third aspect are implemented.
- 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. The steps of a method, or steps of implementing a method as described in the third 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 first device may receive target model information of the target AI model from the second device, and obtain the target AI model according to the target model information, and the target AI model satisfies at least one of the following:
- the target transmission mode is determined according to the capability of the AI model transmission mode supported by the first device;
- the target application time of the target AI model is determined according to the target transmission mode;
- the target AI model triggers the first device to send a target feedback event to the second device, and the target feedback event is used to feedback the application status of the target AI model;
- the above-mentioned target transmission mode includes any one of the first transmission mode and the second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes the model structure information and model parameter information of the target AI model; when the target transmission mode is the second transmission mode, the target model information includes the model parameter information of the target AI model.
- the target transmission mode of the target AI model is determined according to the capability of the AI model transmission mode supported by the first device, it is possible to avoid the situation where the target AI model cannot be obtained according to the target model information due to the first device not supporting the first transmission mode or the second transmission mode, and thus the target AI model cannot be applied; and/or, since the target application time of the target AI model is determined according to the target transmission mode, it is possible to avoid the situation where the target AI model cannot be applied or the waiting time is wasted due to the application time of the target AI model being too long (or too short); and/or, since the target AI model can trigger the first device to send a target feedback event for feedback on the application of the target AI model to the second device, the second device can directly send the AI model that the first device can apply to the first device according to the target feedback event, without the need for the first device and the second device to perform multiple transmissions. Thereby, the number of transmissions between the first device and the second device in the process of obtaining the target AI model can be reduced, so that transmission resources can be saved.
- Figure 1 is a block diagram of a wireless communication system provided by an embodiment of the present application.
- Figure 2 is one of the flow diagrams of the AI model transmission method provided by the embodiment of the present application.
- Figure 3 is the second schematic flow chart of the AI model transmission method provided by the embodiment of the present application.
- Figure 4 is the third schematic flowchart of the AI model transmission method provided by the embodiment of the present application.
- Figure 5 is the fourth schematic flowchart of the AI model transmission method provided by the embodiment of the present application.
- Figure 6 is the fifth schematic flow chart of the AI model transmission method provided by the embodiment of the present application.
- Figure 7 is a schematic structural diagram of a first AI model transmission device provided by an embodiment of the present application.
- Figure 8 is a schematic structural diagram of a second AI model transmission device provided by an embodiment of the present application.
- Figure 9 is a schematic structural diagram of a communication device provided by an embodiment of the present application.
- Figure 10 is a schematic diagram of the hardware structure of a terminal provided by an embodiment of the present application.
- Figure 11 is a schematic diagram of the hardware structure of a network-side device provided by an embodiment of the present application.
- AI models have been widely used in various fields.
- the AI model can be trained using training samples to obtain a trained AI model, so that the trained AI model can be used for communication to improve communication throughput and reduce communication delay.
- the AI model may include at least one of the following: neural network model, decision tree model, support vector machine model, Bayesian classifier model, etc.
- 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 can 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 handheld 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
- Mobile Internet Device MID
- augmented reality augmented reality, AR
- VR virtual reality
- robots wearable devices
- VUE vehicle-mounted equipment
- PUE pedestrian terminal
- smart home home equipment with wireless communication functions, such as refrigerators, TVs, washing machines or furniture, etc.
- game consoles personal computers (personal computers, PC), teller machines or self-service Terminal devices
- wearable devices include: smart watches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart bracelets, smart rings, smart necklaces, smart anklets, smart anklets, etc.), Smart wristbands, smart clothing, etc.
- the network side equipment 12 may include access network equipment or core network equipment, where the access network equipment 12 may also be called wireless access network equipment, radio access network (Radio Access Network, RAN), radio access network function or Wireless access network unit.
- the access network device 12 may include a base station, a WLAN access point or a WiFi node, etc.
- the base station may be called a Node B, an evolved Node B (eNB), an access point, a Base Transceiver Station (BTS), a radio Base station, radio transceiver, Basic Service Set (BSS), Extended Service Set (ESS), Home B-Node, Home Evolved B-Node, Transmitting Receiving Point (TRP) or all
- eNB evolved Node B
- BTS Base Transceiver Station
- BSS Basic Service Set
- ESS Extended Service Set
- Home B-Node Home Evolved B-Node
- TRP Transmitting Receiving Point
- FIG. 2 shows a flow chart of an AI model transmission method provided by an embodiment of the present application.
- the AI model transmission method provided by the embodiment of the present application may include the following steps 101 and 102.
- Step 101 The first device receives target model information of the target AI model from the second device.
- the above-mentioned first device may be any of the following: user equipment (User Equipment, UE) or network side equipment.
- the above-mentioned second device may be any of the following: UE or network side device.
- the first device may be a UE
- the second device may be a network side device.
- the first device may be a network side device
- the second device may be a UE
- both the first device and the second device are UEs.
- both the first device and the second device are network-side devices.
- the first device and the second device may be different nodes on the network side; or, the first device and the second device may be different terminal nodes on the network side.
- the target AI model may be any of the following: a neural network model, a decision tree model, a support vector machine model, a Bayesian classifier model, etc.
- the target model information may include at least one of the following: model structure information of the target AI model, and model parameter information of the target AI model.
- the above model structure information of the target AI model is used to indicate the model structure of the target AI model.
- the model parameter information of the above target AI model may be the weight value of the AI model.
- the model parameter information of the target AI model may be the weight values of at least some neurons of the neural network model.
- the weight value of the neuron can include any of the following: weight coefficient, multiplicative coefficient, bias coefficient, additive coefficient, type of activation function, and coefficient of the activation function.
- the above target AI model satisfies at least one of the following:
- the target transmission mode of the target AI model is determined based on the capabilities of the AI model transmission mode supported by the first device;
- the target application time of the target AI model is determined based on the target transmission mode
- the target AI model triggers the first device to send a target feedback event to the second device, and the target feedback event is used to feed back the application status of the target AI model.
- the above-mentioned target transmission mode includes any one of the following: a first transmission mode, a second transmission mode;
- the target transmission mode is the first transmission mode
- the target model information includes model structure information and model parameter information of the target AI model
- the target transmission mode is the second transmission mode
- the target model information includes the model structure information and model parameter information of the target AI model. Model parameter information.
- the first transmission mode is a transmission mode that transmits the model structure information and model parameter information of the AI model; the second transmission mode is a transmission mode that transmits the model parameter information of the AI model.
- the first device when the target transmission mode is the first transmission mode, the first device will receive the model structure information and model parameter information of the target AI model from the second device.
- the target transmission mode is the second transmission mode, the first device receives model parameter information of the target AI model from the second device.
- the capabilities of the AI model transmission mode supported by the first device include any of the following:
- the target transmission mode may be determined by the first device according to the capability of the AI model transmission mode supported by the first device; or, it may be determined by the second device according to the AI model transmission mode supported by the first device. The ability is determined.
- the above-mentioned target application time is the switching time or effectiveness time of the target AI model.
- switching time of the target AI model can be understood as: the time required for the first device to deactivate the AI model applied before receiving the target AI model.
- effective time of the target AI model can be understood as: the time required for the first device to deactivate the AI model applied before receiving the target AI model and activate the target AI model.
- the target application time may be determined by the first device based on the target transmission mode; or may be determined by the second device based on the target transmission mode.
- the application of the target AI model may include at least one of the following: whether the first device applies the target AI model, the use of the AI model applied by the first device before receiving the target AI model, Model identification information of the target AI model and the reason why the first device did not apply the target AI model.
- Step 102 The first device obtains the target AI model based on the target model information.
- the first device when the target transmission mode is the first transmission mode, the first device will receive the model structure information and model parameter information of the target AI model from the second device, so that the first device can first Compile (or recompile) the model structure information of the target AI model to determine the model structure of the target AI model, and then apply the model parameter information of the target AI model to the model structure of the target AI model to obtain the target AI model.
- the first device When the target transmission mode is the second transmission mode, the first device will receive the model parameter information of the target AI model from the second device, so that the first device can apply the model parameter information of the target AI model to the pre-stored data in the first device. in the model structure to obtain the target AI model.
- the first device does not need to compile (or recompile) the model structure information of the target AI model first.
- the first device can first determine whether the target AI model can be applied, and perform different operations based on different determination results. Examples will be described below.
- the AI model transmission method provided in the embodiment of the present application may also include the following step 201.
- Step 201 The first device adopts the target AI model and performs the first operation.
- the first device when the first device determines that the target AI model can be applied, the first device can use the target AI model to perform the first operation.
- the above-mentioned first operation includes at least one of the following:
- the above-mentioned signal processing operation includes at least one of the following: signal detection operation, signal filtering operation, and signal equalization operation.
- the signal can be any of the following: Demodulation Reference Signal (DMRS), Sounding Reference Signal (SRS), Synchronization Signal Block (SSB), Tracking Reference Signal (Tracking Reference Signal (TRS), Phase Tracking Reference Signal (PTRS), Channel State nformation Reference Signal (CSI-RS), etc.
- DMRS Demodulation Reference Signal
- SRS Sounding Reference Signal
- SSB Synchronization Signal Block
- TRS Tracking Reference Signal
- TRS Phase Tracking Reference Signal
- CSI-RS Channel State nformation Reference Signal
- the signal transmission operation includes at least one of the following: a channel receiving operation and a channel sending operation.
- the channel can be any one of the following: a physical downlink control channel (PDCCH), a physical downlink shared channel (PDSCH), a physical uplink control channel (PUCCH), a physical uplink shared channel (PUSCH), a physical random access channel (PRACH), and a physical broadcast channel (PBCH).
- PDCCH physical downlink control channel
- PDSCH physical downlink shared channel
- PUCCH physical uplink control channel
- PUSCH physical uplink shared channel
- PRACH physical random access channel
- PBCH physical broadcast channel
- the channel state information acquisition operation includes at least one of the following: channel state information feedback, frequency division duplex (Frequency Division Duplex, FDD) uplink and downlink partial reciprocity.
- the channel state information may include any of the following: channel related information, channel matrix related information, channel characteristic information, channel matrix characteristic information, pre-coding matrix indicator (Pre-coding Matrix Indicator, PMI), rank indicator (Rank Indication) , RI), CSI-RS Resource Indicator (CRI), Channel Quality Information (Channel Quality Information, CQI), Layer Indicator (LI).
- the second device can obtain the angle and delay information according to the uplink channel, and notify the first device of the angle information and delay information through CSI-RS precoding or direct indication, so that the first device can report according to the indication of the second device or select and report within the indication range of the base station, thereby reducing the calculation amount of the first device and the overhead of CSI reporting.
- the beam management operation may include at least one of the following: beam measurement operation, beam reporting operation, beam prediction operation, beam failure detection operation, beam failure recovery operation, and new beam indication operation in beam failure recovery.
- the channel prediction operation may include at least one of the following: a prediction operation of channel state information and a beam prediction operation.
- the interference suppression operation may include at least one of the following: intra-cell interference suppression operation, inter-cell interference suppression operation, out-of-band interference suppression operation, and cross-modulation interference suppression operation.
- the positioning operation may specifically include estimating the target position information through a reference signal (such as SRS).
- the target location information may include at least one of the following: specific location information of the first device (including at least one of horizontal location information and vertical location information), possible future location information, information to assist location estimation, and trajectory estimation information. .
- the prediction and management operations of high-level services and the prediction and management operations of high-level parameters may include: prediction and management operations of throughput, prediction and management operations of required data packet size, prediction and management operations of business needs, and movement speed. Prediction and management operations, prediction and management operations of noise information, etc.
- the control signaling may include at least one of the following: power control related signaling, and beam management related signaling.
- the first device can adopt the target AI model to perform signal processing operations, signal transmission operations, channel state information acquisition operations, beam management operations, channel prediction operations, interference suppression operations, positioning operations, and prediction and management operations of high-level services , at least one of a high-level parameter prediction and management operation and a control signaling parsing operation. Therefore, the communication throughput of the first device can be improved and the communication delay of the first device can be reduced.
- the AI model transmission method provided by the embodiment of the present application may also include the following step 301.
- Step 301 If the target AI model does not match the capabilities of the first device, the first device does not apply the target AI model.
- the first device determines that the target AI model cannot be applied, so that the first device does not apply the target AI model.
- the target AI model does not match the capabilities of the first device, the first device may not respond Using the target AI model, therefore, it is possible to avoid the first device being stuck in the process of using the target AI model for communication.
- the first device can receive the target model information of the target AI model from the second device, and obtain the target AI model based on the target model information.
- the target AI model satisfies at least one of the following: Target The target transmission mode of the AI model is determined based on the capability of the AI model transmission mode supported by the first device; the target application time of the target AI model is determined based on the target transmission mode; the target AI model triggers the first device to transmit data to the second device Send a target feedback event, which is used to feedback the application status of the target AI model; wherein the above-mentioned target transmission mode includes any one of the first transmission mode and the second transmission mode; when the target transmission mode is the first transmission mode
- the target model information includes model structure information and model parameter information of the target AI model; when the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
- the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first device, it can be avoided that the first device does not support the first transmission mode or the second transmission mode, causing the failure to perform the target transmission according to the target AI model.
- the model information obtains the target AI model, which leads to the inability to apply the target AI model; and/or, because the target application time of the target AI model is determined based on the target transmission mode, therefore, the application of the target AI model can be avoided.
- the second device can directly send the AI model that the first device can apply to the first device based on the target feedback event, without the need for the first device to perform multiple transmissions with the second device. This can reduce the number of transmissions between the first device and the second device in the process of obtaining the target AI model, thus saving transmission resources.
- the AI model transmission method provided by the embodiment of the present application may also include the following step 401.
- Step 401 The first device sends target information to the second device.
- the above target information is used to indicate the capability of the AI model transmission mode supported by the first device.
- the second device can determine the target transmission mode according to the capability of the AI model transmission mode supported by the first device.
- the target transmission mode when the AI model transmission mode capability supported by the first device is to support the first transmission mode, the target transmission mode may specifically be the first transmission mode; when the AI model supported by the first device When the capability of the model transmission mode is to support the second transmission mode, the target transmission mode may specifically be the second transmission mode; when the capability of the AI model transmission mode supported by the first device is to support the first transmission mode and the second transmission mode, The target transmission mode may specifically be the first transmission mode or the second transmission mode.
- the first device can send the target information to the second device, so that the second device can determine the target transmission mode according to the AI model transmission mode capability supported by the first device, therefore, it is possible to avoid the problem that the first device does not support the transmission mode.
- the first transmission mode or the second transmission mode results in the inability to obtain the target AI model, which in turn results in the inability to apply the target AI model.
- the target application time of the target AI model that satisfies the target AI model is determined based on the target transmission mode.
- the AI model transmission method provided by the embodiment of the present application may further include the following step 402.
- Step 402 The first device determines a target application time according to a target transmission mode.
- step 402 can be specifically implemented by the following step 402a.
- Step 402a The first device determines the application time corresponding to the target transmission mode as the target application time.
- each corresponding relationship is a corresponding relationship between a transmission mode and an application time, so that the first device can select from multiple transmission modes, Determine a transmission mode that is the same as the target transmission mode, and determine the application time corresponding to the transmission mode as the target application time.
- At least one of the following is satisfied between the first application time and the second application time:
- the first application time is greater than the second application time
- the first application time is greater than or equal to the third application time, and the third application time is determined based on the second application time and the compilation time of the AI model.
- the above-mentioned first application time is: the AI model application time corresponding to the first transmission mode;
- the above-mentioned second application time Time is: the AI model application time corresponding to the second transmission mode.
- the third application time may be the sum of the second application time and the compilation time of the AI model.
- the compilation time of the AI model includes: a first compilation time and a second compilation time; where the first compilation time is: the time from the AI model execution module of the first device to the AI model compilation module of the first device, and the second compilation time is: The compilation time is: the time from the AI model execution module of the second device to the AI model compilation module of the second device.
- the first device can determine the application time corresponding to the target transmission mode as the target application time; wherein the first application time is greater than the second application time, and/or the first application time is greater than or equal to the third application time, therefore , which can avoid situations where the target AI model cannot be applied or waiting time is wasted due to the application time of the target AI model being too large (or too small).
- step 402 can be specifically implemented through the following step 402b.
- Step 402b When the target transmission mode is the first transmission mode, if the target AI model matches the pre-stored AI model in the first device, the first device determines the fourth application time as the target application time.
- the above-mentioned fourth application time is less than the first application time; the first application time is: the AI model application time corresponding to the first transmission mode.
- target AI model and pre-stored AI model match can be understood as: the target AI model and the pre-stored AI model are the same, or the similarity of the model structures of the target AI model and the pre-stored AI model is greater than or equal to the preset Threshold.
- the target AI model and the pre-stored AI model are the same: the target AI model and the pre-stored AI model have the same number of model layers, and the types of neurons in each layer of the target AI model and the pre-stored AI model are the same.
- the similarity between the model structures of the target AI model and the pre-stored AI model is greater than or equal to a preset threshold, which can be any of the following: the model structure of the pre-stored AI model is a subset or sub-model of the model structure of the target AI model, the model structure of the pre-stored AI model includes the model structure of the target AI model, and the model structure of the pre-stored AI model includes a subset or sub-model of the model structure of the target AI model.
- the first device can determine a smaller application time as the target application time when the target AI model matches the pre-stored AI model, the delay in applying the target AI model can be reduced.
- the above target AI model satisfies the target AI model and triggers the first device to send a target feedback event to the second device.
- the above-mentioned target feedback event is any one of the following: a first feedback event, a second feedback event; the first feedback event is used to feedback the unapplied target AI model; the second feedback event is used to feedback the applied target AI model.
- the above-mentioned first feedback event is any of the following:
- the first device does not apply the target AI model
- the first device does not apply the target AI model, and the first device still applies the first AI model
- the first device does not apply the target AI model, and the first device does not apply the AI model
- the first device does not support the target model information sent by the second device through the target transmission mode
- the first device does not support the first model information of the target AI model.
- the fact that the first device does not apply the target AI model can be understood as: the target AI model fails to be activated, and/or fails to take effect, and/or fails to switch.
- the first AI model is: the AI model applied before the first device receives the target AI model. Understandably, the first device still uses the old A model.
- the first device not applying the AI model can be understood as: the first device uses a non-AI algorithm to perform the operation (eg, the first operation).
- the first device does not support the target model information sent by the second device through the target transmission mode, which can be specifically: the first device does not support the target model information sent by the second device through the first transmission mode. It can be understood that the first device does not support the model structure indicated by the model structure information of the target AI model sent by the second device, and the target transmission mode is the first transmission mode.
- the third feedback event is: the first device does not support the target model information sent by the second device through the target transmission mode; the third feedback event carries the first information, and the first information is the information of the model structure in the target AI model that the first device does not support. It can be understood that the first device can feedback to the second device the information of the model structure in the target AI model that the first device does not support, and the target transmission mode is the first transmission mode.
- the model structure information of the neural network model indicates The model structure is 5 fully connected layers, 2 convolutional layers and 1 Long-Short-Term Memory (LSTM) layer.
- the first device does not support the LSTM layer, so the third feedback event is the first device
- the second device is not supported to send the model structure information of the neural network model through the first transmission mode.
- the third feedback event carries the first information, and the first information is the information of the LSTM layer.
- the above-mentioned first model information includes at least one of the following: model size, model complexity, and model operands (such as floating-point operations (Floating-point operations, FLOP)).
- model size such as floating-point operations (Floating-point operations, FLOP)
- model operands such as floating-point operations (Floating-point operations, FLOP)
- the above-mentioned second feedback event is any of the following:
- the first device has the target AI model applied
- the first device has applied the target AI model and has replaced the first AI model
- the first device supports the target model information sent by the second device through the target transmission mode
- the first device has compiled the target AI model.
- the first device has applied the target AI model can be understood as: the target AI model has been activated, and/or has taken effect, and/or has been switched.
- the above-mentioned fourth feedback event is: the first device has applied the target AI model; the fourth feedback event carries second information, and the second information is model identification information of the target AI model.
- the target transmission mode is the first transmission mode.
- the model identification information may specifically be: model identity (Identity, ID).
- the above-mentioned first AI model is: the AI model applied before the first device receives the target AI model.
- the first device supporting the target model information sent by the second device through the target transmission mode can be understood as: the first device supports the model structure indicated by the model structure information of the target AI model sent by the second device through the first transmission mode.
- the first device has compiled the target AI model can be understood as: the first device has compiled the model structure information of the target AI model sent by the second device through the first transmission mode. Therefore, when the first device requires other AI models, the second device can transmit the model parameter information of the other AI models through the second transmission mode, so that the first device can apply the model parameter information of the other AI models to the target AI.
- the model structure information of the model indicates the model structure to obtain the other AI model.
- the first device may consider that the model structure information of the AI model to be transmitted by the second device is the same as the model structure information of the target AI model.
- the first device can also transmit model information of the AI model to the second device, so that the second device can transmit target model information to the first device according to the AI model information transmitted by the first device.
- model information of the AI model can also transmit target model information to the first device according to the AI model information transmitted by the first device.
- the AI model transmission method provided by the embodiment of the present application may also include the following step 501.
- Step 501 The first device sends the second model information of the second AI model to the second device according to the third transmission mode.
- the above-mentioned third transmission mode is the first transmission mode or the fourth transmission mode; when the target transmission mode is the first transmission mode, the second model information includes the model structure information and the model of the second AI model. Parameter information; when the target transmission mode is the fourth transmission mode, the second model information includes model structure information of the second AI model; the above target AI model matches the model structure information of the second AI model.
- the fourth transmission mode is: a transmission mode for transmitting model structure information of the AI model.
- the target AI model matches the model structure information of the second AI model can be understood as: the target AI model and the second AI model have the same model structure information, or the target AI model and the second AI model The similarity of the model structure information is greater than or equal to the preset threshold.
- the model structure information of the target AI model and the second AI model is the same: the target AI model and the second AI model have the same number of model layers, and the types of neurons in each layer of the target AI model and the second AI model are the same.
- the similarity between the model structure information of the target AI model and the second AI model is greater than or equal to the preset threshold, which can be any of the following:
- the model structure indicated by the model structure information of the second AI model is a subset or sub-model of the model structure indicated by the model structure information of the target AI model;
- the model structure indicated by the model structure information of the second AI model includes the model structure indicated by the model structure information of the target AI model
- the model structure indicated by the model structure information of the second AI model includes a subset or sub-model of the model structure indicated by the model structure information of the target AI model.
- the first device since the model structure information of the target AI model and the second AI model match, therefore, the first device
- the model parameter information of the target AI model can be directly received from the second device to obtain the target AI model without repeatedly receiving the model structure information of the target AI model. In this way, transmission resources can be saved.
- FIG 6 shows a flow chart of an AI model transmission method provided by an embodiment of the present application.
- the AI model transmission method provided by the embodiment of the present application may include the following step 601.
- Step 601 The second device sends target model information of the target AI model to the first device.
- the above-mentioned target model information is used by the first device to obtain the target AI model.
- the above target AI model satisfies at least one of the following:
- the target transmission mode of the target AI model is determined based on the capabilities of the AI model transmission mode supported by the first device;
- the target application time of the target AI model is determined based on the target transmission mode
- the target AI model triggers the first device to send a target feedback event to the second device, and the target feedback event is used to feed back the application status of the target AI model.
- the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
- the first transmission mode is: a transmission mode for transmitting model structure information and model parameter information of the AI model
- the second transmission mode is: a transmission mode for transmitting model parameter information of the AI model.
- the second device can send target model information of the target AI model to the first device, and the target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined according to the capability of the AI model transmission mode supported by the first device; the target application time of the target AI model is determined according to the target transmission mode; the target AI model triggers the first device to send a target feedback event to the second device, and the target feedback event is used to feedback the application status of the target AI model; wherein the above-mentioned target transmission mode includes any one of the first transmission mode and the second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model; when the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
- the target transmission mode of the target AI model is determined according to the capability of the AI model transmission mode supported by the first device, it is possible to avoid the situation where the target AI model cannot be obtained according to the target model information due to the first device not supporting the first transmission mode or the second transmission mode, and thus the target AI model cannot be applied; and/or, since the target application time of the target AI model is determined according to the target transmission mode, it is possible to avoid the situation where the target AI model cannot be applied or the waiting time is wasted due to the application time of the target AI model being too long (or too short); and/or, since the target AI model can trigger the first device to send a target feedback event for feedback on the application of the target AI model to the second device, the second device can directly send the AI model that the first device can apply to the first device according to the target feedback event, without the need for the first device and the second device to perform multiple transmissions. Thereby, the number of transmissions between the first device and the second device in the process of obtaining the target AI model can be reduced, so that transmission resources can be saved.
- the AI model transmission method provided by the embodiment of the present application may also include the following steps 701 to 703.
- Step 701 The second device receives target information from the first device.
- the above target information is used to indicate the capability of the AI model transmission mode supported by the first device.
- Step 702 The second device determines the target transmission mode according to the target information.
- the second device when the capability of the AI model transmission mode supported by the first device is to support the first transmission mode, the second device may determine the first transmission mode as the target transmission mode; when the capability of the AI model transmission mode supported by the first device is to support the second transmission mode, the second device may determine the second transmission mode as the target transmission mode; when the capability of the AI model transmission mode supported by the first device is to support the first transmission mode and the second transmission mode, the second device may determine the first transmission mode or the second transmission mode as the target transmission mode.
- Step 703 The second device sends target model information to the first device based on the target transmission mode.
- the second device when the target transmission mode is the first transmission mode, can send the model structure information and model parameter information of the target AI model to the first device, so that the first device can Obtain the target AI model; when the target transmission mode is the second transmission mode, the second device can send model parameter information of the target AI model to the first device, so that the first device can obtain the target AI model.
- the transmission mode is to send the target model information to the first device according to the target transmission mode. Therefore, it is possible to avoid the situation where the first device cannot obtain the target AI model because the first device does not support the first transmission mode or the second transmission mode. This leads to a situation where the target AI model cannot be applied.
- the target transmission mode is the first transmission mode.
- the AI model transmission method provided in the embodiment of the present application may further include the following steps 801 and 802.
- Step 801 The second device determines whether the target AI model matches the pre-stored AI model in the first device.
- the second device can receive the model identification information of the pre-stored AI model in the first device from the first device, and determine the pre-stored AI model based on the model identification information, so that the second device can determine Whether the target AI model matches the pre-stored AI model.
- Step 802 If the target AI model matches the pre-stored AI model, the second device sends the target model information to the first device according to the fourth transmission mode.
- the above-mentioned fourth transmission mode is any one of the following: the second transmission mode, the fifth transmission mode; when the fourth transmission mode is the second transmission mode, the target model information includes the model of the target AI model Parameter information; when the fourth transmission mode is the fifth transmission mode, the target model information includes model structure information and model parameter information of the target AI model; the AI model application time corresponding to the fifth transmission mode is shorter than that corresponding to the first transmission mode AI model application time
- the second transmission mode is a transmission mode for transmitting model parameter information of the AI model
- the fifth transmission mode is a transmission mode for transmitting model structure information and model parameter information of the AI model.
- the fifth transmission mode can be understood as a simplified first transmission mode.
- the second device can send the target model information to the first device according to the second transmission mode, that is, send the target AI model to the first device.
- model parameter information of the target AI model without repeatedly sending the model structure information of the target AI model, thus saving transmission resources; or, the second device can send the target model information to the first device according to the fifth transmission mode, so that the first device can
- the smaller application time is determined as the target application time, so the delay in applying the target AI model can be reduced.
- the AI model transmission method provided by the embodiment of the present application may further include the following step 803.
- Step 803 The second device receives the target feedback event from the first device.
- the above-mentioned target feedback event is used to feedback the application status of the target AI model.
- the second device can receive a target feedback event for feedback on the application of the target AI model from the first device, the second device can directly send the first device based on the target feedback event to the first device. Model information of the AI model without the need for multiple transmissions between the first device and the second device.
- the AI model transmission method provided by the embodiment of the present application may also include the following steps 901 and 902.
- Step 901 The second device receives second model information of the second AI model from the first device.
- the above-mentioned second model information is sent by the first device according to the third transmission mode.
- the third transmission mode is the first transmission mode or the fourth transmission mode; when the target transmission mode is the first transmission mode, the second model information includes the model structure information of the second AI model and Model parameter information; when the target transmission mode is the fourth transmission mode, the second model information includes model structure information of the second AI model.
- the fourth transmission mode is: a transmission mode for transmitting model structure information of the AI model.
- Step 902 The second device determines the model structure information that matches the model structure information of the second AI model as the model structure information of the target AI model.
- step 902 can be implemented through the following step 902a or step 902b.
- Step 902a The second device determines the model structure that is the same as the model structure information of the second AI model as the model structure information of the target AI model.
- the structural information of the target AI model and the second AI model is the same: the target AI model and the second AI model have the same number of model layers, and the types of neurons in each layer of the target AI model and the second AI model are the same.
- Step 902b The second device determines the model structure whose similarity to the model structure information of the second AI model is greater than or equal to the preset threshold as the model structure information of the target AI model.
- the similarity between the model structure information of the target AI model and the second AI model is greater than or equal to the preset threshold, which can be any of the following:
- the model structure indicated by the model structure information of the second AI model is a subset or sub-model of the model structure indicated by the model structure information of the target AI model;
- the model structure indicated by the model structure information of the second AI model includes the model structure indicated by the model structure information of the target AI model
- the model structure of the pre-stored AI model includes a subset or sub-model of the model structure indicated by the model structure information of the target AI model.
- the model structure indicated by the model structure information of the second AI model is 5 fully connected layers and 2 convolutional layers
- the model structure indicated by the model structure information of the target AI model is 4 weighted connection layers and 2 layers.
- the convolutional layer that is, the model structure indicated by the model structure information of the second AI model is a subset of the model structure indicated by the model structure information of the target AI model, then it can be considered that the model structure information of the target AI model and the second AI model are similar.
- the degree is greater than or equal to the preset threshold, that is, the target AI model and the second AI model match.
- the model structure indicated by the model structure information of the second AI model is 5 fully connected layers and 2 convolutional layers
- the model structure indicated by the model structure information of the target AI model is 4 weighted connection layers and, that is, the model structure indicated by the model structure information of the second AI model is a sub-model of the model structure indicated by the model structure information of the target AI model
- the similarity between the model structure information of the target AI model and the second AI model is greater than or equal to the preset threshold, that is, the target AI model and the second AI model match.
- the model structure indicated by the model structure information of the second AI model is 6 fully connected layers and 3 convolutional layers
- the model structure indicated by the model structure information of the target AI model is 5 connected layers and 2 layers.
- the convolution layer that is, the model structure indicated by the model structure information of the second AI model includes the model structure indicated by the model structure information of the target AI model
- the similarity between the model structure information of the target AI model and the second AI model is greater than Or equal to the preset threshold, that is, the target AI model matches the second AI model.
- model structure indicated by the model structure information of the second AI model is 5 layers of fully connected layers, 2 layers of convolutional layers, and 1 layer of LSTM
- model structure indicated by the model structure information of the target AI model is 5 layers.
- the fully connected layer and the 2-layer convolution layer, that is, the model structure indicated by the model structure information of the second AI model includes the model structure indicated by the model structure information of the target AI model, then it can be considered that the model of the target AI model and the second AI model
- the similarity of the structural information is greater than or equal to the preset threshold, that is, the target AI model and the second AI model match.
- the first device can directly receive the model parameter information of the target AI model from the second device to obtain the target AI model without repeatedly receiving the target.
- the model structure information of the AI model can save transmission resources.
- the execution subject may be an AI model transmission device.
- the AI model transmission method performed by the AI model transmission device is used as an example to illustrate the AI model transmission device provided by the embodiment of the present application.
- FIG. 7 shows a possible structural schematic diagram of the AI model transmission device involved in the embodiment of the present application.
- the AI model transmission device is the first AI model transmission device.
- the first AI model transmission device 60 may include: a receiving module 61 and a processing module 62 .
- the receiving module 61 is used to receive the target model information of the target AI model from the second AI model transmission device.
- the processing module 62 is used to obtain the target AI model according to the target model information received by the receiving module 61 .
- the target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device 60; the target application time of the target AI model is determined based on the target transmission The mode is determined; the target AI model triggers the first AI model transmission device 60 to send a target feedback event to the second AI model transmission device, and the target feedback event is used to feed back the application status of the target AI model.
- the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
- the target AI model satisfying the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device 60; the first AI model transmission device 60 Also included: Send module.
- the sending module is configured to send target information to the second AI model transmission device, where the target information is used to indicate the capabilities of the AI model transmission mode supported by the first AI model transmission device 60 .
- the AI model transmission mode supported by the first AI model transmission device 60 can The power includes any of the following: supporting the first transmission mode; supporting the second transmission mode; supporting the first transmission mode and the second transmission mode.
- the target application time of the above target AI model that satisfies the target AI model is determined based on the target transmission mode.
- the first AI model transmission device 60 provided by the embodiment of the present application may also include: a determination module. Among them, the determination module is used to determine the target application time according to the target transmission mode.
- the determination module is specifically used to determine the application time corresponding to the target transmission mode as the target application time; the first application time and the second application time satisfy at least one of the following: the first application time is greater than the second application time; the first application time is greater than or equal to the third application time, and the third application time is determined based on the second application time and the compilation time of the AI model.
- the first application time is: the AI model application time corresponding to the first transmission mode; the second application time is: the AI model application time corresponding to the second transmission mode.
- the above-mentioned determination module is specifically used to: when the target transmission mode is the first transmission mode, if the target AI model matches the pre-stored AI model in the first AI model transmission device 60, then The fourth application time is determined as the target application time.
- the above-mentioned fourth application time is less than the first application time; the first application time is: the AI model application time corresponding to the first transmission mode.
- the above target AI model satisfies the target AI model and triggers the first AI model transmission device 60 to send a target feedback event to the second AI model transmission device;
- the target feedback event is any of the following: first feedback event, second feedback event.
- the above-mentioned first feedback event is used to feedback that the target AI model has not been applied;
- the above-mentioned second feedback event is used to feedback that the target AI model has been applied.
- the above-mentioned first feedback event is any one of the following: the first AI model transmission device 60 does not apply the target AI model; the first AI model transmission device 60 does not apply the target AI model, and the first AI
- the model transmission device 60 still applies the first AI model; the first AI model transmission device 60 does not apply the target AI model, and the first AI model transmission device 60 does not apply the AI model; the first AI model transmission device 60 does not support the second AI model
- the above-mentioned first AI model is: the AI model applied before the first AI model transmission device 60 receives the target AI model;
- the above-mentioned third feedback event is: the first AI model transmission device 60 does not support the second AI model transmission device to pass the target Target model information sent in transmission mode;
- the above-mentioned third feedback event carries first information, which is information about a model structure that is not supported by the first AI model transmission device 60 in the target AI model;
- the above-mentioned first model information includes the following At least one item: model size, model complexity, and number of model operations.
- the above-mentioned second feedback event is any one of the following: the first AI model transmission device 60 has applied the target AI model; the fourth feedback event; the first AI model transmission device 60 has applied the target AI model , and the first AI model has been replaced; the first AI model transmission device 60 supports the target model information sent by the second AI model transmission device through the target transmission mode; the first AI model transmission device 60 has compiled the target AI model.
- the above-mentioned fourth feedback event is: the first AI model transmission device 60 has applied the target AI model; the above-mentioned fourth feedback event carries second information, and the second information is the model identification information of the target AI model; the above-mentioned first AI model is: the AI model applied before the first AI model transmission device 60 receives the target AI model.
- the first AI model transmission device 60 provided in the embodiment of the present application may also include: a processing module.
- the processing module is configured to not apply the target AI model when the target AI model does not match the capabilities of the first AI model transmission device 60 .
- the first AI model transmission device 60 may also include: a sending module.
- the sending module is configured to send the second model information of the second AI model to the second AI model transmission device according to the third transmission mode.
- the above-mentioned third transmission mode is the first transmission mode or the fourth transmission mode; when the target transmission mode is the first transmission mode, the second model information includes model structure information and model parameter information of the second AI model; in When the target transmission mode is the fourth transmission mode, the second model information includes model structure information of the second AI model; the target AI model matches the model structure information of the second AI model.
- the processing module is also used to adopt the target AI model to perform the first operation.
- the above-mentioned first operation includes at least one of the following: signal processing operation; signal transmission operation; signal demodulation operation; channel state information acquisition operation; beam management operation; channel prediction operation; interference suppression operation; positioning operation; prediction of high-level services and management operations; high-level parameter prediction and management operations; control signaling parsing operations.
- the target transmission mode of the target AI model is determined based on the capabilities of the AI model transmission mode supported by the first AI model transmission device, it is possible to avoid errors caused by the first AI model transmission device.
- the first transmission mode or the second transmission mode is not supported, resulting in the inability to obtain the target AI based on the target model information.
- the situation of the model which leads to the inability to apply the target AI model; and/or, since the target application time of the target AI model is determined based on the target transmission mode, it is possible to avoid the application time of the target AI model being too long (or is too small), resulting in the inability to apply the target AI model or a waste of waiting time; and/or, because the target AI model can trigger the first AI model transmission device to send feedback to the second AI model transmission device for feedback of the target AI model.
- the target feedback event of the application situation therefore, the second AI model transmission device can directly send the AI model that the first AI model transmission device can apply to the first AI model transmission device according to the target feedback event, without the need for the first AI model transmission device Perform multiple transmissions with the second AI model transmission device. Therefore, the number of transmissions between the first AI model transmission device and the second AI model transmission device during the process of obtaining the target AI model can be reduced, thus saving transmission resources.
- the AI model transmission 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.
- the AI model transmission device provided by the embodiments of this application can implement each process implemented by the method embodiments in Figures 2 to 5 and achieve the same technical effect. To avoid duplication, the details will not be described here.
- FIG. 8 shows a possible structural diagram of the AI model transmission device involved in the embodiment of the present application.
- the AI model transmission device is a second AI model transmission device.
- the second AI model transmission device 70 may include: a sending module 71 .
- the sending module 71 is used to send the target model information of the target AI model to the first AI model transmission device; the target model information is used by the first AI model transmission device to obtain the target AI model; the target AI model satisfies at least the following:
- the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device; the target application time of the target AI model is determined based on the target transmission mode; the target AI model triggers
- the first AI model transmission device sends a target feedback event to the second AI model transmission device 70 , where the target feedback event is used to feed back the application status of the target AI model.
- the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
- the target AI model satisfying the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device.
- the second AI model transmission device 70 provided by the embodiment of the present application may also include: a receiving module and a determining module.
- the receiving module is configured to receive target information from the first AI model transmission device, where the target information is used to indicate the capabilities of the AI model transmission mode supported by the first AI model transmission device.
- the determining module is used to determine the target transmission mode according to the target information received by the receiving module.
- the above-mentioned sending module 71 is also used to send target model information to the first AI model transmission device based on the target transmission mode determined by the determination module.
- the target transmission mode is the first transmission mode.
- the above-mentioned determination module is also used to determine whether the target AI model matches the pre-stored AI model in the first AI model transmission device.
- the above-mentioned sending module 71 is also used to send the target model information to the first AI model transmission device according to the fourth transmission mode if the determination module determines that the target AI model matches the pre-stored AI model.
- the above-mentioned fourth transmission mode is any of the following: the second transmission mode or the fifth transmission mode; when the fourth transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model; in When the fourth transmission mode is the fifth transmission mode, the target model information includes model structure information and model parameter information of the target AI model; the AI model application time corresponding to the fifth transmission mode is shorter than the AI model application time corresponding to the first transmission mode. time.
- the above-mentioned receiving module is also used to receive the target feedback event from the first AI model transmission device.
- the above-mentioned target feedback event is used to feedback the application status of the target AI model.
- the above-mentioned receiving module is also used to receive second model information of the second AI model from the first AI model transmission device.
- the second model information is transmitted by the first AI model transmission device according to the third transmission method. mode sent.
- the above-mentioned determining module is also used to determine the model structure information that matches the model structure information of the second AI model received by the receiving module as the model structure information of the target AI model.
- the above-mentioned third transmission mode is the first transmission mode or the fourth transmission mode; when the target transmission mode is the first transmission mode, the second model information includes model structure information and model parameter information of the second AI model; in When the target transmission mode is the fourth transmission mode, the second model information includes model structure information of the second AI model.
- the above determination module is specifically used for any of the following: determining the model structure information that is the same as the model structure information of the second AI model as the model structure information of the target AI model; The model structure whose similarity between the model structure information of the two AI models is greater than or equal to the preset threshold is determined as the model structure information of the target AI model.
- the AI model transmission device since the target transmission mode of the target AI model is determined according to the capability of the AI model transmission mode supported by the first AI model transmission device, it can avoid the situation that the target AI model cannot be obtained according to the target model information due to the first AI model transmission device not supporting the first transmission mode or the second transmission mode, and thus the situation that the target AI model cannot be applied; and/or, since the target application time of the target AI model is determined according to the target transmission mode, it can avoid the situation that the target AI model cannot be applied or the waiting time is wasted due to the application time of the target AI model being too long (or too short); and/or, since the target AI model can trigger the first AI model transmission device to send a target feedback event for feedback on the application of the target AI model to the second AI model transmission device, the second AI model transmission device can directly send the AI model that the first AI model transmission device can apply to the first AI model transmission device according to the target feedback event, without the need for the first AI model transmission device and the second AI model transmission device to perform multiple transmission
- the AI model transmission 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.
- the AI model transmission device provided in the embodiment of the present application can implement each process implemented by the method embodiment of Figure 6 and achieve the same technical effect. To avoid repetition, it will not be repeated here.
- this embodiment of the present application also provides a communication device 80, which includes a processor 81 and a memory 82.
- the memory 82 stores information that can run on the processor 81.
- a program or instruction For example, when the communication device 80 is a terminal, when the program or instruction is executed by the processor 81, each step of the above-mentioned AI model transmission method embodiment is implemented, and the same technical effect can be achieved.
- the communication device 80 is a network-side device, when the program or instruction is executed by the processor 81, the steps of the above-mentioned AI model transmission method embodiment are implemented, and the same technical effect can be achieved. To avoid duplication, they will not be described again here.
- An embodiment of the present application also provides a terminal.
- the terminal is a first terminal.
- the first terminal includes a processor and a communication interface.
- the communication interface is used to receive target model information of the target AI model from the second terminal.
- the processor is used to calculate the target AI model according to the target. Model information to obtain the target AI model.
- the target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first terminal; the target application time of the target AI model is determined based on the target transmission mode;
- the target AI model triggers the first terminal to send a target feedback event to the second terminal, and the target feedback event is used to feed back the application status of the target AI model.
- the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
- This terminal embodiment corresponds to the above-mentioned terminal-side method embodiment. Each implementation process and implementation manner of the above-mentioned method embodiment can be applied to this terminal embodiment, and can achieve the same technical effect.
- FIG. 10 is a schematic diagram of the hardware structure of a terminal that implements an embodiment of the present application.
- the terminal 800 includes but is not limited to: a radio frequency unit 801, a network module 802, an audio output unit 803, an input unit 804, a sensor 805, a display unit 806, a user input unit 807, an interface unit 808, a memory 809, a processor 810, etc. At least some parts.
- the terminal 800 may also include a power supply (such as a battery) that supplies power to various components.
- the power supply may be logically connected to the processor 810 through a power management system, thereby managing charging, discharging, and power consumption through the power management system. Management and other functions.
- the terminal structure shown in FIG. 10 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 804 may include a graphics processing unit (GPU) 8041 and a microphone 8042.
- the graphics processor 8041 is responsible for the image capture device (GPU) in the video capture mode or the image capture mode. Process the image data of still pictures or videos obtained by cameras (such as cameras).
- the display unit 806 may include a display panel 8061, which may be configured in the form of a liquid crystal display, an organic light emitting diode, or the like.
- the user input unit 807 includes a touch panel 8071 and at least one of other input devices 8072 . Touch panel 8071, Also called a touch screen.
- the touch panel 8071 may include two parts: a touch detection device and a touch controller.
- Other input devices 8072 may include but are not limited to physical keyboards, function keys (such as volume control keys, switch keys, etc.), trackballs, mice, and joysticks, which will not be described again here.
- the radio frequency unit 801 after receiving downlink data from the network side device, the radio frequency unit 801 can transmit it to the processor 810 for processing; in addition, the radio frequency unit 801 can send uplink data to the network side device.
- the radio frequency unit 801 includes, but is not limited to, an antenna, amplifier, transceiver, coupler, low noise amplifier, duplexer, etc.
- Memory 809 may be used to store software programs or instructions as well as various data.
- the memory 809 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first 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 809 may include volatile memory or non-volatile memory, or memory 809 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 810 may include one or more processing units; optionally, the processor 810 integrates an application processor and a modem processor, where the application processor mainly handles operations related to the operating system, user interface, application programs, etc., Modem processors mainly process wireless communication signals, such as baseband processors. It can be understood that the above modem processor may not be integrated into the processor 810.
- the radio frequency unit 801 is used to receive target model information of the target AI model from the second terminal.
- the above-mentioned target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined according to the capabilities of the AI model transmission mode supported by the first terminal; the target application time of the target AI model is determined according to the target transmission mode; the target AI model triggers the first terminal to send a target feedback event to the second terminal, and the target feedback event is used to feedback the application status of the target AI model.
- the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
- the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first terminal, it is possible to avoid errors caused by the first terminal not supporting the first transmission mode or the second transmission mode.
- transmission mode resulting in the inability to obtain the target AI model based on the target model information, resulting in the inability to apply the target AI model; and/or, because the target application time of the target AI model is determined based on the target transmission mode, therefore, It can avoid the situation where the target AI model cannot be applied or the waiting time is wasted because the application time of the target AI model is too long (or too small); and/or, because the target AI model can trigger the first terminal to send data to the second terminal.
- the second terminal can directly send the AI model that the first terminal can apply to the first terminal based on the target feedback event without the need for the first terminal to communicate with the second terminal. Multiple transfers. This can reduce the number of transmissions between the first terminal and the second terminal in the process of obtaining the target AI model, thus saving transmission resources.
- the target AI model satisfying the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first terminal.
- the radio frequency unit 801 is also used to send target information to the second terminal, where the target information is used to indicate the capabilities of the AI model transmission mode supported by the first terminal.
- the first terminal can send the target information to the second terminal, so that the second terminal can determine the target transmission mode according to the capability of the AI model transmission mode supported by the first terminal, therefore, it can be avoided that the first terminal does not support the target transmission mode.
- the first transmission mode or the second transmission mode causes the first terminal to be unable to obtain the target AI model, which further leads to the situation that the target AI model cannot be applied.
- the target application time of the target AI model that satisfies the target AI model is determined based on the target transmission mode.
- the processor 810 is used to determine the target application time according to the target transmission mode.
- the first terminal can determine the application time corresponding to the target transmission mode as the target application time; wherein the first application time is greater than the second application time, and/or the first application time is greater than or equal to the third application time, therefore , which can avoid situations where the target AI model cannot be applied or waiting time is wasted due to the application time of the target AI model being too large (or too small).
- the processor 810 is specifically configured to determine the application time corresponding to the target transmission mode as the target application time.
- At least one of the following is satisfied between the first application time and the second application time: the first application time is greater than the second application time; the first application time is greater than or equal to the third application time, and the third application time is based on the second application time Determined by the compilation time of the AI model.
- the above-mentioned first application time is: the AI model application time corresponding to the first transmission mode; the above-mentioned second application time is: the AI model application time corresponding to the second transmission mode.
- the first terminal can determine the application time corresponding to the target transmission mode as the target application time; wherein the first application time is greater than the second application time, and/or the first application time is greater than or equal to the third application time, therefore , which can avoid situations where the target AI model cannot be applied or waiting time is wasted due to the application time of the target AI model being too large (or too small).
- the processor 810 is specifically configured to, when the target transmission mode is the first transmission mode, if the target AI model matches the pre-stored AI model in the first terminal, then convert the fourth The application time is determined as the target application time.
- the above-mentioned fourth application time is less than the first application time; the first application time is: the AI model application time corresponding to the first transmission mode.
- the first terminal can determine a smaller application time as the target application time when the target AI model matches the pre-stored AI model, the delay in applying the target AI model can be reduced.
- the processor 810 is further configured to not apply the target AI model when the target AI model does not match the capability of the first terminal.
- the first terminal since the first terminal does not need to apply the target AI model when the target AI model does not match the capabilities of the first terminal, it can be avoided that the first terminal uses the target AI model for communication. Stuttering occurs.
- the radio frequency unit 801 is also configured to send the second model information of the second AI model to the second terminal according to the third transmission mode.
- the above-mentioned third transmission mode is the first transmission mode or the fourth transmission mode; when the target transmission mode is the first transmission mode, the second model information includes model structure information and model parameter information of the second AI model; in When the target transmission mode is the fourth transmission mode, the second model information includes model structure information of the second AI model; the target AI model matches the model structure information of the second AI model.
- the first terminal can directly receive the model parameter information of the target AI model from the second terminal to obtain the target AI model without repeatedly receiving the target.
- the model structure information of the AI model can save transmission resources.
- the processor 810 is also configured to use the target AI model to perform the first operation.
- the above-mentioned first operation includes at least one of the following: signal processing operation; signal transmission operation; signal demodulation operation; channel state information acquisition operation; beam management operation; channel prediction operation; interference suppression operation; positioning operation; prediction of high-level services and management operations; high-level parameter prediction and management operations; control signaling parsing operations.
- the first terminal can use the target AI model to perform signal processing operations, signal transmission operations, channel state information acquisition operations, beam management operations, channel prediction operations, interference suppression operations, positioning operations, and prediction and management operations of high-level services , at least one of a high-level parameter prediction and management operation and a control signaling parsing operation. Therefore, the throughput of the first terminal communication can be improved and the delay of the first terminal communication can be reduced.
- An embodiment of the present application also provides a terminal, which is a second terminal.
- the second terminal includes a processor and a communication interface.
- the communication interface is used to send target model information of the target AI model to the first terminal.
- the target model information is used by the first terminal to obtain the target AI model; the target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first terminal; the target AI The target application time of the model is Determined according to the target transmission mode; the target AI model triggers the first terminal to send a target feedback event to the second terminal, and the target feedback event is used to feed back the application status of the target AI model.
- the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
- This terminal embodiment corresponds to the above-mentioned terminal-side method embodiment. Each implementation process and implementation manner of the above-mentioned method embodiment can be applied to this terminal embodiment, and can achieve the same technical effect.
- FIG. 10 is a schematic diagram of the hardware structure of a terminal that implements an embodiment of the present application.
- the terminal 800 includes but is not limited to: a radio frequency unit 801, a network module 802, an audio output unit 803, an input unit 804, a sensor 805, a display unit 806, a user input unit 807, an interface unit 808, a memory 809 and at least some of the components of a processor 810.
- the terminal 800 may also include a power supply (such as a battery) that supplies power to various components.
- the power supply may be logically connected to the processor 810 through a power management system, thereby managing charging, discharging, and power consumption through the power management system. Management and other functions.
- the terminal structure shown in FIG. 10 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 804 may include a graphics processing unit (Graphics Processing Unit, GPU) 8041 and a microphone 8042.
- the graphics processor 8041 is responsible for the image capture device (GPU) in the video capture mode or the image capture mode. Process the image data of still pictures or videos obtained by cameras (such as cameras).
- the display unit 806 may include a display panel 8061, which may be configured in the form of a liquid crystal display, an organic light emitting diode, or the like.
- the user input unit 807 includes a touch panel 8071 and at least one of other input devices 8072 .
- Touch panel 8071 also known as touch screen.
- the touch panel 8071 may include two parts: a touch detection device and a touch controller.
- Other input devices 8072 may include but are not limited to physical keyboards, function keys (such as volume control keys, switch keys, etc.), trackballs, mice, and joysticks, which will not be described again here.
- the radio frequency unit 801 after receiving downlink data from the network side device, can transmit the data to the processor 810 for processing; in addition, the radio frequency unit 801 can send uplink data to the network side device.
- the radio frequency unit 801 includes but is not limited to an antenna, an amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, etc.
- Memory 809 may be used to store software programs or instructions as well as various data.
- the memory 809 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first 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 809 may include volatile memory or non-volatile memory, or memory 809 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 810 may include one or more processing units; optionally, the processor 810 integrates an application processor and a modem processor, where the application processor mainly handles operations related to the operating system, user interface, application programs, etc., Modem processors mainly process wireless communication signals, such as baseband processors. It can be understood that the above modem processor may not be integrated into the processor 810.
- the radio frequency unit 801 is used to send target model information of the target AI model to the first terminal.
- the above target model information is used by the first terminal to obtain the target AI model.
- the above target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first terminal; the target application time of the target AI model is determined based on the target transmission mode; The target AI model triggers the first terminal to send a target feedback event to the second terminal, and the target feedback event is used to feed back the application status of the target AI model.
- the above-mentioned target transmission mode includes any of the following: a first transmission mode and a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameters of the target AI model. data information; when the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
- the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first terminal, it is possible to avoid errors caused by the first terminal not supporting the first transmission mode or the second transmission mode.
- transmission mode resulting in the inability to obtain the target AI model based on the target model information, resulting in the inability to apply the target AI model; and/or, because the target application time of the target AI model is determined based on the target transmission mode, therefore, It can avoid the situation where the target AI model cannot be applied or the waiting time is wasted because the application time of the target AI model is too long (or too small); and/or, because the target AI model can trigger the first terminal to send data to the second terminal.
- the second terminal can directly send the AI model that the first terminal can apply to the first terminal based on the target feedback event without the need for the first terminal to communicate with the second terminal. Multiple transfers. This can reduce the number of transmissions between the first terminal and the second terminal in the process of obtaining the target AI model, thus saving transmission resources.
- the target AI model satisfying the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first terminal.
- the radio frequency unit 801 is also configured to receive target information from the first terminal, where the target information is used to indicate the capability of the AI model transmission mode supported by the first terminal.
- the processor 810 is configured to determine a target transmission mode according to the target information.
- the radio frequency unit 801 is also used to send target model information to the first terminal based on the target transmission mode.
- the second terminal can receive the target information from the first terminal, determine the target transmission mode according to the target information, and send the target model information to the first terminal according to the target transmission mode, it can avoid the problem that the first terminal does not support The first transmission mode or the second transmission mode causes the first device to be unable to obtain the target AI model, which further leads to the situation that the target AI model cannot be applied.
- the target transmission mode is the first transmission mode.
- the processor 810 is also used to determine whether the target AI model matches the pre-stored AI model in the first terminal.
- the radio frequency unit 801 is also used to send the target model information to the first terminal according to the fourth transmission mode if the target AI model matches the pre-stored AI model.
- the above-mentioned fourth transmission mode is any of the following: the second transmission mode or the fifth transmission mode; when the fourth transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model; in When the fourth transmission mode is the fifth transmission mode, the target model information includes model structure information and model parameter information of the target AI model.
- the AI model application time corresponding to the fifth transmission mode is shorter than the AI model application time corresponding to the first transmission mode. time.
- the second terminal can send the target model information to the first terminal according to the second transmission mode, that is, send the target AI model to the first terminal. model parameter information without repeatedly sending the model structure information of the target AI model, thus saving transmission resources; or, the second terminal can send the target model information to the first terminal according to the fifth transmission mode, so that the first terminal can The smaller application time is determined as the target application time, so the delay in applying the target AI model can be reduced.
- the radio frequency unit 801 is also used to receive a target feedback event from the first terminal.
- the above-mentioned target feedback event is used to feedback the application status of the target AI model.
- the second terminal can receive a target feedback event for feedback on the application status of the target AI model from the first terminal, the second terminal can directly send to the first terminal the information that the first terminal can apply based on the target feedback event. AI model without the need for multiple transmissions between the first terminal and the second terminal.
- the radio frequency unit 801 is also configured to receive second model information of the second AI model from the first terminal, where the second model information is sent by the first terminal according to the third transmission mode.
- the processor 810 is also configured to determine the model structure information that matches the model structure information of the second AI model as the model structure information of the target AI model.
- the above-mentioned third transmission mode is the first transmission mode or the fourth transmission mode; when the target transmission mode is the first transmission mode, the second model information includes model structure information and model parameter information of the second AI model; in When the target transmission mode is the fourth transmission mode, the second model information includes model structure information of the second AI model.
- the first terminal can directly receive the model parameter information of the target AI model from the second terminal to receive the target AI model without repeatedly receiving the target.
- the model structure information of the AI model can save transmission resources.
- the processor 810 is specifically used for any of the following:
- the model structure whose similarity to the model structure information of the second AI model is greater than or equal to the preset threshold is determined as the model structure information of the target AI model.
- 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 target model information of the target AI model from the second device, and the processor is used to obtain the target AI model according to the target model information; or , used to send the target model information of the target AI model to the first device; the target model information is used by the first device to obtain the target AI model; the target AI model satisfies at least one of the following: the target transmission mode of the target AI model is based on The capability of the AI model transmission mode supported by the first device is determined; the target application time of the target AI model is determined based on the target transmission mode; the target AI model triggers the first device to send a target feedback event to the second device, and the target feedback event Used to provide feedback on the application of the target AI model.
- the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
- This network-side device embodiment corresponds to the above-mentioned network-side device 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 900 includes: an antenna 901, a radio frequency device 902, a baseband device 903, a processor 904 and a memory 905.
- the antenna 901 is connected to the radio frequency device 902 .
- the radio frequency device 902 receives information through the antenna 901 and sends the received information to the baseband device 903 for processing.
- the baseband device 903 processes the information to be sent and sends it to the radio frequency device 902.
- the radio frequency device 902 processes the received information and then sends it out through the antenna 901.
- the method performed by the network side device in the above embodiment can be implemented in the baseband device 903, which includes a baseband processor.
- the baseband device 903 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 906, which is, for example, a common public radio interface (CPRI).
- a network interface 906 which is, for example, a common public radio interface (CPRI).
- CPRI common public radio interface
- the network side device 900 in the embodiment of the present application also includes: instructions or programs stored in the memory 905 and executable on the processor 904.
- the processor 904 calls the instructions or programs in the memory 905 to execute each of the steps shown in Figure 11. The method of module execution and achieving the same technical effect will not be described in detail here to avoid duplication.
- An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored.
- a program or instruction is stored.
- the various processes of the above-mentioned AI model transmission method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.
- the processor is the processor in the terminal described in the above embodiment.
- the readable storage medium includes computer readable storage media, such as computer read-only memory ROM, random access memory RAM, magnetic disk or optical disk, etc.
- An embodiment of the present application further provides a chip.
- the chip includes a processor and a communication interface.
- the communication interface is coupled to the processor.
- the processor is used to run programs or instructions to implement the above embodiment of the AI model transmission method. Each process can achieve the same technical effect. To avoid repetition, we will not go into details here.
- chips mentioned in the embodiments of this application may also be called system-on-chip, system-on-a-chip, system-on-chip or system-on-chip, etc.
- Embodiments of the present application further provide a computer program/program product.
- the computer program/program product is stored in a storage medium.
- the computer program/program product is executed by at least one processor to implement the above-mentioned AI model transmission method.
- Each process in the example can achieve the same technical effect. To avoid repetition, we will not repeat it here.
- Embodiments of the present application also provide an AI model transmission system, including: a first device and a second device.
- the first device can be used to perform the steps of the AI model transmission method as described above.
- the second device can be used to perform the above steps. The steps of the AI model transmission method.
- 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)
- Signal Processing (AREA)
- Computer Networks & Wireless Communication (AREA)
- Physics & Mathematics (AREA)
- Mathematical Physics (AREA)
- Artificial Intelligence (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Medical Informatics (AREA)
- Software Systems (AREA)
- Mobile Radio Communication Systems (AREA)
Abstract
The present application belongs to the technical field of communications. Disclosed are an artificial intelligence (AI) model transmission method and apparatus, and a terminal and a medium. The AI model transmission method in the embodiments of the present application comprises: a first device receiving target model information of a target AI model from a second device; the first device obtaining the target AI model according to the target model information, wherein the target AI model satisfies at least one of the following: a target transmission mode of the target AI model being determined according to the capability of an AI model transmission mode supported by the first device, and a target application time of the target AI model being determined according to the target transmission mode; and the target AI model triggering the first device to send a target feedback event to the second device, wherein the target feedback event is used for feeding back an application situation of the target AI model, wherein the target transmission mode comprises either of the following: a first transmission mode and a second transmission mode.
Description
本申请要求于2022年9月23日提交国家知识产权局、申请号为202211167997.0、申请名称为“人工智能AI模型传输方法、装置、终端及介质”的中国专利申请的优先权,其全部内容通过引用结合在本申请中。This application requests the priority of the Chinese patent application submitted to the State Intellectual Property Office on September 23, 2022, with the application number 202211167997.0 and the application name "Artificial Intelligence AI Model Transmission Method, Device, Terminal and Medium", and its entire content has been approved This reference is incorporated into this application.
本申请属于通信技术领域,具体涉及一种AI模型传输方法、装置、终端及介质。The present application belongs to the field of communication technology, and specifically relates to an AI model transmission method, device, terminal and medium.
目前,在无线通信网络中,终端设备可以从其他设备接收训练好的人工智能(Artificial Intelligence,AI)模型,并使用训练好的AI模型进行通信,从而可以提升通信的吞吐量,且降低通信的时延。Currently, in wireless communication networks, terminal devices can receive trained artificial intelligence (AI) models from other devices and use the trained AI models to communicate, thereby improving communication throughput and reducing communication costs. time delay.
但是,由于可能会出现终端设备无法应用从其他设备接收的训练好的AI模型的情况,此时,终端设备可能需要与其他设备进行多次传输,才可以得到终端设备能够应用的训练好的AI模型,以使用训练好的AI模型进行通信。However, there may be situations where the terminal device cannot apply the trained AI model received from other devices. At this time, the terminal device may need to perform multiple transmissions with other devices before it can obtain the trained AI model that the terminal device can apply. model to communicate using the trained AI model.
因此,可能会导致传输资源的浪费。Therefore, transmission resources may be wasted.
发明内容Contents of the invention
本申请实施例提供一种AI模型传输方法、装置、终端及介质,能够解决传输资源浪费的问题。Embodiments of the present application provide an AI model transmission method, device, terminal and medium, which can solve the problem of waste of transmission resources.
第一方面,提供了一种AI模型传输方法,应用于第一设备,该方法包括:第一设备从第二设备接收目标AI模型的目标模型信息;第一设备根据目标模型信息,得到目标AI模型;该目标AI模型满足以下至少一项:目标AI模型的目标传输模式,是根据第一设备支持的AI模型传输模式的能力确定的;目标AI模型的目标应用时间,是根据目标传输模式确定的;目标AI模型触发第一设备向第二设备发送目标反馈事件,该目标反馈事件用于反馈目标AI模型的应用情况。其中,上述目标传输模式包括以下任一项:第一传输模式、第二传输模式;在目标传输模式为第一传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;在目标传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息。In a first aspect, an AI model transmission method is provided, which is applied to a first device. The method includes: the first device receives target model information of the target AI model from a second device; the first device obtains the target AI based on the target model information. model; the target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first device; the target application time of the target AI model is determined based on the target transmission mode ; the target AI model triggers the first device to send a target feedback event to the second device, and the target feedback event is used to feed back the application status of the target AI model. Wherein, the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
第二方面,提供了一种AI模型传输装置,该AI模型传输装置为第一AI模型传输装置,该第一AI模型传输装置包括:接收模块。其中,接收模块,用于从第二AI模型传输装置接收目标AI模型的目标模型信息;第一AI模型传输装置根据目标模型信息,得到目标AI模型;该目标AI模型满足以下至少一项:目标AI模型的目标传输模式,是根据第一AI模型传输装置支持的AI模型传输模式的能力确定的;目标AI模型的目标应用时间,是根据目标传输模式确定的;目标AI模型触发第一AI模型传输装置向第二AI模型传输装置发送目标反馈事件,该目标反馈事件用于反馈目标AI模型的应用情况。其中,上述目标传输模式包括以下任一项:第一传输模式、第二传输模式;在目标传输模式为第一传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;在目标传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息。In a second aspect, an AI model transmission device is provided. The AI model transmission device is a first AI model transmission device. The first AI model transmission device includes: a receiving module. Among them, the receiving module is used to receive the target model information of the target AI model from the second AI model transmission device; the first AI model transmission device obtains the target AI model according to the target model information; the target AI model satisfies at least one of the following: target The target transmission mode of the AI model is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device; the target application time of the target AI model is determined based on the target transmission mode; the target AI model triggers the first AI model The transmission device sends a target feedback event to the second AI model transmission device, where the target feedback event is used to feed back the application status of the target AI model. Wherein, the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
第三方面,提供了一种AI模型传输方法,应用于第二设备,该方法包括:第二设备向第一设备发送目标AI模型的目标模型信息;该目标模型信息用于第一设备得到目标AI模型;该目标AI模型满足以下至少一项:目标AI模型的目标传输模式,是根据第一设备支持的AI模型传输模式的能力确定的;目标AI模型的目标应用时间,是根据目标传输模式确定的;目标AI模型触发第一设备向第二设备发送目标反馈事件,该目标反馈事件用于反馈目标AI模型的应用情况。其中,上述目标传输模式包括以下任一项:第一传输模式、第二传输模式;在目标传输模式为第一传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;在目标传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息。In a third aspect, an AI model transmission method is provided, applied to a second device. The method includes: the second device sends target model information of the target AI model to the first device; the target model information is used by the first device to obtain the target AI model; the target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first device; the target application time of the target AI model is determined based on the target transmission mode Determined; the target AI model triggers the first device to send a target feedback event to the second device, and the target feedback event is used to feed back the application status of the target AI model. Wherein, the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
第四方面,提供了一种AI模型传输装置,该AI模型传输装置为第二AI模型传输装置,
该第二AI模型传输装置包括:发送模块。其中,发送模块,用于向第一AI模型传输装置发送目标AI模型的目标模型信息;该目标模型信息用于第一设备得到目标AI模型;该目标AI模型满足以下至少一项:目标AI模型的目标传输模式,是根据第一AI模型传输装置支持的AI模型传输模式的能力确定的;目标AI模型的目标应用时间,是根据目标传输模式确定的;目标AI模型触发第一AI模型传输装置向第二AI模型传输装置发送目标反馈事件,该目标反馈事件用于反馈目标AI模型的应用情况。其中,上述目标传输模式包括以下任一项:第一传输模式、第二传输模式;在目标传输模式为第一传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;在目标传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息。In a fourth aspect, an AI model transmission device is provided, and the AI model transmission device is a second AI model transmission device, The second AI model transmission device includes: a sending module. Among them, the sending module is used to send the target model information of the target AI model to the first AI model transmission device; the target model information is used by the first device to obtain the target AI model; the target AI model satisfies at least one of the following: target AI model The target transmission mode is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device; the target application time of the target AI model is determined based on the target transmission mode; the target AI model triggers the first AI model transmission device A target feedback event is sent to the second AI model transmission device, where the target feedback event is used to feed back the application status of the target AI model. Wherein, the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
第五方面,提供了一种终端,该终端包括处理器和存储器,所述存储器存储可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如第一方面所述的方法的步骤,或实现如第三方面所述的方法的步骤。In a fifth aspect, a terminal is provided, which includes a processor and a memory, wherein the memory stores a program or instruction that can be run on the processor, and when the program or instruction is executed by the processor, the steps of the method described in the first aspect are implemented, or the steps of the method described in the third aspect are implemented.
第六方面,提供了一种终端,包括处理器及通信接口,其中,该通信接口用于从第二设备接收目标AI模型的目标模型信息,该处理器用于根据目标模型信息,得到目标AI模型;或者,该通信结构用于向第一设备发送目标AI模型的目标模型信息,该目标AI模型用于第一设备得到目标AI模型。其中,该目标AI模型满足以下至少一项:目标AI模型的目标传输模式,是根据第一AI模型传输装置支持的AI模型传输模式的能力确定的;目标AI模型的目标应用时间,是根据目标传输模式确定的;目标AI模型触发第一AI模型传输装置向第二AI模型传输装置发送目标反馈事件,该目标反馈事件用于反馈目标AI模型的应用情况。其中,上述目标传输模式包括以下任一项:第一传输模式、第二传输模式;在目标传输模式为第一传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;在目标传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息。In a sixth aspect, a terminal is provided, including a processor and a communication interface, wherein the communication interface is used to receive target model information of the target AI model from the second device, and the processor is used to obtain the target AI model based on the target model information. ; Or, the communication structure is used to send target model information of the target AI model to the first device, and the target AI model is used by the first device to obtain the target AI model. Wherein, the target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device; the target application time of the target AI model is determined based on the target The transmission mode is determined; the target AI model triggers the first AI model transmission device to send a target feedback event to the second AI model transmission device, and the target feedback event is used to feed back the application status of the target AI model. Wherein, the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
第七方面,提供了一种网络侧设备,该网络侧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,或实现如第三方面所述的方法的步骤。In a seventh aspect, a network side device is provided. The network side device includes a processor and a memory. The memory stores programs or instructions that can be run on the processor. The program or instructions are executed by the processor. When implementing the steps of the method described in the first aspect, or implementing the steps of the method described in the third aspect.
第八方面,提供了一种网络侧设备,包括处理器及通信接口,其中,该通信接口用于从第二设备接收目标AI模型的目标模型信息,该处理器用于根据目标模型信息,得到目标AI模型;或者,该通信结构用于向第一设备发送目标AI模型的目标模型信息,该目标AI模型用于第一设备得到目标AI模型。其中,该目标AI模型满足以下至少一项:目标AI模型的目标传输模式,是根据第一AI模型传输装置支持的AI模型传输模式的能力确定的;目标AI模型的目标应用时间,是根据目标传输模式确定的;目标AI模型触发第一AI模型传输装置向第二AI模型传输装置发送目标反馈事件,该目标反馈事件用于反馈目标AI模型的应用情况。其中,上述目标传输模式包括以下任一项:第一传输模式、第二传输模式;在目标传输模式为第一传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;在目标传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息。In an eighth aspect, a network side device is provided, including a processor and a communication interface, wherein the communication interface is used to receive target model information of the target AI model from the second device, and the processor is used to obtain the target based on the target model information. AI model; or, the communication structure is used to send target model information of the target AI model to the first device, and the target AI model is used by the first device to obtain the target AI model. Wherein, the target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device; the target application time of the target AI model is determined based on the target The transmission mode is determined; the target AI model triggers the first AI model transmission device to send a target feedback event to the second AI model transmission device, and the target feedback event is used to feed back the application status of the target AI model. Wherein, the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
第九方面,提供了一种AI模型传输***,包括:第一终端及第二终端,该第一终端可用于执行如第一方面所述的方法的步骤,该第二终端可用于执行如第三方面所述的方法的步骤。A ninth aspect provides an AI model transmission system, including: a first terminal and a second terminal. The first terminal can be used to perform the steps of the method described in the first aspect, and the second terminal can be used to perform the steps of the method described in the first aspect. The steps of the method described in three aspects.
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,或者实现如第三方面所述的方法的步骤。In the tenth aspect, a readable storage medium is provided, on which a program or instruction is stored. When the program or instruction is executed by a processor, the steps of the method described in the first aspect are implemented, or the steps of the method described in the third aspect are implemented.
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的步骤,或实现如第三方面所述的方法的步骤。In an eleventh aspect, a chip is provided. 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 method described in the first aspect. The steps of a method, or steps of implementing a method as described in the third aspect.
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方法的步骤,或实现如第三方面所述的方法的步骤。In a twelfth 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 method, or the steps of implementing the method as described in the third aspect.
在本申请实施例中,第一设备可以从第二设备接收目标AI模型的目标模型信息,并根据目标模型信息,得到目标AI模型,该目标AI模型满足以下至少一项:目标AI模型的目
标传输模式,是根据第一设备支持的AI模型传输模式的能力确定的;目标AI模型的目标应用时间,是根据目标传输模式确定的;目标AI模型触发第一设备向第二设备发送目标反馈事件,该目标反馈事件用于反馈目标AI模型的应用情况;其中,上述目标传输模式包括第一传输模式和第二传输模式中的任一个;在目标传输模式为第一传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;在目标传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息。由于目标AI模型的目标传输模式,是根据第一设备支持的AI模型传输模式的能力确定的,因此,可以避免因第一设备不支持第一传输模式或第二传输模式,而导致无法根据目标模型信息得到目标AI模型的情况,进而导致无法应用目标AI模型的情况;和/或,由于目标AI模型的目标应用时间,是根据目标传输模式确定的,因此,可以避免因目标AI模型的应用时间过大(或过小),而导致无法应用目标AI模型、或浪费等待时间的情况;和/或,由于目标AI模型可以触发第一设备向第二设备发送用于反馈目标AI模型的应用情况的目标反馈事件,因此,第二设备可以直接根据目标反馈事件,向第一设备发送第一设备能够应用的AI模型,而无需第一设备与第二设备进行多次传输。从而可以减少在得到目标AI模型的过程中,第一设备与第二设备进行传输的次数,如此,可以节省传输资源。In an embodiment of the present application, the first device may receive target model information of the target AI model from the second device, and obtain the target AI model according to the target model information, and the target AI model satisfies at least one of the following: The target transmission mode is determined according to the capability of the AI model transmission mode supported by the first device; the target application time of the target AI model is determined according to the target transmission mode; the target AI model triggers the first device to send a target feedback event to the second device, and the target feedback event is used to feedback the application status of the target AI model; wherein, the above-mentioned target transmission mode includes any one of the first transmission mode and the second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes the model structure information and model parameter information of the target AI model; when the target transmission mode is the second transmission mode, the target model information includes the model parameter information of the target AI model. Since the target transmission mode of the target AI model is determined according to the capability of the AI model transmission mode supported by the first device, it is possible to avoid the situation where the target AI model cannot be obtained according to the target model information due to the first device not supporting the first transmission mode or the second transmission mode, and thus the target AI model cannot be applied; and/or, since the target application time of the target AI model is determined according to the target transmission mode, it is possible to avoid the situation where the target AI model cannot be applied or the waiting time is wasted due to the application time of the target AI model being too long (or too short); and/or, since the target AI model can trigger the first device to send a target feedback event for feedback on the application of the target AI model to the second device, the second device can directly send the AI model that the first device can apply to the first device according to the target feedback event, without the need for the first device and the second device to perform multiple transmissions. Thereby, the number of transmissions between the first device and the second device in the process of obtaining the target AI model can be reduced, so that transmission resources can be saved.
图1是本申请实施例提供的一种无线通信***的框图;Figure 1 is a block diagram of a wireless communication system provided by an embodiment of the present application;
图2是本申请实施例提供的AI模型传输方法的流程示意图之一;Figure 2 is one of the flow diagrams of the AI model transmission method provided by the embodiment of the present application;
图3是本申请实施例提供的AI模型传输方法的流程示意图之二;Figure 3 is the second schematic flow chart of the AI model transmission method provided by the embodiment of the present application;
图4是本申请实施例提供的AI模型传输方法的流程示意图之三;Figure 4 is the third schematic flowchart of the AI model transmission method provided by the embodiment of the present application;
图5是本申请实施例提供的AI模型传输方法的流程示意图之四;Figure 5 is the fourth schematic flowchart of the AI model transmission method provided by the embodiment of the present application;
图6是本申请实施例提供的AI模型传输方法的流程示意图之五;Figure 6 is the fifth schematic flow chart of the AI model transmission method provided by the embodiment of the present application;
图7是本申请实施例提供的第一AI模型传输装置的结构示意图;Figure 7 is a schematic structural diagram of a first AI model transmission device provided by an embodiment of the present application;
图8是本申请实施例提供的第二AI模型传输装置的结构示意图;Figure 8 is a schematic structural diagram of a second AI model transmission device provided by an embodiment of the present application;
图9是本申请实施例提供的通信设备的结构示意图;Figure 9 is a schematic structural diagram of a communication device provided by an embodiment of the present application;
图10是本申请实施例提供的终端的硬件结构示意图;Figure 10 is a schematic diagram of the hardware structure of a terminal provided by an embodiment of the present application;
图11是本申请实施例提供的网络侧设备的硬件结构示意图。Figure 11 is a schematic diagram of the hardware structure of a network-side device provided by an embodiment of the present application.
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。The technical solutions in the embodiments of the present application will be clearly described below with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, but not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those of ordinary skill in the art fall within the scope of protection of this application.
以下将对本申请实施例涉及的术语进行说明。The terms used in the embodiments of this application will be described below.
1、AI模型1. AI model
目前,AI模型已经在各个领域得到了广泛的应用。通常,可以采用训练样本对AI模型进行训练,以得到训练好的AI模型,从而可以使用训练好的AI模型进行通信,以提升通信的吞吐量,且降低通信的时延。At present, AI models have been widely used in various fields. Generally, the AI model can be trained using training samples to obtain a trained AI model, so that the trained AI model can be used for communication to improve communication throughput and reduce communication delay.
其中,AI模型可以包括以下至少一项:神经网络模型、决策树模型、支持向量机模型、贝叶斯分类器模型等。Among them, the AI model may include at least one of the following: neural network model, decision tree model, support vector machine model, Bayesian classifier model, etc.
2、其他术语2. Other terms
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。The terms "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. For example, the first object can be one or multiple. In addition, "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.
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)***,还可用于其他无线通信***,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***。本申请实施例中的术语“***”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***和无线电技术,也可用于其他***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)***,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***应用以外的应用,如第6代(6th Generation,6G)通信***。It is worth pointing out that the technology described in the embodiments of this application is not limited to Long Term Evolution (LTE)/LTE Evolution (LTE-Advanced, LTE-A) systems, and can also be used in other wireless communication systems, such as code Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Frequency Division Multiple Access (FDMA), Orthogonal Frequency Division Multiple Access, OFDMA), single-carrier frequency division multiple access (Single-carrier Frequency Division Multiple Access, SC-FDMA) and other systems. The terms "system" and "network" in the embodiments of this application are often used interchangeably, and the described technology can be used not only for the above-mentioned systems and radio technologies, but also for other systems and radio technologies. The following description describes a New Radio (NR) system for example purposes, and NR terminology is used in much of the following description, but these techniques can also be applied to applications other than NR system applications, such as 6th generation Generation, 6G) communication system.
图1示出本申请实施例可应用的一种无线通信***的框图。无线通信***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(VUE)、行人终端(PUE)、智能家居(具有无线通信功能的家居设备,如冰箱、电视、洗衣机或者家具等)、游戏机、个人计算机(personal computer,PC)、柜员机或者自助机等终端侧设备,可穿戴式设备包括:智能手表、智能手环、智能耳机、智能眼镜、智能首饰(智能手镯、智能手链、智能戒指、智能项链、智能脚镯、智能脚链等)、智能腕带、智能服装等。需要说明的是,在本申请实施例并不限定终端11的具体类型。网络侧设备12可以包括接入网设备或核心网设备,其中,接入网设备12也可以称为无线接入网设备、无线接入网(Radio Access Network,RAN)、无线接入网功能或无线接入网单元。接入网设备12可以包括基站、WLAN接入点或WiFi节点等,基站可被称为节点B、演进节点B(eNB)、接入点、基收发机站(Base Transceiver Station,BTS)、无线电基站、无线电收发机、基本服务集(Basic Service Set,BSS)、扩展服务集(Extended Service Set,ESS)、家用B节点、家用演进型B节点、发送接收点(Transmitting Receiving Point,TRP)或所述领域中其他某个合适的术语,只要达到相同的技术效果,所述基站不限于特定技术词汇,需要说明的是,在本申请实施例中仅以NR***中的基站为例进行介绍,并不限定基站的具体类型。Figure 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. Among them, the terminal 11 can 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 handheld computer, a netbook, or a super mobile personal computer. (ultra-mobile personal computer, UMPC), mobile Internet device (Mobile Internet Device, MID), augmented reality (augmented reality, AR)/virtual reality (VR) equipment, robots, wearable devices (Wearable Device) , vehicle-mounted equipment (VUE), pedestrian terminal (PUE), smart home (home equipment with wireless communication functions, such as refrigerators, TVs, washing machines or furniture, etc.), game consoles, personal computers (personal computers, PC), teller machines or self-service Terminal devices such as mobile phones, wearable devices include: smart watches, smart bracelets, smart headphones, smart glasses, smart jewelry (smart bracelets, smart bracelets, smart rings, smart necklaces, smart anklets, smart anklets, etc.), Smart wristbands, smart clothing, etc. It should be noted that the embodiment of the present application does not limit the specific type of the terminal 11. The network side equipment 12 may include access network equipment or core network equipment, where the access network equipment 12 may also be called wireless access network equipment, radio access network (Radio Access Network, RAN), radio access network function or Wireless access network unit. The access network device 12 may include a base station, a WLAN access point or a WiFi node, etc. The base station may be called a Node B, an evolved Node B (eNB), an access point, a Base Transceiver Station (BTS), a radio Base station, radio transceiver, Basic Service Set (BSS), Extended Service Set (ESS), Home B-Node, Home Evolved B-Node, Transmitting Receiving Point (TRP) or all Some other appropriate terminology in the above 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 the base station in the NR system is used as an example for introduction, and The specific type of base station is not limited.
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AI模型传输方法、装置、终端及介质进行详细地说明。The AI model transmission method, device, terminal and medium provided by the embodiments of the present application will be described in detail below through some embodiments and application scenarios with reference to the accompanying drawings.
图2示出了本申请实施例提供的一种AI模型传输方法的流程图。如图2所示,本申请实施例提供的AI模型传输方法可以包括下述的步骤101和步骤102。Figure 2 shows a flow chart of an AI model transmission method provided by an embodiment of the present application. As shown in Figure 2, the AI model transmission method provided by the embodiment of the present application may include the following steps 101 and 102.
步骤101、第一设备从第二设备接收目标AI模型的目标模型信息。Step 101: The first device receives target model information of the target AI model from the second device.
可选地,本申请实施例中,上述第一设备可以为以下任一项:用户设备(User Equipment,UE)、网络侧设备。上述第二设备可以为以下任一项:UE、网络侧设备。Optionally, in this embodiment of the present application, the above-mentioned first device may be any of the following: user equipment (User Equipment, UE) or network side equipment. The above-mentioned second device may be any of the following: UE or network side device.
在一种示例中,第一设备可以为UE,且第二设备可以为网络侧设备。In an example, the first device may be a UE, and the second device may be a network side device.
在另一种示例中,第一设备可以为网络侧设备,第二设备可以为UE。In another example, the first device may be a network side device, and the second device may be a UE.
在又一种示例中,第一设备和第二设备均为UE。In yet another example, both the first device and the second device are UEs.
在又一种示例中,第一设备和第二设备均为网络侧设备。例如,第一设备和第二设备可以为网络侧的不同节点;或者,第一设备和第二设备可以为网络侧的不同终端节点。In yet another example, both the first device and the second device are network-side devices. For example, the first device and the second device may be different nodes on the network side; or, the first device and the second device may be different terminal nodes on the network side.
可选地,本申请实施例中,上述目标AI模型可以为以下任一项:神经网络模型、决策树模型、支持向量机模型、贝叶斯分类器模型等。Optionally, in this embodiment of the present application, the target AI model may be any of the following: a neural network model, a decision tree model, a support vector machine model, a Bayesian classifier model, etc.
可选地,本申请实施例中,目标模型信息可以包括以下至少一项:目标AI模型的模型结构信息、目标AI模型的模型参数信息。Optionally, in this embodiment of the present application, the target model information may include at least one of the following: model structure information of the target AI model, and model parameter information of the target AI model.
其中,上述目标AI模型的模型结构信息用于指示目标AI模型的模型结构。上述目标AI模型的模型参数信息可以为AI模型的权重值。Wherein, the above model structure information of the target AI model is used to indicate the model structure of the target AI model. The model parameter information of the above target AI model may be the weight value of the AI model.
示例性地,假设目标AI模型为神经网络模型,则目标AI模型的模型参数信息可以为该神经网络模型的至少部分神经元的权重值。这里,神经元的权重值可以包括以下任一项:权值系数、乘性系数、偏置系数、加性系数、激活函数的类型、激活函数的系数。For example, assuming that the target AI model is a neural network model, the model parameter information of the target AI model may be the weight values of at least some neurons of the neural network model. Here, the weight value of the neuron can include any of the following: weight coefficient, multiplicative coefficient, bias coefficient, additive coefficient, type of activation function, and coefficient of the activation function.
本申请实施例中,上述目标AI模型满足以下至少一项:In the embodiment of this application, the above target AI model satisfies at least one of the following:
目标AI模型的目标传输模式,是根据第一设备支持的AI模型传输模式的能力确定的;The target transmission mode of the target AI model is determined based on the capabilities of the AI model transmission mode supported by the first device;
目标AI模型的目标应用时间,是根据目标传输模式确定的;The target application time of the target AI model is determined based on the target transmission mode;
目标AI模型触发第一设备向第二设备发送目标反馈事件,该目标反馈事件用于反馈目标AI模型的应用情况。The target AI model triggers the first device to send a target feedback event to the second device, and the target feedback event is used to feed back the application status of the target AI model.
本申请实施例中,上述目标传输模式包括以下任一项:第一传输模式、第二传输模式;
在目标传输模式为第一传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;在目标传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息。In the embodiment of the present application, the above-mentioned target transmission mode includes any one of the following: a first transmission mode, a second transmission mode; When the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model; when the target transmission mode is the second transmission mode, the target model information includes the model structure information and model parameter information of the target AI model. Model parameter information.
其中,第一传输模式为:传输AI模型的模型结构信息和模型参数信息的传输模式;该第二传输模式为:传输AI模型的模型参数信息的传输模式。The first transmission mode is a transmission mode that transmits the model structure information and model parameter information of the AI model; the second transmission mode is a transmission mode that transmits the model parameter information of the AI model.
可以理解,在目标传输模式为第一传输模式的情况下,第一设备会从第二设备接收目标AI模型的模型结构信息和模型参数信息。在目标传输模式为第二传输模式的情况下,第一设备会从第二设备接收目标AI模型的模型参数信息。It can be understood that when the target transmission mode is the first transmission mode, the first device will receive the model structure information and model parameter information of the target AI model from the second device. When the target transmission mode is the second transmission mode, the first device receives model parameter information of the target AI model from the second device.
其中,上述第一设备支持的AI模型传输模式的能力包括以下任一项:Among them, the capabilities of the AI model transmission mode supported by the first device include any of the following:
支持第一传输模式;Support the first transmission mode;
支持第二传输模式;Support the second transmission mode;
支持第一传输模式和第二传输模式。Supports first transmission mode and second transmission mode.
可选地,本申请实施例中,目标传输模式可以是第一设备根据第一设备支持的AI模型传输模式的能力确定的;或者,可以是第二设备根据第一设备支持的AI模型传输模式的能力确定的。Optionally, in this embodiment of the present application, the target transmission mode may be determined by the first device according to the capability of the AI model transmission mode supported by the first device; or, it may be determined by the second device according to the AI model transmission mode supported by the first device. The ability is determined.
本申请实施例中,上述目标应用时间为目标AI模型的切换时间或生效时间。In an embodiment of the present application, the above-mentioned target application time is the switching time or effectiveness time of the target AI model.
需要说明的是,上述“目标AI模型的切换时间”可以理解为:第一设备将接收目标AI模型之前应用的AI模型去激活所需的时长。上述“目标AI模型的生效时间”可以理解为:第一设备将接收目标AI模型之前应用的AI模型去激活,且将目标AI模型激活所需的时长。It should be noted that the above "switching time of the target AI model" can be understood as: the time required for the first device to deactivate the AI model applied before receiving the target AI model. The above "effective time of the target AI model" can be understood as: the time required for the first device to deactivate the AI model applied before receiving the target AI model and activate the target AI model.
可选地,本申请实施例中,目标应用时间可以是第一设备根据目标传输模式确定的;或者,可以是第二设备根据目标传输模式确定的。Optionally, in this embodiment of the present application, the target application time may be determined by the first device based on the target transmission mode; or may be determined by the second device based on the target transmission mode.
可选地,本申请实施例中,上述目标AI模型的应用情况可以包括以下至少一项:第一设备是否应用目标AI模型、第一设备对接收目标AI模型之前应用的AI模型的使用情况、目标AI模型的模型标识信息、第一设备未应用目标AI模型的原因。Optionally, in this embodiment of the present application, the application of the target AI model may include at least one of the following: whether the first device applies the target AI model, the use of the AI model applied by the first device before receiving the target AI model, Model identification information of the target AI model and the reason why the first device did not apply the target AI model.
步骤102、第一设备根据目标模型信息,得到目标AI模型。Step 102: The first device obtains the target AI model based on the target model information.
可选地,本申请实施例中,在目标传输模式为第一传输模式的情况下,第一设备会从第二设备接收目标AI模型的模型结构信息和模型参数信息,从而第一设备可以先对目标AI模型的模型结构信息进行编译(或重编译),以确定目标AI模型的模型结构,再将目标AI模型的模型参数信息应用于目标AI模型的模型结构中,以得到目标AI模型。Optionally, in this embodiment of the present application, when the target transmission mode is the first transmission mode, the first device will receive the model structure information and model parameter information of the target AI model from the second device, so that the first device can first Compile (or recompile) the model structure information of the target AI model to determine the model structure of the target AI model, and then apply the model parameter information of the target AI model to the model structure of the target AI model to obtain the target AI model.
在目标传输模式为第二传输模式的情况下,第一设备会从第二设备接收目标AI模型的模型参数信息,从而第一设备可以将目标AI模型的模型参数信息应用于第一设备中预存的模型结构中,以得到目标AI模型。这里,第一设备不需要先对目标AI模型的模型结构信息进行编译(或重编译)。When the target transmission mode is the second transmission mode, the first device will receive the model parameter information of the target AI model from the second device, so that the first device can apply the model parameter information of the target AI model to the pre-stored data in the first device. in the model structure to obtain the target AI model. Here, the first device does not need to compile (or recompile) the model structure information of the target AI model first.
本申请实施例中,在第一设备得到目标AI模型之后,第一设备可以先确定是否可以应用目标AI模型,并根据不同的确定结果,执行不同的操作,以下将分别举例说明。In the embodiment of the present application, after the first device obtains the target AI model, the first device can first determine whether the target AI model can be applied, and perform different operations based on different determination results. Examples will be described below.
可选地,本申请实施例中,结合图2,如图3所示,在上述步骤102之后,本申请实施例提供的AI模型传输方法还可以包括下述的步骤201。Optionally, in an embodiment of the present application, in combination with Figure 2, as shown in Figure 3, after the above-mentioned step 102, the AI model transmission method provided in the embodiment of the present application may also include the following step 201.
步骤201、第一设备采用目标AI模型,执行第一操作。Step 201: The first device adopts the target AI model and performs the first operation.
本申请实施例中,在第一设备确定可以应用目标AI模型的情况下,第一设备可以采用目标AI模型,执行第一操作。In this embodiment of the present application, when the first device determines that the target AI model can be applied, the first device can use the target AI model to perform the first operation.
本申请实施例中,上述第一操作包括以下至少一项:In the embodiment of the present application, the above-mentioned first operation includes at least one of the following:
信号处理操作;signal processing operations;
信号传输操作;signal transmission operations;
信号解调操作;signal demodulation operation;
信道状态信息获取操作;Channel status information acquisition operation;
波束管理操作;beam management operations;
信道预测操作;Channel prediction operation;
干扰抑制操作;interference suppression operations;
定位操作;Positioning operations;
高层业务的预测和管理操作;
High-level business forecasting and management operations;
高层参数的预测和管理操作;Prediction and management of high-level parameters;
控制信令解析操作。Control signaling parsing operations.
需要说明的是,针对第一设备采用目标AI模型执行第一操作中的各个操作的描述,可以参考相关技术中的具体说明,本申请实施例在此不予赘述。It should be noted that, for the description of the first device using the target AI model to perform each operation in the first operation, reference can be made to the specific descriptions in the related art, and the embodiments of the present application will not be repeated here.
其中,上述信号处理操作包括以下至少一项:信号的检测操作、信号的滤波操作、信号的均衡操作。这里,该信号可以为以下任一项:解调参考信号(Demodulation Reference Signal,DMRS)、探测参考信号(Sounding Reference Signal,SRS)、同步信号块(Synchronization Signal Block,SSB)、跟踪参考信号(Tracking Reference Signal,TRS)、相位跟踪参考信号(Phase Tracking Reference Signal,PTRS)、信道状态指示参考信号(Channel State nformation Reference Signal,CSI-RS)等。Wherein, the above-mentioned signal processing operation includes at least one of the following: signal detection operation, signal filtering operation, and signal equalization operation. Here, the signal can be any of the following: Demodulation Reference Signal (DMRS), Sounding Reference Signal (SRS), Synchronization Signal Block (SSB), Tracking Reference Signal (Tracking Reference Signal (TRS), Phase Tracking Reference Signal (PTRS), Channel State nformation Reference Signal (CSI-RS), etc.
其中,上述信号传输操作包括以下至少一项:信道的接收操作、信道的发送操作。这里,该信道可以为以下任一项:物理下行控制信道(Physical Downlink Control Channel,PDCCH)、物理下行共享信道(Physical Downlink Share Channel,PDSCH)、物理上行控制信道(Physical Uplink Control Channel,PUCCH)、物理上行共享信道(Physical Uplink Share Channel,PUSCH)、物理随机接入信道(Physical Random Access Channel,PRACH)、物理广播信道(Physical Broadcast Channel,PBCH)。The signal transmission operation includes at least one of the following: a channel receiving operation and a channel sending operation. Here, the channel can be any one of the following: a physical downlink control channel (PDCCH), a physical downlink shared channel (PDSCH), a physical uplink control channel (PUCCH), a physical uplink shared channel (PUSCH), a physical random access channel (PRACH), and a physical broadcast channel (PBCH).
其中,信道状态信息获取操作包括以下至少一项:信道状态信息反馈、频分双工(Frequency Division Duplex,FDD)上下行部分互易性。这里,信道状态信息可以包括以下任一项:信道相关信息、信道矩阵相关信息、信道特征信息、信道矩阵特征信息、预编码矩阵指示符(Pre-coding Matrix Indicator,PMI)、秩指示(Rank Indication,RI)、CSI-RS资源指示符(CSI-RS Resource Indicator,CRI)、信道质量信息(Channel Quality Information,CQI)、层指示(Layer Indicator,LI)。Among them, the channel state information acquisition operation includes at least one of the following: channel state information feedback, frequency division duplex (Frequency Division Duplex, FDD) uplink and downlink partial reciprocity. Here, the channel state information may include any of the following: channel related information, channel matrix related information, channel characteristic information, channel matrix characteristic information, pre-coding matrix indicator (Pre-coding Matrix Indicator, PMI), rank indicator (Rank Indication) , RI), CSI-RS Resource Indicator (CRI), Channel Quality Information (Channel Quality Information, CQI), Layer Indicator (LI).
在FDD***中,基于上下行部分互易性,第二设备可以根据上行信道,获取角度和时延信息,并通过CSI-RS预编码或者直接指示的方法,将角度信息和时延信息通知第一设备,从而第一设备可以根据第二设备的指示上报或者在基站的指示范围内选择并上报,从而减少第一设备的计算量和CSI上报的开销。In the FDD system, based on the partial reciprocity of uplink and downlink, the second device can obtain the angle and delay information according to the uplink channel, and notify the first device of the angle information and delay information through CSI-RS precoding or direct indication, so that the first device can report according to the indication of the second device or select and report within the indication range of the base station, thereby reducing the calculation amount of the first device and the overhead of CSI reporting.
其中,波束管理操作可以包括以下至少一项:波束测量操作、波束上报操作、波束预测操作、波束失败检测操作、波束失败恢复操作、波束失败恢复中的新波束指示操作。The beam management operation may include at least one of the following: beam measurement operation, beam reporting operation, beam prediction operation, beam failure detection operation, beam failure recovery operation, and new beam indication operation in beam failure recovery.
其中,信道预测操作可以包括以下至少一项:信道状态信息的预测操作、波束预测操作。The channel prediction operation may include at least one of the following: a prediction operation of channel state information and a beam prediction operation.
其中,干扰抑制操作可以包括以下至少一项:小区内干扰抑制操作、小区间干扰抑制操作、带外干扰抑制操作、交调干扰抑制操作。The interference suppression operation may include at least one of the following: intra-cell interference suppression operation, inter-cell interference suppression operation, out-of-band interference suppression operation, and cross-modulation interference suppression operation.
其中,定位操作具体可以为:通过参考信号(例如SRS),估计出目标位置信息。这里,目标位置信息可以包括以下至少一项:第一设备的具***置信息(包括水平位置信息和垂直位置信息中的至少一个)、未来可能的位置信息、辅助位置估计的信息、轨迹估计的信息。The positioning operation may specifically include estimating the target position information through a reference signal (such as SRS). Here, the target location information may include at least one of the following: specific location information of the first device (including at least one of horizontal location information and vertical location information), possible future location information, information to assist location estimation, and trajectory estimation information. .
其中,高层业务的预测和管理操作和高层参数的预测和管理操作可以包括:吞吐量的预测和管理操作、所需数据包大小的预测和管理操作、业务需求的预测和管理操作、移动速度的预测和管理操作、噪声信息的预测和管理操作等。Among them, the prediction and management operations of high-level services and the prediction and management operations of high-level parameters may include: prediction and management operations of throughput, prediction and management operations of required data packet size, prediction and management operations of business needs, and movement speed. Prediction and management operations, prediction and management operations of noise information, etc.
其中,控制信令可以包括以下至少一项:功率控制的相关信令,波束管理的相关信令。The control signaling may include at least one of the following: power control related signaling, and beam management related signaling.
如此可知,由于第一设备可以采用目标AI模型,执行信号处理操作、信号传输操作、信道状态信息获取操作、波束管理操作、信道预测操作、干扰抑制操作、定位操作、高层业务的预测和管理操作、高层参数的预测和管理操作以及控制信令解析操作中的至少一个操作,因此,可以提升第一设备通信的吞吐量,且降低第一设备通信的时延。It can be seen that because the first device can adopt the target AI model to perform signal processing operations, signal transmission operations, channel state information acquisition operations, beam management operations, channel prediction operations, interference suppression operations, positioning operations, and prediction and management operations of high-level services , at least one of a high-level parameter prediction and management operation and a control signaling parsing operation. Therefore, the communication throughput of the first device can be improved and the communication delay of the first device can be reduced.
可选地,本申请实施例中,结合图2,如图4所示,在上述步骤102之后,本申请实施例提供的AI模型传输方法还可以包括下述的步骤301。Optionally, in this embodiment of the present application, as shown in Figure 4 in conjunction with Figure 2, after the above step 102, the AI model transmission method provided by the embodiment of the present application may also include the following step 301.
步骤301、在目标AI模型与第一设备的能力不匹配的情况下,第一设备不应用目标AI模型。Step 301: If the target AI model does not match the capabilities of the first device, the first device does not apply the target AI model.
本申请实施例中,在目标AI模型与第一设备的能力不匹配的情况下,第一设备确定不可以应用目标AI模型,从而第一设备不应用目标AI模型。In the embodiment of the present application, when the target AI model does not match the capabilities of the first device, the first device determines that the target AI model cannot be applied, so that the first device does not apply the target AI model.
如此可知,由于在目标AI模型与第一设备的能力不匹配的情况下,第一设备可以不应
用目标AI模型,因此,可以避免第一设备使用目标AI模型进行通信的过程中,第一设备发生卡顿的情况。It can be seen that, when the target AI model does not match the capabilities of the first device, the first device may not respond Using the target AI model, therefore, it is possible to avoid the first device being stuck in the process of using the target AI model for communication.
本申请实施例提供的AI模型传输方法,第一设备可以从第二设备接收目标AI模型的目标模型信息,并根据目标模型信息,得到目标AI模型,该目标AI模型满足以下至少一项:目标AI模型的目标传输模式,是根据第一设备支持的AI模型传输模式的能力确定的;目标AI模型的目标应用时间,是根据目标传输模式确定的;目标AI模型触发第一设备向第二设备发送目标反馈事件,该目标反馈事件用于反馈目标AI模型的应用情况;其中,上述目标传输模式包括第一传输模式和第二传输模式中的任一个;在目标传输模式为第一传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;在目标传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息。由于目标AI模型的目标传输模式,是根据第一设备支持的AI模型传输模式的能力确定的,因此,可以避免因第一设备不支持第一传输模式或第二传输模式,而导致无法根据目标模型信息得到目标AI模型的情况,进而导致无法应用目标AI模型的情况;和/或,由于目标AI模型的目标应用时间,是根据目标传输模式确定的,因此,可以避免因目标AI模型的应用时间过大(或过小),而导致无法应用目标AI模型、或浪费等待时间的情况;和/或,由于目标AI模型可以触发第一设备向第二设备发送用于反馈目标AI模型的应用情况的目标反馈事件,因此,第二设备可以直接根据目标反馈事件,向第一设备发送第一设备能够应用的AI模型,而无需第一设备与第二设备进行多次传输。从而可以减少在得到目标AI模型的过程中,第一设备与第二设备进行传输的次数,如此,可以节省传输资源。According to the AI model transmission method provided by the embodiment of the present application, the first device can receive the target model information of the target AI model from the second device, and obtain the target AI model based on the target model information. The target AI model satisfies at least one of the following: Target The target transmission mode of the AI model is determined based on the capability of the AI model transmission mode supported by the first device; the target application time of the target AI model is determined based on the target transmission mode; the target AI model triggers the first device to transmit data to the second device Send a target feedback event, which is used to feedback the application status of the target AI model; wherein the above-mentioned target transmission mode includes any one of the first transmission mode and the second transmission mode; when the target transmission mode is the first transmission mode In this case, the target model information includes model structure information and model parameter information of the target AI model; when the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model. Since the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first device, it can be avoided that the first device does not support the first transmission mode or the second transmission mode, causing the failure to perform the target transmission according to the target AI model. The model information obtains the target AI model, which leads to the inability to apply the target AI model; and/or, because the target application time of the target AI model is determined based on the target transmission mode, therefore, the application of the target AI model can be avoided. The time is too long (or too short), resulting in the inability to apply the target AI model, or a waste of waiting time; and/or, because the target AI model can trigger the first device to send an application for feedback of the target AI model to the second device Therefore, the second device can directly send the AI model that the first device can apply to the first device based on the target feedback event, without the need for the first device to perform multiple transmissions with the second device. This can reduce the number of transmissions between the first device and the second device in the process of obtaining the target AI model, thus saving transmission resources.
下面将分别以目标AI模型满足不同的条件为例,进行具体的举例说明。The following will give specific examples using the target AI model meeting different conditions as examples.
可选地,本申请实施例中,上述目标AI模型满足目标AI模型的目标传输模式,是根据第一设备支持的AI模型传输模式的能力确定的。具体地,在上述步骤102之后,本申请实施例提供的AI模型传输方法还可以包括下述的步骤401。Optionally, in this embodiment of the present application, whether the target AI model satisfies the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first device. Specifically, after the above step 102, the AI model transmission method provided by the embodiment of the present application may also include the following step 401.
步骤401、第一设备向第二设备发送目标信息。Step 401: The first device sends target information to the second device.
本申请实施例中,上述目标信息用于指示第一设备支持的AI模型传输模式的能力。In this embodiment of the present application, the above target information is used to indicate the capability of the AI model transmission mode supported by the first device.
可以理解,在第一设备向第二信息发送目标信息之后,第二设备可以根据第一设备支持的AI模型传输模式的能力,确定目标传输模式。It can be understood that after the first device sends the target information to the second information, the second device can determine the target transmission mode according to the capability of the AI model transmission mode supported by the first device.
可选地,本申请实施例中,在第一设备支持的AI模型传输模式的能力为支持第一传输模式的情况下,目标传输模式具体可以为第一传输模式;在第一设备支持的AI模型传输模式的能力为支持第二传输模式的情况下,目标传输模式具体可以为第二传输模式;在第一设备支持的AI模型传输模式的能力为支持第一传输模式和第二传输模式,目标传输模式具体可以为第一传输模式或第二传输模式。Optionally, in this embodiment of the present application, when the AI model transmission mode capability supported by the first device is to support the first transmission mode, the target transmission mode may specifically be the first transmission mode; when the AI model supported by the first device When the capability of the model transmission mode is to support the second transmission mode, the target transmission mode may specifically be the second transmission mode; when the capability of the AI model transmission mode supported by the first device is to support the first transmission mode and the second transmission mode, The target transmission mode may specifically be the first transmission mode or the second transmission mode.
如此可知,由于第一设备可以向第二设备发送目标信息,以使得第二设备可以根据第一设备支持的AI模型传输模式的能力,确定目标传输模式,因此,可以避免因第一设备不支持第一传输模式或第二传输模式,而导致无法得到目标AI模型,进而导致无法应用目标AI模型的情况。It can be seen that since the first device can send the target information to the second device, so that the second device can determine the target transmission mode according to the AI model transmission mode capability supported by the first device, therefore, it is possible to avoid the problem that the first device does not support the transmission mode. The first transmission mode or the second transmission mode results in the inability to obtain the target AI model, which in turn results in the inability to apply the target AI model.
可选地,本申请实施例中,上述目标AI模型满足目标AI模型的目标应用时间,是根据目标传输模式确定的。具体地,在上述步骤102之后,本申请实施例提供的AI模型传输方法还可以包括下述的步骤402。Optionally, in this embodiment of the present application, the target application time of the target AI model that satisfies the target AI model is determined based on the target transmission mode. Specifically, after the above-mentioned step 102, the AI model transmission method provided by the embodiment of the present application may further include the following step 402.
步骤402、第一设备根据目标传输模式,确定目标应用时间。Step 402: The first device determines a target application time according to a target transmission mode.
可选地,本申请实施例中,上述步骤402具体可以通过下述的步骤402a实现。Optionally, in the embodiment of the present application, the above step 402 can be specifically implemented by the following step 402a.
步骤402a、第一设备将目标传输模式对应的应用时间,确定为目标应用时间。Step 402a: The first device determines the application time corresponding to the target transmission mode as the target application time.
可选地,本申请实施例中,在第一设备中预存有多个对应关系,每个对应关系为一个传输模式和一个应用时间的对应关系,从而第一设备可以从多个传输模式中,确定与目标传输模式相同的一个传输模式,并将该一个传输模式对应的应用时间,确定为目标应用时间。Optionally, in this embodiment of the present application, multiple corresponding relationships are pre-stored in the first device, and each corresponding relationship is a corresponding relationship between a transmission mode and an application time, so that the first device can select from multiple transmission modes, Determine a transmission mode that is the same as the target transmission mode, and determine the application time corresponding to the transmission mode as the target application time.
本申请实施例中,第一应用时间和第二应用时间之间满足以下至少一项:In the embodiment of the present application, at least one of the following is satisfied between the first application time and the second application time:
第一应用时间大于第二应用时间;The first application time is greater than the second application time;
第一应用时间大于或等于第三应用时间,该第三应用时间是根据第二应用时间和AI模型的编译时间确定的。The first application time is greater than or equal to the third application time, and the third application time is determined based on the second application time and the compilation time of the AI model.
其中,上述第一应用时间为:第一传输模式对应的AI模型应用时间;上述第二应用时
间为:第二传输模式对应的AI模型应用时间。Among them, the above-mentioned first application time is: the AI model application time corresponding to the first transmission mode; the above-mentioned second application time Time is: the AI model application time corresponding to the second transmission mode.
其中,上述第三应用时间具体可以是根据第二应用时间和AI模型的编译时间之和。Specifically, the third application time may be the sum of the second application time and the compilation time of the AI model.
这里,AI模型的编译时间包括:第一编译时间和第二编译时间;其中,第一编译时间为:从第一设备的AI模型执行模块到第一设备的AI模型编译模块的时间,第二编译时间为:从第二设备的AI模型执行模块到第二设备的AI模型编译模块的时间。Here, the compilation time of the AI model includes: a first compilation time and a second compilation time; where the first compilation time is: the time from the AI model execution module of the first device to the AI model compilation module of the first device, and the second compilation time is: The compilation time is: the time from the AI model execution module of the second device to the AI model compilation module of the second device.
如此可知,由于第一设备可以将目标传输模式对应的应用时间确定为目标应用时间;其中,第一应用时间大于第二应用时间、和/或第一应用时间大于或等于第三应用时间,因此,可以避免因目标AI模型的应用时间过大(或过小),而导致无法应用目标AI模型、或浪费等待时间的情况。It can be seen that since the first device can determine the application time corresponding to the target transmission mode as the target application time; wherein the first application time is greater than the second application time, and/or the first application time is greater than or equal to the third application time, therefore , which can avoid situations where the target AI model cannot be applied or waiting time is wasted due to the application time of the target AI model being too large (or too small).
可选地,本申请实施例中,上述步骤402具体可以通过下述的步骤402b实现。Optionally, in this embodiment of the present application, the above step 402 can be specifically implemented through the following step 402b.
步骤402b、在目标传输模式为第一传输模式的情况下,若目标AI模型和第一设备中的预存AI模型相匹配,则第一设备将第四应用时间确定为目标应用时间。Step 402b: When the target transmission mode is the first transmission mode, if the target AI model matches the pre-stored AI model in the first device, the first device determines the fourth application time as the target application time.
本申请实施例中,上述第四应用时间小于第一应用时间;该第一应用时间为:第一传输模式对应的AI模型应用时间。In the embodiment of the present application, the above-mentioned fourth application time is less than the first application time; the first application time is: the AI model application time corresponding to the first transmission mode.
需要说明的是,上述“目标AI模型和预存AI模型相匹配”可以理解为:目标AI模型和预存AI模型相同,或者,目标AI模型和预存AI模型的模型结构的相似度大于或等于预设门限。It should be noted that the above "target AI model and pre-stored AI model match" can be understood as: the target AI model and the pre-stored AI model are the same, or the similarity of the model structures of the target AI model and the pre-stored AI model is greater than or equal to the preset Threshold.
其中,目标AI模型和预存AI模型相同具体可以为:目标AI模型和预存AI模型的模型层数相同,且目标AI模型和预存AI模型的每层神经元类型相同。Specifically, the target AI model and the pre-stored AI model are the same: the target AI model and the pre-stored AI model have the same number of model layers, and the types of neurons in each layer of the target AI model and the pre-stored AI model are the same.
其中,目标AI模型和预存AI模型的模型结构的相似度大于或等于预设门限具体可以为以下任一项:预存AI模型的模型结构是目标AI模型的模型结构的子集或子模型、预存AI模型的模型结构包括了目标AI模型的模型结构、预存AI模型的模型结构包括了目标AI模型的模型结构的子集或子模型。Among them, the similarity between the model structures of the target AI model and the pre-stored AI model is greater than or equal to a preset threshold, which can be any of the following: the model structure of the pre-stored AI model is a subset or sub-model of the model structure of the target AI model, the model structure of the pre-stored AI model includes the model structure of the target AI model, and the model structure of the pre-stored AI model includes a subset or sub-model of the model structure of the target AI model.
如此可知,由于在目标AI模型和预存AI模型相匹配的情况下,第一设备可以将较小的应用时间,确定为目标应用时间,因此,可以减少应用目标AI模型的时延。It can be seen that, since the first device can determine a smaller application time as the target application time when the target AI model matches the pre-stored AI model, the delay in applying the target AI model can be reduced.
可选地,本申请实施例中,上述目标AI模型满足目标AI模型触发第一设备向第二设备发送目标反馈事件。其中,上述目标反馈事件为以下任一项:第一反馈事件、第二反馈事件;该第一反馈事件用于反馈未应用目标AI模型;该第二反馈事件用于反馈已应用目标AI模型。Optionally, in this embodiment of the present application, the above target AI model satisfies the target AI model and triggers the first device to send a target feedback event to the second device. Wherein, the above-mentioned target feedback event is any one of the following: a first feedback event, a second feedback event; the first feedback event is used to feedback the unapplied target AI model; the second feedback event is used to feedback the applied target AI model.
其中,上述第一反馈事件为以下任一项:Among them, the above-mentioned first feedback event is any of the following:
第一设备未应用目标AI模型;The first device does not apply the target AI model;
第一设备未应用目标AI模型,且第一设备仍应用第一AI模型;The first device does not apply the target AI model, and the first device still applies the first AI model;
第一设备未应用目标AI模型,且第一设备不应用AI模型;The first device does not apply the target AI model, and the first device does not apply the AI model;
第一设备不支持第二设备通过目标传输模式发送的目标模型信息;The first device does not support the target model information sent by the second device through the target transmission mode;
第三反馈事件;Third feedback event;
第一设备不支持目标AI模型的第一模型信息。The first device does not support the first model information of the target AI model.
这里,第一设备未应用目标AI模型可以理解为:目标AI模型未能激活,和/或未能生效,和/或未能切换。Here, the fact that the first device does not apply the target AI model can be understood as: the target AI model fails to be activated, and/or fails to take effect, and/or fails to switch.
这里,第一AI模型为:第一设备接收目标AI模型之前应用的AI模型。可以理解,第一设备仍应用旧的A模型。Here, the first AI model is: the AI model applied before the first device receives the target AI model. Understandably, the first device still uses the old A model.
这里,第一设备不应用AI模型可以理解为:第一设备使用非AI算法执行操作(例如第一操作)。Here, the first device not applying the AI model can be understood as: the first device uses a non-AI algorithm to perform the operation (eg, the first operation).
这里,第一设备不支持第二设备通过目标传输模式发送的目标模型信息具体可以为:第一设备不支持第二设备通过第一传输模式发送的目标模型信息。可以理解,第一设备不支持第二设备发送的目标AI模型的模型结构信息指示的模型结构,目标传输模式为第一传输模式。Here, the first device does not support the target model information sent by the second device through the target transmission mode, which can be specifically: the first device does not support the target model information sent by the second device through the first transmission mode. It can be understood that the first device does not support the model structure indicated by the model structure information of the target AI model sent by the second device, and the target transmission mode is the first transmission mode.
这里,上述第三反馈事件为:第一设备不支持第二设备通过目标传输模式发送的目标模型信息;该第三反馈事件携带第一信息,该第一信息为目标AI模型中的第一设备不支持的模型结构的信息。可以理解,第一设备可以向第二设备反馈目标AI模型中的第一设备不支持的模型结构的信息,目标传输模式为第一传输模式。Here, the third feedback event is: the first device does not support the target model information sent by the second device through the target transmission mode; the third feedback event carries the first information, and the first information is the information of the model structure in the target AI model that the first device does not support. It can be understood that the first device can feedback to the second device the information of the model structure in the target AI model that the first device does not support, and the target transmission mode is the first transmission mode.
示例性地,假设目标AI模型为神经网络模型,该神经网络模型的模型结构信息指示的
模型结构为5层全连接层、2层卷积层以及1层长短期记忆网络(Long-Short-Term Memory,LSTM)层,第一设备不支持LSTM层,则第三反馈事件为第一设备不支持第二设备通过第一传输模式发送神经网络模型的模型结构信息,该第三反馈事件携带第一信息,该第一信息为LSTM层的信息。For example, assuming that the target AI model is a neural network model, the model structure information of the neural network model indicates The model structure is 5 fully connected layers, 2 convolutional layers and 1 Long-Short-Term Memory (LSTM) layer. The first device does not support the LSTM layer, so the third feedback event is the first device The second device is not supported to send the model structure information of the neural network model through the first transmission mode. The third feedback event carries the first information, and the first information is the information of the LSTM layer.
这里,上述第一模型信息包括以下至少一项:模型大小、模型复杂度、模型操作数(例如浮点运算次数(Floating-point operations,FLOP))。Here, the above-mentioned first model information includes at least one of the following: model size, model complexity, and model operands (such as floating-point operations (Floating-point operations, FLOP)).
其中,上述第二反馈事件为以下任一项:Among them, the above-mentioned second feedback event is any of the following:
第一设备已应用目标AI模型;The first device has the target AI model applied;
第四反馈事件;The fourth feedback event;
第一设备已应用目标AI模型,且已替换第一AI模型;The first device has applied the target AI model and has replaced the first AI model;
第一设备支持第二设备通过目标传输模式发送的目标模型信息;The first device supports the target model information sent by the second device through the target transmission mode;
第一设备已编译目标AI模型。The first device has compiled the target AI model.
这里,第一设备已应用目标AI模型可以理解为:目标AI模型已经激活,和/或已经生效,和/或已经切换。Here, the first device has applied the target AI model can be understood as: the target AI model has been activated, and/or has taken effect, and/or has been switched.
这里,上述第四反馈事件为:第一设备已应用目标AI模型;该第四反馈事件携带第二信息,该第二信息为目标AI模型的模型标识信息。可以理解,目标传输模式为第一传输模式。其中,模型标识信息具体可以为:模型身份标识(Identity,ID)。Here, the above-mentioned fourth feedback event is: the first device has applied the target AI model; the fourth feedback event carries second information, and the second information is model identification information of the target AI model. It can be understood that the target transmission mode is the first transmission mode. The model identification information may specifically be: model identity (Identity, ID).
这里,上述第一AI模型为:第一设备接收目标AI模型之前应用的AI模型。Here, the above-mentioned first AI model is: the AI model applied before the first device receives the target AI model.
这里,第一设备支持第二设备通过目标传输模式发送的目标模型信息可以理解为:第一设备支持第二设备通过第一传输模式发送的目标AI模型的模型结构信息指示的模型结构。Here, the first device supporting the target model information sent by the second device through the target transmission mode can be understood as: the first device supports the model structure indicated by the model structure information of the target AI model sent by the second device through the first transmission mode.
这里,第一设备已编译目标AI模型可以理解为:第一设备已编译好第二设备通过第一传输模式发送的目标AI模型的模型结构信息。从而在第一设备需求其他AI模型时,第二设备可以通过第二传输模式,传输其他AI模型的模型参数信息,以使得第一设备可以将该其他AI模型的模型参数信息,应用于目标AI模型的模型结构信息指示的模型结构,以得到该其他AI模型。Here, the first device has compiled the target AI model can be understood as: the first device has compiled the model structure information of the target AI model sent by the second device through the first transmission mode. Therefore, when the first device requires other AI models, the second device can transmit the model parameter information of the other AI models through the second transmission mode, so that the first device can apply the model parameter information of the other AI models to the target AI. The model structure information of the model indicates the model structure to obtain the other AI model.
可以理解,后续若第二设备通过第二传输模式传输的模型参数信息,则第一设备可以认为第二设备要传输的AI模型的模型结构信息,与目标AI模型的模型结构信息相同。It can be understood that if the second device subsequently transmits model parameter information through the second transmission mode, the first device may consider that the model structure information of the AI model to be transmitted by the second device is the same as the model structure information of the target AI model.
当然,第一设备也可以向第二设备传输AI模型的模型信息,以使得第二设备可以按照第一设备传输的AI模型信息,向第一设备传输目标模型信息,以下将举例说明。Of course, the first device can also transmit model information of the AI model to the second device, so that the second device can transmit target model information to the first device according to the AI model information transmitted by the first device. An example will be given below.
可选地,本申请实施例中,结合图2,如图5所示,在上述步骤101之前,本申请实施例提供的AI模型传输方法还可以包括下述的步骤501。Optionally, in this embodiment of the present application, as shown in Figure 5 in conjunction with Figure 2, before the above step 101, the AI model transmission method provided by the embodiment of the present application may also include the following step 501.
步骤501、第一设备按照第三传输模式,向第二设备发送第二AI模型的第二模型信息。Step 501: The first device sends the second model information of the second AI model to the second device according to the third transmission mode.
本申请实施例中,上述第三传输模式为第一传输模式或第四传输模式;在目标传输模式为第一传输模式的情况下,第二模型信息包括第二AI模型的模型结构信息和模型参数信息;在目标传输模式为第四传输模式的情况下,第二模型信息包括第二AI模型的模型结构信息;上述目标AI模型与第二AI模型的模型结构信息相匹配。In the embodiment of the present application, the above-mentioned third transmission mode is the first transmission mode or the fourth transmission mode; when the target transmission mode is the first transmission mode, the second model information includes the model structure information and the model of the second AI model. Parameter information; when the target transmission mode is the fourth transmission mode, the second model information includes model structure information of the second AI model; the above target AI model matches the model structure information of the second AI model.
其中,第四传输模式为:传输AI模型的模型结构信息的传输模式。Among them, the fourth transmission mode is: a transmission mode for transmitting model structure information of the AI model.
需要说明的是,上述“目标AI模型与第二AI模型的模型结构信息相匹配”可以理解为:目标AI模型和第二AI模型的模型结构信息相同,或者,目标AI模型和第二AI模型的模型结构信息的相似度大于或等于预设门限。It should be noted that the above "the target AI model matches the model structure information of the second AI model" can be understood as: the target AI model and the second AI model have the same model structure information, or the target AI model and the second AI model The similarity of the model structure information is greater than or equal to the preset threshold.
其中,目标AI模型和第二AI模型的模型结构信息相同具体可以为:目标AI模型和第二AI模型的模型层数相同,且目标AI模型和第二AI模型的每层神经元类型相同。Specifically, the model structure information of the target AI model and the second AI model is the same: the target AI model and the second AI model have the same number of model layers, and the types of neurons in each layer of the target AI model and the second AI model are the same.
其中,目标AI模型和第二AI模型的模型结构信息的相似度大于或等于预设门限具体可以为以下任一项:Among them, the similarity between the model structure information of the target AI model and the second AI model is greater than or equal to the preset threshold, which can be any of the following:
第二AI模型的模型结构信息指示的模型结构是目标AI模型的模型结构信息指示的模型结构的子集或子模型;The model structure indicated by the model structure information of the second AI model is a subset or sub-model of the model structure indicated by the model structure information of the target AI model;
第二AI模型的模型结构信息指示的模型结构包括了目标AI模型的模型结构信息指示的模型结构;The model structure indicated by the model structure information of the second AI model includes the model structure indicated by the model structure information of the target AI model;
第二AI模型的模型结构信息指示的模型结构包括了目标AI模型的模型结构信息指示的模型结构的子集或子模型。The model structure indicated by the model structure information of the second AI model includes a subset or sub-model of the model structure indicated by the model structure information of the target AI model.
如此可知,由于目标AI模型和第二AI模型的模型结构信息相匹配,因此,第一设备
可以直接从第二设备接收目标AI模型的模型参数信息,以得到目标AI模型,而无需重复接收目标AI模型的模型结构信息,如此,可以节省传输资源。It can be seen that since the model structure information of the target AI model and the second AI model match, therefore, the first device The model parameter information of the target AI model can be directly received from the second device to obtain the target AI model without repeatedly receiving the model structure information of the target AI model. In this way, transmission resources can be saved.
图6示出了本申请实施例提供的一种AI模型传输方法的流程图。如图6所示,本申请实施例提供的AI模型传输方法可以包括下述的步骤601。Figure 6 shows a flow chart of an AI model transmission method provided by an embodiment of the present application. As shown in Figure 6, the AI model transmission method provided by the embodiment of the present application may include the following step 601.
步骤601、第二设备向第一设备发送目标AI模型的目标模型信息。Step 601: The second device sends target model information of the target AI model to the first device.
本申请实施例中,上述目标模型信息用于第一设备得到目标AI模型。In an embodiment of the present application, the above-mentioned target model information is used by the first device to obtain the target AI model.
本申请实施例中,上述目标AI模型满足以下至少一项:In the embodiment of this application, the above target AI model satisfies at least one of the following:
目标AI模型的目标传输模式,是根据第一设备支持的AI模型传输模式的能力确定的;The target transmission mode of the target AI model is determined based on the capabilities of the AI model transmission mode supported by the first device;
目标AI模型的目标应用时间,是根据目标传输模式确定的;The target application time of the target AI model is determined based on the target transmission mode;
目标AI模型触发第一设备向第二设备发送目标反馈事件,该目标反馈事件用于反馈目标AI模型的应用情况。The target AI model triggers the first device to send a target feedback event to the second device, and the target feedback event is used to feed back the application status of the target AI model.
其中,上述目标传输模式包括以下任一项:第一传输模式、第二传输模式;在目标传输模式为第一传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;在目标传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息。Wherein, the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
可以理解,第一传输模式为:传输AI模型的模型结构信息和模型参数信息的传输模式;该第二传输模式为:传输AI模型的模型参数信息的传输模式。It can be understood that the first transmission mode is: a transmission mode for transmitting model structure information and model parameter information of the AI model; the second transmission mode is: a transmission mode for transmitting model parameter information of the AI model.
本申请实施例提供的AI模型传输方法,第二设备可以向第一设备发送目标AI模型的目标模型信息,该目标AI模型满足以下至少一项:目标AI模型的目标传输模式,是根据第一设备支持的AI模型传输模式的能力确定的;目标AI模型的目标应用时间,是根据目标传输模式确定的;目标AI模型触发第一设备向第二设备发送目标反馈事件,该目标反馈事件用于反馈目标AI模型的应用情况;其中,上述目标传输模式包括第一传输模式和第二传输模式中的任一个;在目标传输模式为第一传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;在目标传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息。由于目标AI模型的目标传输模式,是根据第一设备支持的AI模型传输模式的能力确定的,因此,可以避免因第一设备不支持第一传输模式或第二传输模式,而导致无法根据目标模型信息得到目标AI模型的情况,进而导致无法应用目标AI模型的情况;和/或,由于目标AI模型的目标应用时间,是根据目标传输模式确定的,因此,可以避免因目标AI模型的应用时间过大(或过小),而导致无法应用目标AI模型、或浪费等待时间的情况;和/或,由于目标AI模型可以触发第一设备向第二设备发送用于反馈目标AI模型的应用情况的目标反馈事件,因此,第二设备可以直接根据目标反馈事件,向第一设备发送第一设备能够应用的AI模型,而无需第一设备与第二设备进行多次传输。从而可以减少在得到目标AI模型的过程中,第一设备与第二设备进行传输的次数,如此,可以节省传输资源。In the AI model transmission method provided in the embodiment of the present application, the second device can send target model information of the target AI model to the first device, and the target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined according to the capability of the AI model transmission mode supported by the first device; the target application time of the target AI model is determined according to the target transmission mode; the target AI model triggers the first device to send a target feedback event to the second device, and the target feedback event is used to feedback the application status of the target AI model; wherein the above-mentioned target transmission mode includes any one of the first transmission mode and the second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model; when the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model. Since the target transmission mode of the target AI model is determined according to the capability of the AI model transmission mode supported by the first device, it is possible to avoid the situation where the target AI model cannot be obtained according to the target model information due to the first device not supporting the first transmission mode or the second transmission mode, and thus the target AI model cannot be applied; and/or, since the target application time of the target AI model is determined according to the target transmission mode, it is possible to avoid the situation where the target AI model cannot be applied or the waiting time is wasted due to the application time of the target AI model being too long (or too short); and/or, since the target AI model can trigger the first device to send a target feedback event for feedback on the application of the target AI model to the second device, the second device can directly send the AI model that the first device can apply to the first device according to the target feedback event, without the need for the first device and the second device to perform multiple transmissions. Thereby, the number of transmissions between the first device and the second device in the process of obtaining the target AI model can be reduced, so that transmission resources can be saved.
可选地,本申请实施例中,上述目标AI模型满足目标AI模型的目标传输模式,是根据第一设备支持的AI模型传输模式的能力确定的。具体地,在上述步骤601之后,本申请实施例提供的AI模型传输方法还可以包括下述的步骤701至步骤703。Optionally, in this embodiment of the present application, whether the target AI model satisfies the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first device. Specifically, after the above step 601, the AI model transmission method provided by the embodiment of the present application may also include the following steps 701 to 703.
步骤701、第二设备从第一设备接收目标信息。Step 701: The second device receives target information from the first device.
本申请实施例中,上述目标信息用于指示第一设备支持的AI模型传输模式的能力。In this embodiment of the present application, the above target information is used to indicate the capability of the AI model transmission mode supported by the first device.
步骤702、第二设备根据目标信息,确定目标传输模式。Step 702: The second device determines the target transmission mode according to the target information.
可选地,本申请实施例中,在第一设备支持的AI模型传输模式的能力为支持第一传输模式的情况下,第二设备可以将第一传输模式确定为目标传输模式;在第一设备支持的AI模型传输模式的能力为支持第二传输模式的情况下,第二设备可以将第二传输模式确定为目标传输模式;在第一设备支持的AI模型传输模式的能力为支持第一传输模式和第二传输模式,第二设备可以将第一传输模式或第二传输模式确定为目标传输模式。Optionally, in an embodiment of the present application, when the capability of the AI model transmission mode supported by the first device is to support the first transmission mode, the second device may determine the first transmission mode as the target transmission mode; when the capability of the AI model transmission mode supported by the first device is to support the second transmission mode, the second device may determine the second transmission mode as the target transmission mode; when the capability of the AI model transmission mode supported by the first device is to support the first transmission mode and the second transmission mode, the second device may determine the first transmission mode or the second transmission mode as the target transmission mode.
步骤703、第二设备基于目标传输模式,向第一设备发送目标模型信息。Step 703: The second device sends target model information to the first device based on the target transmission mode.
可选地,本申请实施例中,在目标传输模式为第一传输模式的情况下,第二设备可以向第一设备发送目标AI模型的模型结构信息和模型参数信息,以使得第一设备可以得到目标AI模型;在目标传输模式为第二传输模式的情况下,第二设备可以向第一设备发送目标AI模型的模型参数信息,以使得第一设备可以得到目标AI模型。Optionally, in this embodiment of the present application, when the target transmission mode is the first transmission mode, the second device can send the model structure information and model parameter information of the target AI model to the first device, so that the first device can Obtain the target AI model; when the target transmission mode is the second transmission mode, the second device can send model parameter information of the target AI model to the first device, so that the first device can obtain the target AI model.
如此可知,由于第二设备可以从第一设备接收目标信息,并根据目标信息,确定目标
传输模式,以按照目标传输模式向第一设备发送目标模型信息,因此,可以避免因第一设备不支持第一传输模式或第二传输模式,而导致第一设备无法得到目标AI模型的情况,进而导致无法应用目标AI模型的情况。It can be seen that since the second device can receive target information from the first device and determine the target based on the target information, The transmission mode is to send the target model information to the first device according to the target transmission mode. Therefore, it is possible to avoid the situation where the first device cannot obtain the target AI model because the first device does not support the first transmission mode or the second transmission mode. This leads to a situation where the target AI model cannot be applied.
可选地,本申请实施例中,上述目标传输模式为第一传输模式。具体地,在上述步骤601之后,本申请实施例提供的AI模型传输方法还可以包括下述的步骤801和步骤802。Optionally, in an embodiment of the present application, the target transmission mode is the first transmission mode. Specifically, after the above step 601, the AI model transmission method provided in the embodiment of the present application may further include the following steps 801 and 802.
步骤801、第二设备确定目标AI模型和第一设备中的预存AI模型是否相匹配。Step 801: The second device determines whether the target AI model matches the pre-stored AI model in the first device.
可选地,本申请实施例中,第二设备可以从第一设备接收第一设备中的预存AI模型的模型标识信息,并根据该模型标识信息,确定预存AI模型,从而第二设备可以确定目标AI模型和预存AI模型是否相匹配。Optionally, in this embodiment of the present application, the second device can receive the model identification information of the pre-stored AI model in the first device from the first device, and determine the pre-stored AI model based on the model identification information, so that the second device can determine Whether the target AI model matches the pre-stored AI model.
步骤802、若目标AI模型和预存AI模型相匹配,则第二设备按照第四传输模式,向第一设备发送目标模型信息。Step 802: If the target AI model matches the pre-stored AI model, the second device sends the target model information to the first device according to the fourth transmission mode.
本申请实施例中,上述第四传输模式为以下任一项:第二传输模式、第五传输模式;在第四传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息;在第四传输模式为第五传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;该第五传输模式对应的AI模型应用时间小于第一传输模式对应的AI模型应用时间In the embodiment of the present application, the above-mentioned fourth transmission mode is any one of the following: the second transmission mode, the fifth transmission mode; when the fourth transmission mode is the second transmission mode, the target model information includes the model of the target AI model Parameter information; when the fourth transmission mode is the fifth transmission mode, the target model information includes model structure information and model parameter information of the target AI model; the AI model application time corresponding to the fifth transmission mode is shorter than that corresponding to the first transmission mode AI model application time
可以理解,第二传输模式为:传输AI模型的模型参数信息的传输模式;该第五传输模式为:传输AI模型的模型结构信息和模型参数信息的传输模式。该第五传输模式可以理解为简化的第一传输模式。It can be understood that the second transmission mode is a transmission mode for transmitting model parameter information of the AI model, and the fifth transmission mode is a transmission mode for transmitting model structure information and model parameter information of the AI model. The fifth transmission mode can be understood as a simplified first transmission mode.
如此可知,由于在目标AI模型和第一设备中的预存AI模型相匹配的情况下,第二设备可以按照第二传输模式向第一设备发送目标模型信息,即向第一设备发送目标AI模型的模型参数信息,而无需重复发送目标AI模型的模型结构信息,因此,可以节省传输资源;或者,第二设备可以按照第五传输模式向第一设备发送目标模型信息,以使得第一设备可以将较小的应用时间,确定为目标应用时间,因此,可以减少应用目标AI模型的时延。It can be seen that, when the target AI model matches the pre-stored AI model in the first device, the second device can send the target model information to the first device according to the second transmission mode, that is, send the target AI model to the first device. model parameter information of the target AI model without repeatedly sending the model structure information of the target AI model, thus saving transmission resources; or, the second device can send the target model information to the first device according to the fifth transmission mode, so that the first device can The smaller application time is determined as the target application time, so the delay in applying the target AI model can be reduced.
可选地,本申请实施例中,在上述步骤601之后,本申请实施例提供的AI模型传输方法还可以包括下述的步骤803。Optionally, in the embodiment of the present application, after the above step 601, the AI model transmission method provided by the embodiment of the present application may further include the following step 803.
步骤803、第二设备从第一设备接收目标反馈事件。Step 803: The second device receives the target feedback event from the first device.
本申请实施例中,上述目标反馈事件用于反馈目标AI模型的应用情况。In the embodiment of the present application, the above-mentioned target feedback event is used to feedback the application status of the target AI model.
需要说明的是,针对目标反馈事件的说明,可以参考上述实施例中的具体描述,本申请实施例在此不再赘述。It should be noted that for the description of the target feedback event, reference may be made to the specific description in the above embodiments, and the embodiments of the present application will not be repeated here.
如此可知,由于第二设备可以从第一设备接收用于反馈目标AI模型的应用情况的目标反馈事件,因此,第二设备可以直接根据目标反馈事件,向第一设备发送第一设备能够应用的AI模型的模型信息,而无需第一设备与第二设备进行多次传输。It can be seen that since the second device can receive a target feedback event for feedback on the application of the target AI model from the first device, the second device can directly send the first device based on the target feedback event to the first device. Model information of the AI model without the need for multiple transmissions between the first device and the second device.
可选地,本申请实施例中,在上述步骤601之前,本申请实施例提供的AI模型传输方法还可以包括下述的步骤901和步骤902。Optionally, in the embodiment of the present application, before the above step 601, the AI model transmission method provided by the embodiment of the present application may also include the following steps 901 and 902.
步骤901、第二设备从第一设备接收第二AI模型的第二模型信息。Step 901: The second device receives second model information of the second AI model from the first device.
本申请实施例中,上述第二模型信息是第一设备按照第三传输模式发送的。In this embodiment of the present application, the above-mentioned second model information is sent by the first device according to the third transmission mode.
本申请实施例中,第三传输模式为所述第一传输模式或第四传输模式;在目标传输模式为第一传输模式的情况下,第二模型信息包括第二AI模型的模型结构信息和模型参数信息;在目标传输模式为所述第四传输模式的情况下,第二模型信息包括第二AI模型的模型结构信息。In the embodiment of the present application, the third transmission mode is the first transmission mode or the fourth transmission mode; when the target transmission mode is the first transmission mode, the second model information includes the model structure information of the second AI model and Model parameter information; when the target transmission mode is the fourth transmission mode, the second model information includes model structure information of the second AI model.
可以理解,第四传输模式为:传输AI模型的模型结构信息的传输模式。It can be understood that the fourth transmission mode is: a transmission mode for transmitting model structure information of the AI model.
步骤902、第二设备将与第二AI模型的模型结构信息相匹配的模型结构信息,确定为目标AI模型的模型结构信息。Step 902: The second device determines the model structure information that matches the model structure information of the second AI model as the model structure information of the target AI model.
可选地,本申请实施例中,上述步骤902可以通过下述的步骤902a或步骤902b实现。Optionally, in this embodiment of the present application, the above step 902 can be implemented through the following step 902a or step 902b.
步骤902a、第二设备将与第二AI模型的模型结构信息相同的模型结构,确定为目标AI模型的模型结构信息。Step 902a: The second device determines the model structure that is the same as the model structure information of the second AI model as the model structure information of the target AI model.
其中,目标AI模型和第二AI模型的结构信息相同具体可以为:目标AI模型和第二AI模型的模型层数相同,且目标AI模型和第二AI模型的每层神经元类型相同。Specifically, the structural information of the target AI model and the second AI model is the same: the target AI model and the second AI model have the same number of model layers, and the types of neurons in each layer of the target AI model and the second AI model are the same.
步骤902b、第二设备将与第二AI模型的模型结构信息的相似度大于或等于预设门限的模型结构,确定为目标AI模型的模型结构信息。
Step 902b: The second device determines the model structure whose similarity to the model structure information of the second AI model is greater than or equal to the preset threshold as the model structure information of the target AI model.
其中,目标AI模型和第二AI模型的模型结构信息的相似度大于或等于预设门限具体可以为以下任一项:Among them, the similarity between the model structure information of the target AI model and the second AI model is greater than or equal to the preset threshold, which can be any of the following:
第二AI模型的模型结构信息指示的模型结构是目标AI模型的模型结构信息指示的模型结构的子集或子模型;The model structure indicated by the model structure information of the second AI model is a subset or sub-model of the model structure indicated by the model structure information of the target AI model;
第二AI模型的模型结构信息指示的模型结构包括了目标AI模型的模型结构信息指示的模型结构;The model structure indicated by the model structure information of the second AI model includes the model structure indicated by the model structure information of the target AI model;
预存AI模型的模型结构包括了目标AI模型的模型结构信息指示的模型结构的子集或子模型。The model structure of the pre-stored AI model includes a subset or sub-model of the model structure indicated by the model structure information of the target AI model.
示例性地,假设第二AI模型的模型结构信息指示的模型结构为5层全连接层和2层卷积层,目标AI模型的模型结构信息指示的模型结构为4层权连接层和2层卷积层,即第二AI模型的模型结构信息指示的模型结构是目标AI模型的模型结构信息指示的模型结构的子集,则可以认为目标AI模型和第二AI模型的模型结构信息的相似度大于或等于预设门限,即目标AI模型和第二AI模型相匹配。For example, assume that the model structure indicated by the model structure information of the second AI model is 5 fully connected layers and 2 convolutional layers, and the model structure indicated by the model structure information of the target AI model is 4 weighted connection layers and 2 layers. The convolutional layer, that is, the model structure indicated by the model structure information of the second AI model is a subset of the model structure indicated by the model structure information of the target AI model, then it can be considered that the model structure information of the target AI model and the second AI model are similar. The degree is greater than or equal to the preset threshold, that is, the target AI model and the second AI model match.
又示例性地,假设第二AI模型的模型结构信息指示的模型结构为5层全连接层和2层卷积层,目标AI模型的模型结构信息指示的模型结构为4层权连接层和,即第二AI模型的模型结构信息指示的模型结构是目标AI模型的模型结构信息指示的模型结构的子模型,则可以认为目标AI模型和第二AI模型的模型结构信息的相似度大于或等于预设门限,即目标AI模型和第二AI模型相匹配。As another example, assuming that the model structure indicated by the model structure information of the second AI model is 5 fully connected layers and 2 convolutional layers, and the model structure indicated by the model structure information of the target AI model is 4 weighted connection layers and, that is, the model structure indicated by the model structure information of the second AI model is a sub-model of the model structure indicated by the model structure information of the target AI model, then it can be considered that the similarity between the model structure information of the target AI model and the second AI model is greater than or equal to the preset threshold, that is, the target AI model and the second AI model match.
又示例性地,假设第二AI模型的模型结构信息指示的模型结构为6层全连接层和3层卷积层,目标AI模型的模型结构信息指示的模型结构为5层连接层和2层卷积层,即第二AI模型的模型结构信息指示的模型结构包括了目标AI模型的模型结构信息指示的模型结构,则可以认为目标AI模型和第二AI模型的模型结构信息的相似度大于或等于预设门限,即目标AI模型和第二AI模型相匹配。As another example, assume that the model structure indicated by the model structure information of the second AI model is 6 fully connected layers and 3 convolutional layers, and the model structure indicated by the model structure information of the target AI model is 5 connected layers and 2 layers. The convolution layer, that is, the model structure indicated by the model structure information of the second AI model includes the model structure indicated by the model structure information of the target AI model, then it can be considered that the similarity between the model structure information of the target AI model and the second AI model is greater than Or equal to the preset threshold, that is, the target AI model matches the second AI model.
又示例性地,假设第二AI模型的模型结构信息指示的模型结构为5层全连接层、2层卷积层以及1层LSTM层,目标AI模型的模型结构信息指示的模型结构为5层全连接层和2层卷积层,即第二AI模型的模型结构信息指示的模型结构包括了目标AI模型的模型结构信息指示的模型结构,则可以认为目标AI模型和第二AI模型的模型结构信息的相似度大于或等于预设门限,即目标AI模型和第二AI模型相匹配。As another example, assume that the model structure indicated by the model structure information of the second AI model is 5 layers of fully connected layers, 2 layers of convolutional layers, and 1 layer of LSTM, and the model structure indicated by the model structure information of the target AI model is 5 layers. The fully connected layer and the 2-layer convolution layer, that is, the model structure indicated by the model structure information of the second AI model includes the model structure indicated by the model structure information of the target AI model, then it can be considered that the model of the target AI model and the second AI model The similarity of the structural information is greater than or equal to the preset threshold, that is, the target AI model and the second AI model match.
如此可知,由于目标AI模型和第二AI模型的模型结构信息相匹配,因此,第一设备可以直接从第二设备接收目标AI模型的模型参数信息,以得到目标AI模型,而无需重复接收目标AI模型的模型结构信息,如此,可以节省传输资源。It can be seen that since the model structure information of the target AI model and the second AI model match, the first device can directly receive the model parameter information of the target AI model from the second device to obtain the target AI model without repeatedly receiving the target. The model structure information of the AI model can save transmission resources.
本申请实施例提供的AI模型传输方法,执行主体可以为AI模型传输装置。本申请实施例中以AI模型传输装置执行AI模型传输方法为例,说明本申请实施例提供的AI模型传输装置的。For the AI model transmission method provided by the embodiments of this application, the execution subject may be an AI model transmission device. In the embodiment of the present application, the AI model transmission method performed by the AI model transmission device is used as an example to illustrate the AI model transmission device provided by the embodiment of the present application.
图7示出了本申请实施例中涉及的AI模型传输装置的一种可能的结构示意图,该AI模型传输装置为第一AI模型传输装置。如图7所示,该第一AI模型传输装置60可以包括:接收模块61和处理模块62。Figure 7 shows a possible structural schematic diagram of the AI model transmission device involved in the embodiment of the present application. The AI model transmission device is the first AI model transmission device. As shown in FIG. 7 , the first AI model transmission device 60 may include: a receiving module 61 and a processing module 62 .
其中,接收模块61,用于从第二AI模型传输装置接收目标AI模型的目标模型信息。处理模块62,用于根据接收模块61接收的目标模型信息,得到目标AI模型。该目标AI模型满足以下至少一项:目标AI模型的目标传输模式,是根据第一AI模型传输装置60支持的AI模型传输模式的能力确定的;目标AI模型的目标应用时间,是根据目标传输模式确定的;目标AI模型触发第一AI模型传输装置60向第二AI模型传输装置发送目标反馈事件,该目标反馈事件用于反馈目标AI模型的应用情况。其中,上述目标传输模式包括以下任一项:第一传输模式、第二传输模式;在目标传输模式为第一传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;在目标传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息。Among them, the receiving module 61 is used to receive the target model information of the target AI model from the second AI model transmission device. The processing module 62 is used to obtain the target AI model according to the target model information received by the receiving module 61 . The target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device 60; the target application time of the target AI model is determined based on the target transmission The mode is determined; the target AI model triggers the first AI model transmission device 60 to send a target feedback event to the second AI model transmission device, and the target feedback event is used to feed back the application status of the target AI model. Wherein, the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
在一种可能的实现方式中,上述目标AI模型满足目标AI模型的目标传输模式,是根据第一AI模型传输装置60支持的AI模型传输模式的能力确定的;上述第一AI模型传输装置60还包括:发送模块。其中,发送模块,用于向第二AI模型传输装置发送目标信息,该目标信息用于指示第一AI模型传输装置60支持的AI模型传输模式的能力。In a possible implementation, the target AI model satisfying the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device 60; the first AI model transmission device 60 Also included: Send module. The sending module is configured to send target information to the second AI model transmission device, where the target information is used to indicate the capabilities of the AI model transmission mode supported by the first AI model transmission device 60 .
在一种可能的实现方式中,上述第一AI模型传输装置60支持的AI模型传输模式的能
力包括以下任一项:支持第一传输模式;支持第二传输模式;支持第一传输模式和第二传输模式。In a possible implementation, the AI model transmission mode supported by the first AI model transmission device 60 can The power includes any of the following: supporting the first transmission mode; supporting the second transmission mode; supporting the first transmission mode and the second transmission mode.
在一种可能的实现方式中,上述目标AI模型满足目标AI模型的目标应用时间,是根据目标传输模式确定的。本申请实施例提供的第一AI模型传输装置60还可以包括:确定模块。其中,确定模块,用于根据目标传输模式,确定目标应用时间。In a possible implementation manner, the target application time of the above target AI model that satisfies the target AI model is determined based on the target transmission mode. The first AI model transmission device 60 provided by the embodiment of the present application may also include: a determination module. Among them, the determination module is used to determine the target application time according to the target transmission mode.
在一种可能的实现方式中,上述确定模块,具体用于将目标传输模式对应的应用时间,确定为目标应用时间;第一应用时间和第二应用时间之间满足以下至少一项:第一应用时间大于第二应用时间;第一应用时间大于或等于第三应用时间,该第三应用时间是根据第二应用时间和AI模型的编译时间确定的。其中,上述第一应用时间为:第一传输模式对应的AI模型应用时间;上述第二应用时间为:第二传输模式对应的AI模型应用时间。In a possible implementation, the determination module is specifically used to determine the application time corresponding to the target transmission mode as the target application time; the first application time and the second application time satisfy at least one of the following: the first application time is greater than the second application time; the first application time is greater than or equal to the third application time, and the third application time is determined based on the second application time and the compilation time of the AI model. The first application time is: the AI model application time corresponding to the first transmission mode; the second application time is: the AI model application time corresponding to the second transmission mode.
在一种可能的实现方式中,上述确定模块,具体用于在目标传输模式为第一传输模式的情况下,若目标AI模型和第一AI模型传输装置60中的预存AI模型相匹配,则将第四应用时间确定为目标应用时间。其中,上述第四应用时间小于第一应用时间;该第一应用时间为:第一传输模式对应的AI模型应用时间。In a possible implementation, the above-mentioned determination module is specifically used to: when the target transmission mode is the first transmission mode, if the target AI model matches the pre-stored AI model in the first AI model transmission device 60, then The fourth application time is determined as the target application time. Wherein, the above-mentioned fourth application time is less than the first application time; the first application time is: the AI model application time corresponding to the first transmission mode.
在一种可能的实现方式中,上述目标AI模型满足目标AI模型触发第一AI模型传输装置60向第二AI模型传输装置发送目标反馈事件;该目标反馈事件为以下任一项:第一反馈事件、第二反馈事件。其中,上述第一反馈事件用于反馈未应用目标AI模型;上述第二反馈事件用于反馈已应用目标AI模型。In a possible implementation, the above target AI model satisfies the target AI model and triggers the first AI model transmission device 60 to send a target feedback event to the second AI model transmission device; the target feedback event is any of the following: first feedback event, second feedback event. Among them, the above-mentioned first feedback event is used to feedback that the target AI model has not been applied; the above-mentioned second feedback event is used to feedback that the target AI model has been applied.
在一种可能的实现方式中,上述第一反馈事件为以下任一项:第一AI模型传输装置60未应用目标AI模型;第一AI模型传输装置60未应用目标AI模型,且第一AI模型传输装置60仍应用第一AI模型;第一AI模型传输装置60未应用目标AI模型,且第一AI模型传输装置60不应用AI模型;第一AI模型传输装置60不支持第二AI模型传输装置通过目标传输模式发送的目标模型信息;第三反馈事件;第一AI模型传输装置60不支持目标AI模型的第一模型信息。其中,上述第一AI模型为:第一AI模型传输装置60接收目标AI模型之前应用的AI模型;上述第三反馈事件为:第一AI模型传输装置60不支持第二AI模型传输装置通过目标传输模式发送的目标模型信息;上述第三反馈事件携带第一信息,该第一信息为目标AI模型中的第一AI模型传输装置60不支持的模型结构的信息;上述第一模型信息包括以下至少一项:模型大小、模型复杂度、模型操作数。In a possible implementation, the above-mentioned first feedback event is any one of the following: the first AI model transmission device 60 does not apply the target AI model; the first AI model transmission device 60 does not apply the target AI model, and the first AI The model transmission device 60 still applies the first AI model; the first AI model transmission device 60 does not apply the target AI model, and the first AI model transmission device 60 does not apply the AI model; the first AI model transmission device 60 does not support the second AI model The target model information sent by the transmission device through the target transmission mode; the third feedback event; the first AI model transmission device 60 does not support the first model information of the target AI model. Wherein, the above-mentioned first AI model is: the AI model applied before the first AI model transmission device 60 receives the target AI model; the above-mentioned third feedback event is: the first AI model transmission device 60 does not support the second AI model transmission device to pass the target Target model information sent in transmission mode; the above-mentioned third feedback event carries first information, which is information about a model structure that is not supported by the first AI model transmission device 60 in the target AI model; the above-mentioned first model information includes the following At least one item: model size, model complexity, and number of model operations.
在一种可能的实现方式中,上述第二反馈事件为以下任一项:第一AI模型传输装置60已应用目标AI模型;第四反馈事件;第一AI模型传输装置60已应用目标AI模型,且已替换第一AI模型;第一AI模型传输装置60支持第二AI模型传输装置通过目标传输模式发送的目标模型信息;第一AI模型传输装置60已编译目标AI模型。其中,上述第四反馈事件为:第一AI模型传输装置60已应用目标AI模型;上述第四反馈事件携带第二信息,该第二信息为目标AI模型的模型标识信息;上述第一AI模型为:第一AI模型传输装置60接收目标AI模型之前应用的AI模型。In a possible implementation, the above-mentioned second feedback event is any one of the following: the first AI model transmission device 60 has applied the target AI model; the fourth feedback event; the first AI model transmission device 60 has applied the target AI model , and the first AI model has been replaced; the first AI model transmission device 60 supports the target model information sent by the second AI model transmission device through the target transmission mode; the first AI model transmission device 60 has compiled the target AI model. Wherein, the above-mentioned fourth feedback event is: the first AI model transmission device 60 has applied the target AI model; the above-mentioned fourth feedback event carries second information, and the second information is the model identification information of the target AI model; the above-mentioned first AI model is: the AI model applied before the first AI model transmission device 60 receives the target AI model.
在一种可能的实现方式中,本申请实施例提供的第一AI模型传输装置60还可以包括:处理模块。其中,处理模块,用于在目标AI模型与第一AI模型传输装置60的能力不匹配的情况下,不应用目标AI模型。In a possible implementation, the first AI model transmission device 60 provided in the embodiment of the present application may also include: a processing module. The processing module is configured to not apply the target AI model when the target AI model does not match the capabilities of the first AI model transmission device 60 .
在一种可能的实现方式中,本申请实施例提供的第一AI模型传输装置60还可以包括:发送模块。其中,发送模块,用于按照第三传输模式,向第二AI模型传输装置发送第二AI模型的第二模型信息。其中,上述第三传输模式为第一传输模式或第四传输模式;在目标传输模式为第一传输模式的情况下,第二模型信息包括第二AI模型的模型结构信息和模型参数信息;在目标传输模式为第四传输模式的情况下,第二模型信息包括第二AI模型的模型结构信息;目标AI模型与第二AI模型的模型结构信息相匹配。In a possible implementation, the first AI model transmission device 60 provided in the embodiment of the present application may also include: a sending module. The sending module is configured to send the second model information of the second AI model to the second AI model transmission device according to the third transmission mode. Wherein, the above-mentioned third transmission mode is the first transmission mode or the fourth transmission mode; when the target transmission mode is the first transmission mode, the second model information includes model structure information and model parameter information of the second AI model; in When the target transmission mode is the fourth transmission mode, the second model information includes model structure information of the second AI model; the target AI model matches the model structure information of the second AI model.
在一种可能的实现方式中,处理模块,还用于采用目标AI模型,执行第一操作。其中,上述第一操作包括以下至少一项:信号处理操作;信号传输操作;信号解调操作;信道状态信息获取操作;波束管理操作;信道预测操作;干扰抑制操作;定位操作;高层业务的预测和管理操作;高层参数的预测和管理操作;控制信令解析操作。In a possible implementation, the processing module is also used to adopt the target AI model to perform the first operation. Wherein, the above-mentioned first operation includes at least one of the following: signal processing operation; signal transmission operation; signal demodulation operation; channel state information acquisition operation; beam management operation; channel prediction operation; interference suppression operation; positioning operation; prediction of high-level services and management operations; high-level parameter prediction and management operations; control signaling parsing operations.
本申请实施例提供的AI模型传输装置,由于目标AI模型的目标传输模式,是根据第一AI模型传输装置支持的AI模型传输模式的能力确定的,因此,可以避免因第一AI模型传输装置不支持第一传输模式或第二传输模式,而导致无法根据目标模型信息得到目标AI
模型的情况,进而导致无法应用目标AI模型的情况;和/或,由于目标AI模型的目标应用时间,是根据目标传输模式确定的,因此,可以避免因目标AI模型的应用时间过大(或过小),而导致无法应用目标AI模型、或浪费等待时间的情况;和/或,由于目标AI模型可以触发第一AI模型传输装置向第二AI模型传输装置发送用于反馈目标AI模型的应用情况的目标反馈事件,因此,第二AI模型传输装置可以直接根据目标反馈事件,向第一AI模型传输装置发送第一AI模型传输装置能够应用的AI模型,而无需第一AI模型传输装置与第二AI模型传输装置进行多次传输。从而可以减少在得到目标AI模型的过程中,第一AI模型传输装置与第二AI模型传输装置进行传输的次数,如此,可以节省传输资源。In the AI model transmission device provided by the embodiment of the present application, since the target transmission mode of the target AI model is determined based on the capabilities of the AI model transmission mode supported by the first AI model transmission device, it is possible to avoid errors caused by the first AI model transmission device. The first transmission mode or the second transmission mode is not supported, resulting in the inability to obtain the target AI based on the target model information. The situation of the model, which leads to the inability to apply the target AI model; and/or, since the target application time of the target AI model is determined based on the target transmission mode, it is possible to avoid the application time of the target AI model being too long (or is too small), resulting in the inability to apply the target AI model or a waste of waiting time; and/or, because the target AI model can trigger the first AI model transmission device to send feedback to the second AI model transmission device for feedback of the target AI model. The target feedback event of the application situation, therefore, the second AI model transmission device can directly send the AI model that the first AI model transmission device can apply to the first AI model transmission device according to the target feedback event, without the need for the first AI model transmission device Perform multiple transmissions with the second AI model transmission device. Therefore, the number of transmissions between the first AI model transmission device and the second AI model transmission device during the process of obtaining the target AI model can be reduced, thus saving transmission resources.
本申请实施例中的AI模型传输装置可以是电子设备,例如具有操作***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。示例性地,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。The AI model transmission 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. For example, 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.
本申请实施例提供的AI模型传输装置能够实现图2至图5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。The AI model transmission device provided by the embodiments of this application can implement each process implemented by the method embodiments in Figures 2 to 5 and achieve the same technical effect. To avoid duplication, the details will not be described here.
图8示出了本申请实施例中涉及的AI模型传输装置的一种可能的结构示意图,该AI模型传输装置为第二AI模型传输装置。如图8所示,该第二AI模型传输装置70可以包括:发送模块71。FIG. 8 shows a possible structural diagram of the AI model transmission device involved in the embodiment of the present application. The AI model transmission device is a second AI model transmission device. As shown in FIG. 8 , the second AI model transmission device 70 may include: a sending module 71 .
其中,发送模块71,用于向第一AI模型传输装置发送目标AI模型的目标模型信息;该目标模型信息用于第一AI模型传输装置得到所述目标AI模型;该目标AI模型满足以下至少一项:目标AI模型的目标传输模式,是根据第一AI模型传输装置支持的AI模型传输模式的能力确定的;目标AI模型的目标应用时间,是根据目标传输模式确定的;目标AI模型触发第一AI模型传输装置向第二AI模型传输装置70发送目标反馈事件,该目标反馈事件用于反馈目标AI模型的应用情况。其中,上述目标传输模式包括以下任一项:第一传输模式、第二传输模式;在目标传输模式为第一传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;在目标传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息。Among them, the sending module 71 is used to send the target model information of the target AI model to the first AI model transmission device; the target model information is used by the first AI model transmission device to obtain the target AI model; the target AI model satisfies at least the following: One item: The target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device; the target application time of the target AI model is determined based on the target transmission mode; the target AI model triggers The first AI model transmission device sends a target feedback event to the second AI model transmission device 70 , where the target feedback event is used to feed back the application status of the target AI model. Wherein, the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
在一种可能的实现方式中,上述目标AI模型满足目标AI模型的目标传输模式,是根据第一AI模型传输装置支持的AI模型传输模式的能力确定的。本申请实施例提供的第二AI模型传输装置70还可以包括:接收模块和确定模块。其中,接收模块,用于从第一AI模型传输装置接收目标信息,该目标信息用于指示第一AI模型传输装置支持的AI模型传输模式的能力。确定模块,用于根据接收模块接收的目标信息,确定目标传输模式。上述发送模块71,还用于基于确定模块确定的目标传输模式,向第一AI模型传输装置发送目标模型信息。In a possible implementation manner, the target AI model satisfying the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device. The second AI model transmission device 70 provided by the embodiment of the present application may also include: a receiving module and a determining module. The receiving module is configured to receive target information from the first AI model transmission device, where the target information is used to indicate the capabilities of the AI model transmission mode supported by the first AI model transmission device. The determining module is used to determine the target transmission mode according to the target information received by the receiving module. The above-mentioned sending module 71 is also used to send target model information to the first AI model transmission device based on the target transmission mode determined by the determination module.
在一种可能的实现方式中,上述目标传输模式为第一传输模式。上述确定模块,还用于确定目标AI模型和第一AI模型传输装置中的预存AI模型是否相匹配。上述发送模块71,还用于若确定模块确定目标AI模型和预存AI模型相匹配,则按照第四传输模式,向第一AI模型传输装置发送目标模型信息。其中,上述第四传输模式为以下任一项:第二传输模式、第五传输模式;在第四传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息;在第四传输模式为第五传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;该第五传输模式对应的AI模型应用时间小于第一传输模式对应的AI模型应用时间。In a possible implementation manner, the target transmission mode is the first transmission mode. The above-mentioned determination module is also used to determine whether the target AI model matches the pre-stored AI model in the first AI model transmission device. The above-mentioned sending module 71 is also used to send the target model information to the first AI model transmission device according to the fourth transmission mode if the determination module determines that the target AI model matches the pre-stored AI model. Wherein, the above-mentioned fourth transmission mode is any of the following: the second transmission mode or the fifth transmission mode; when the fourth transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model; in When the fourth transmission mode is the fifth transmission mode, the target model information includes model structure information and model parameter information of the target AI model; the AI model application time corresponding to the fifth transmission mode is shorter than the AI model application time corresponding to the first transmission mode. time.
在一种可能的实现方式中,上述接收模块,还用于从第一AI模型传输装置接收目标反馈事件。其中,上述目标反馈事件用于反馈目标AI模型的应用情况。In a possible implementation, the above-mentioned receiving module is also used to receive the target feedback event from the first AI model transmission device. Among them, the above-mentioned target feedback event is used to feedback the application status of the target AI model.
在一种可能的实现方式中,上述接收模块,还用于从第一AI模型传输装置接收第二AI模型的第二模型信息,该第二模型信息是第一AI模型传输装置按照第三传输模式发送的。上述确定模块,还用于将与接收模块接收的第二AI模型的模型结构信息相匹配的模型结构信息,确定为目标AI模型的模型结构信息。其中,上述第三传输模式为第一传输模式或第四传输模式;在目标传输模式为第一传输模式的情况下,第二模型信息包括第二AI模型的模型结构信息和模型参数信息;在目标传输模式为第四传输模式的情况下,第二模型信息包括第二AI模型的模型结构信息。
In a possible implementation, the above-mentioned receiving module is also used to receive second model information of the second AI model from the first AI model transmission device. The second model information is transmitted by the first AI model transmission device according to the third transmission method. mode sent. The above-mentioned determining module is also used to determine the model structure information that matches the model structure information of the second AI model received by the receiving module as the model structure information of the target AI model. Wherein, the above-mentioned third transmission mode is the first transmission mode or the fourth transmission mode; when the target transmission mode is the first transmission mode, the second model information includes model structure information and model parameter information of the second AI model; in When the target transmission mode is the fourth transmission mode, the second model information includes model structure information of the second AI model.
在一种可能的实现方式中,上述确定模块,具体用于以下任一项:将与第二AI模型的模型结构信息相同的模型结构信息,确定为目标AI模型的模型结构信息;将与第二AI模型的模型结构信息的相似度大于或等于预设门限的模型结构,确定为目标AI模型的模型结构信息。In a possible implementation, the above determination module is specifically used for any of the following: determining the model structure information that is the same as the model structure information of the second AI model as the model structure information of the target AI model; The model structure whose similarity between the model structure information of the two AI models is greater than or equal to the preset threshold is determined as the model structure information of the target AI model.
本申请实施例提供的AI模型传输装置,由于目标AI模型的目标传输模式,是根据第一AI模型传输装置支持的AI模型传输模式的能力确定的,因此,可以避免因第一AI模型传输装置不支持第一传输模式或第二传输模式,而导致无法根据目标模型信息得到目标AI模型的情况,进而导致无法应用目标AI模型的情况;和/或,由于目标AI模型的目标应用时间,是根据目标传输模式确定的,因此,可以避免因目标AI模型的应用时间过大(或过小),而导致无法应用目标AI模型、或浪费等待时间的情况;和/或,由于目标AI模型可以触发第一AI模型传输装置向第二AI模型传输装置发送用于反馈目标AI模型的应用情况的目标反馈事件,因此,第二AI模型传输装置可以直接根据目标反馈事件,向第一AI模型传输装置发送第一AI模型传输装置能够应用的AI模型,而无需第一AI模型传输装置与第二AI模型传输装置进行多次传输。从而可以减少在得到目标AI模型的过程中,第一AI模型传输装置与第二AI模型传输装置进行传输的次数,如此,可以节省传输资源。The AI model transmission device provided in the embodiment of the present application, since the target transmission mode of the target AI model is determined according to the capability of the AI model transmission mode supported by the first AI model transmission device, it can avoid the situation that the target AI model cannot be obtained according to the target model information due to the first AI model transmission device not supporting the first transmission mode or the second transmission mode, and thus the situation that the target AI model cannot be applied; and/or, since the target application time of the target AI model is determined according to the target transmission mode, it can avoid the situation that the target AI model cannot be applied or the waiting time is wasted due to the application time of the target AI model being too long (or too short); and/or, since the target AI model can trigger the first AI model transmission device to send a target feedback event for feedback on the application of the target AI model to the second AI model transmission device, the second AI model transmission device can directly send the AI model that the first AI model transmission device can apply to the first AI model transmission device according to the target feedback event, without the need for the first AI model transmission device and the second AI model transmission device to perform multiple transmissions. Thereby, the number of transmissions between the first AI model transmission device and the second AI model transmission device in the process of obtaining the target AI model can be reduced, so that transmission resources can be saved.
本申请实施例中的AI模型传输装置可以是电子设备,例如具有操作***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。示例性地,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。The AI model transmission 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. For example, 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.
本申请实施例提供的AI模型传输装置能够实现图6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。The AI model transmission device provided in the embodiment of the present application can implement each process implemented by the method embodiment of Figure 6 and achieve the same technical effect. To avoid repetition, it will not be repeated here.
可选地,本申请实施例中,如图9所示,本申请实施例还提供一种通信设备80,包括处理器81和存储器82,存储器82上存储有可在所述处理器81上运行的程序或指令,例如,该通信设备80为终端时,该程序或指令被处理器81执行时实现上述AI模型传输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80为网络侧设备时,该程序或指令被处理器81执行时实现上述AI模型传输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。Optionally, in this embodiment of the present application, as shown in Figure 9, this embodiment of the present application also provides a communication device 80, which includes a processor 81 and a memory 82. The memory 82 stores information that can run on the processor 81. A program or instruction. For example, when the communication device 80 is a terminal, when the program or instruction is executed by the processor 81, each step of the above-mentioned AI model transmission method embodiment is implemented, and the same technical effect can be achieved. When the communication device 80 is a network-side device, when the program or instruction is executed by the processor 81, the steps of the above-mentioned AI model transmission method embodiment are implemented, and the same technical effect can be achieved. To avoid duplication, they will not be described again here.
本申请实施例还提供一种终端,该终端为第一终端,该第一终端包括处理器和通信接口,通信接口用于从第二终端接收目标AI模型的目标模型信息,处理器用于根据目标模型信息,得到目标AI模型。该目标AI模型满足以下至少一项:目标AI模型的目标传输模式,是根据第一终端支持的AI模型传输模式的能力确定的;目标AI模型的目标应用时间,是根据目标传输模式确定的;目标AI模型触发第一终端向第二终端发送目标反馈事件,该目标反馈事件用于反馈目标AI模型的应用情况。其中,上述目标传输模式包括以下任一项:第一传输模式、第二传输模式;在目标传输模式为第一传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;在目标传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息。该终端实施例与上述终端侧方法实施例对应,上述方法实施例的各个实施过程和实现方式均可适用于该终端实施例中,且能达到相同的技术效果。具体地,图10为实现本申请实施例的一种终端的硬件结构示意图。An embodiment of the present application also provides a terminal. The terminal is a first terminal. The first terminal includes a processor and a communication interface. The communication interface is used to receive target model information of the target AI model from the second terminal. The processor is used to calculate the target AI model according to the target. Model information to obtain the target AI model. The target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first terminal; the target application time of the target AI model is determined based on the target transmission mode; The target AI model triggers the first terminal to send a target feedback event to the second terminal, and the target feedback event is used to feed back the application status of the target AI model. Wherein, the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model. This terminal embodiment corresponds to the above-mentioned terminal-side method embodiment. Each implementation process and implementation manner of the above-mentioned method embodiment can be applied to this terminal embodiment, and can achieve the same technical effect. Specifically, FIG. 10 is a schematic diagram of the hardware structure of a terminal that implements an embodiment of the present application.
该终端800包括但不限于:射频单元801、网络模块802、音频输出单元803、输入单元804、传感器805、显示单元806、用户输入单元807、接口单元808、存储器809以及处理器810等中的至少部分部件。The terminal 800 includes but is not limited to: a radio frequency unit 801, a network module 802, an audio output unit 803, an input unit 804, a sensor 805, a display unit 806, a user input unit 807, an interface unit 808, a memory 809, a processor 810, etc. At least some parts.
本领域技术人员可以理解,终端800还可以包括给各个部件供电的电源(比如电池),电源可以通过电源管理***与处理器810逻辑相连,从而通过电源管理***实现管理充电、放电、以及功耗管理等功能。图10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。Those skilled in the art can understand that the terminal 800 may also include a power supply (such as a battery) that supplies power to various components. The power supply may be logically connected to the processor 810 through a power management system, thereby managing charging, discharging, and power consumption through the power management system. Management and other functions. The terminal structure shown in FIG. 10 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.
应理解的是,本申请实施例中,输入单元804可以包括图形处理单元(Graphics Processing Unit,GPU)8041和麦克风8042,图形处理器8041对在视频捕获模式或图像捕获模式中由图像捕获装置(如摄像头)获得的静态图片或视频的图像数据进行处理。显示单元806可包括显示面板8061,可以采用液晶显示器、有机发光二极管等形式来配置显示面板8061。用户输入单元807包括触控面板8071以及其他输入设备8072中的至少一种。触控面板8071,
也称为触摸屏。触控面板8071可包括触摸检测装置和触摸控制器两个部分。其他输入设备8072可以包括但不限于物理键盘、功能键(比如音量控制按键、开关按键等)、轨迹球、鼠标、操作杆,在此不再赘述。It should be understood that in the embodiment of the present application, the input unit 804 may include a graphics processing unit (GPU) 8041 and a microphone 8042. The graphics processor 8041 is responsible for the image capture device (GPU) in the video capture mode or the image capture mode. Process the image data of still pictures or videos obtained by cameras (such as cameras). The display unit 806 may include a display panel 8061, which may be configured in the form of a liquid crystal display, an organic light emitting diode, or the like. The user input unit 807 includes a touch panel 8071 and at least one of other input devices 8072 . Touch panel 8071, Also called a touch screen. The touch panel 8071 may include two parts: a touch detection device and a touch controller. Other input devices 8072 may include but are not limited to physical keyboards, function keys (such as volume control keys, switch keys, etc.), trackballs, mice, and joysticks, which will not be described again here.
本申请实施例中,射频单元801接收来自网络侧设备的下行数据后,可以传输给处理器810进行处理;另外,射频单元801可以向网络侧设备发送上行数据。通常,射频单元801包括但不限于天线、放大器、收发信机、耦合器、低噪声放大器、双工器等。In this embodiment of the present application, after receiving downlink data from the network side device, the radio frequency unit 801 can transmit it to the processor 810 for processing; in addition, the radio frequency unit 801 can send uplink data to the network side device. Generally, the radio frequency unit 801 includes, but is not limited to, an antenna, amplifier, transceiver, coupler, low noise amplifier, duplexer, etc.
存储器809可用于存储软件程序或指令以及各种数据。存储器809可主要包括存储程序或指令的第一存储区和存储数据的第二存储区,其中,第一存储区可存储操作***、至少一个功能所需的应用程序或指令(比如声音播放功能、图像播放功能等)等。此外,存储器809可以包括易失性存储器或非易失性存储器,或者,存储器809可以包括易失性和非易失性存储器两者。其中,非易失性存储器可以是只读存储器(Read-Only Memory,ROM)、可编程只读存储器(Programmable ROM,PROM)、可擦除可编程只读存储器(Erasable PROM,EPROM)、电可擦除可编程只读存储器(Electrically EPROM,EEPROM)或闪存。易失性存储器可以是随机存取存储器(Random Access Memory,RAM),静态随机存取存储器(Static RAM,SRAM)、动态随机存取存储器(Dynamic RAM,DRAM)、同步动态随机存取存储器(Synchronous DRAM,SDRAM)、双倍数据速率同步动态随机存取存储器(Double Data Rate SDRAM,DDRSDRAM)、增强型同步动态随机存取存储器(Enhanced SDRAM,ESDRAM)、同步连接动态随机存取存储器(Synch link DRAM,SLDRAM)和直接内存总线随机存取存储器(Direct Rambus RAM,DRRAM)。本申请实施例中的存储器809包括但不限于这些和任意其它适合类型的存储器。Memory 809 may be used to store software programs or instructions as well as various data. The memory 809 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first 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. Additionally, memory 809 may include volatile memory or non-volatile memory, or memory 809 may include both volatile and non-volatile memory. Among them, 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). Memory 809 in embodiments of the present application includes, but is not limited to, these and any other suitable types of memory.
处理器810可包括一个或多个处理单元;可选地,处理器810集成应用处理器和调制解调处理器,其中,应用处理器主要处理涉及操作***、用户界面和应用程序等的操作,调制解调处理器主要处理无线通信信号,如基带处理器。可以理解的是,上述调制解调处理器也可以不集成到处理器810中。The processor 810 may include one or more processing units; optionally, the processor 810 integrates an application processor and a modem processor, where the application processor mainly handles operations related to the operating system, user interface, application programs, etc., Modem processors mainly process wireless communication signals, such as baseband processors. It can be understood that the above modem processor may not be integrated into the processor 810.
其中,射频单元801,用于从第二终端接收目标AI模型的目标模型信息。Among them, the radio frequency unit 801 is used to receive target model information of the target AI model from the second terminal.
上述目标AI模型满足以下至少一项:目标AI模型的目标传输模式,是根据第一终端支持的AI模型传输模式的能力确定的;目标AI模型的目标应用时间,是根据目标传输模式确定的;目标AI模型触发第一终端向第二终端发送目标反馈事件,该目标反馈事件用于反馈目标AI模型的应用情况。The above-mentioned target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined according to the capabilities of the AI model transmission mode supported by the first terminal; the target application time of the target AI model is determined according to the target transmission mode; the target AI model triggers the first terminal to send a target feedback event to the second terminal, and the target feedback event is used to feedback the application status of the target AI model.
其中,上述目标传输模式包括以下任一项:第一传输模式、第二传输模式;在目标传输模式为第一传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;在目标传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息。Wherein, the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
本申请实施例提供的终端,由于目标AI模型的目标传输模式,是根据第一终端支持的AI模型传输模式的能力确定的,因此,可以避免因第一终端不支持第一传输模式或第二传输模式,而导致无法根据目标模型信息得到目标AI模型的情况,进而导致无法应用目标AI模型的情况;和/或,由于目标AI模型的目标应用时间,是根据目标传输模式确定的,因此,可以避免因目标AI模型的应用时间过大(或过小),而导致无法应用目标AI模型、或浪费等待时间的情况;和/或,由于目标AI模型可以触发第一终端向第二终端发送用于反馈目标AI模型的应用情况的目标反馈事件,因此,第二终端可以直接根据目标反馈事件,向第一终端发送第一终端能够应用的AI模型,而无需第一终端与第二终端进行多次传输。从而可以减少在得到目标AI模型的过程中,第一终端与第二终端进行传输的次数,如此,可以节省传输资源。In the terminal provided by the embodiment of the present application, since the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first terminal, it is possible to avoid errors caused by the first terminal not supporting the first transmission mode or the second transmission mode. transmission mode, resulting in the inability to obtain the target AI model based on the target model information, resulting in the inability to apply the target AI model; and/or, because the target application time of the target AI model is determined based on the target transmission mode, therefore, It can avoid the situation where the target AI model cannot be applied or the waiting time is wasted because the application time of the target AI model is too long (or too small); and/or, because the target AI model can trigger the first terminal to send data to the second terminal. A target feedback event used to feed back the application status of the target AI model. Therefore, the second terminal can directly send the AI model that the first terminal can apply to the first terminal based on the target feedback event without the need for the first terminal to communicate with the second terminal. Multiple transfers. This can reduce the number of transmissions between the first terminal and the second terminal in the process of obtaining the target AI model, thus saving transmission resources.
可选地,本申请实施例中,上述目标AI模型满足目标AI模型的目标传输模式,是根据第一终端支持的AI模型传输模式的能力确定的。Optionally, in this embodiment of the present application, the target AI model satisfying the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first terminal.
射频单元801,还用于向所述第二终端发送目标信息,所述目标信息用于指示所述第一终端支持的AI模型传输模式的能力。The radio frequency unit 801 is also used to send target information to the second terminal, where the target information is used to indicate the capabilities of the AI model transmission mode supported by the first terminal.
如此可知,由于第一终端可以向第二终端发送目标信息,以使得第二终端可以根据第一终端支持的AI模型传输模式的能力,确定目标传输模式,因此,可以避免因第一终端不支持第一传输模式或第二传输模式,而导致第一终端无法得到目标AI模型的情况,进而导致无法应用目标AI模型的情况。
It can be seen from this that since the first terminal can send the target information to the second terminal, so that the second terminal can determine the target transmission mode according to the capability of the AI model transmission mode supported by the first terminal, therefore, it can be avoided that the first terminal does not support the target transmission mode. The first transmission mode or the second transmission mode causes the first terminal to be unable to obtain the target AI model, which further leads to the situation that the target AI model cannot be applied.
可选地,本申请实施例中,上述目标AI模型满足目标AI模型的目标应用时间,是根据目标传输模式确定的。Optionally, in this embodiment of the present application, the target application time of the target AI model that satisfies the target AI model is determined based on the target transmission mode.
处理器810,用于根据目标传输模式,确定目标应用时间。The processor 810 is used to determine the target application time according to the target transmission mode.
如此可知,由于第一终端可以将目标传输模式对应的应用时间确定为目标应用时间;其中,第一应用时间大于第二应用时间、和/或第一应用时间大于或等于第三应用时间,因此,可以避免因目标AI模型的应用时间过大(或过小),而导致无法应用目标AI模型、或浪费等待时间的情况。It can be seen from this that since the first terminal can determine the application time corresponding to the target transmission mode as the target application time; wherein the first application time is greater than the second application time, and/or the first application time is greater than or equal to the third application time, therefore , which can avoid situations where the target AI model cannot be applied or waiting time is wasted due to the application time of the target AI model being too large (or too small).
可选地,本申请实施例中,处理器810,具体用于将目标传输模式对应的应用时间,确定为目标应用时间。Optionally, in this embodiment of the present application, the processor 810 is specifically configured to determine the application time corresponding to the target transmission mode as the target application time.
第一应用时间和第二应用时间之间满足以下至少一项:第一应用时间大于第二应用时间;第一应用时间大于或等于第三应用时间,该第三应用时间是根据第二应用时间和AI模型的编译时间确定的。At least one of the following is satisfied between the first application time and the second application time: the first application time is greater than the second application time; the first application time is greater than or equal to the third application time, and the third application time is based on the second application time Determined by the compilation time of the AI model.
其中,上述第一应用时间为:第一传输模式对应的AI模型应用时间;上述第二应用时间为:第二传输模式对应的AI模型应用时间。Wherein, the above-mentioned first application time is: the AI model application time corresponding to the first transmission mode; the above-mentioned second application time is: the AI model application time corresponding to the second transmission mode.
如此可知,由于第一终端可以将目标传输模式对应的应用时间确定为目标应用时间;其中,第一应用时间大于第二应用时间、和/或第一应用时间大于或等于第三应用时间,因此,可以避免因目标AI模型的应用时间过大(或过小),而导致无法应用目标AI模型、或浪费等待时间的情况。It can be seen from this that since the first terminal can determine the application time corresponding to the target transmission mode as the target application time; wherein the first application time is greater than the second application time, and/or the first application time is greater than or equal to the third application time, therefore , which can avoid situations where the target AI model cannot be applied or waiting time is wasted due to the application time of the target AI model being too large (or too small).
可选地,本申请实施例中,处理器810,具体用于在目标传输模式为第一传输模式的情况下,若目标AI模型和第一终端中的预存AI模型相匹配,则将第四应用时间确定为目标应用时间。Optionally, in this embodiment of the present application, the processor 810 is specifically configured to, when the target transmission mode is the first transmission mode, if the target AI model matches the pre-stored AI model in the first terminal, then convert the fourth The application time is determined as the target application time.
其中,上述第四应用时间小于第一应用时间;该第一应用时间为:第一传输模式对应的AI模型应用时间。Wherein, the above-mentioned fourth application time is less than the first application time; the first application time is: the AI model application time corresponding to the first transmission mode.
如此可知,由于在目标AI模型和预存AI模型相匹配的情况下,第一终端可以将较小的应用时间,确定为目标应用时间,因此,可以减少应用目标AI模型的时延。It can be seen that, since the first terminal can determine a smaller application time as the target application time when the target AI model matches the pre-stored AI model, the delay in applying the target AI model can be reduced.
可选地,本申请实施例中,处理器810,还用于在目标AI模型与第一终端的能力不匹配的情况下,不应用目标AI模型。Optionally, in the embodiment of the present application, the processor 810 is further configured to not apply the target AI model when the target AI model does not match the capability of the first terminal.
如此可知,由于在目标AI模型与第一终端的能力不匹配的情况下,第一终端可以不应用目标AI模型,因此,可以避免第一终端使用目标AI模型进行通信的过程中,第一终端发生卡顿的情况。It can be seen from this that since the first terminal does not need to apply the target AI model when the target AI model does not match the capabilities of the first terminal, it can be avoided that the first terminal uses the target AI model for communication. Stuttering occurs.
可选地,本申请实施例中,射频单元801,还用于按照第三传输模式,向第二终端发送第二AI模型的第二模型信息。Optionally, in this embodiment of the present application, the radio frequency unit 801 is also configured to send the second model information of the second AI model to the second terminal according to the third transmission mode.
其中,上述第三传输模式为第一传输模式或第四传输模式;在目标传输模式为第一传输模式的情况下,第二模型信息包括第二AI模型的模型结构信息和模型参数信息;在目标传输模式为第四传输模式的情况下,第二模型信息包括第二AI模型的模型结构信息;目标AI模型与第二AI模型的模型结构信息相匹配。Wherein, the above-mentioned third transmission mode is the first transmission mode or the fourth transmission mode; when the target transmission mode is the first transmission mode, the second model information includes model structure information and model parameter information of the second AI model; in When the target transmission mode is the fourth transmission mode, the second model information includes model structure information of the second AI model; the target AI model matches the model structure information of the second AI model.
如此可知,由于目标AI模型和第二AI模型的模型结构信息相匹配,因此,第一终端可以直接从第二终端接收目标AI模型的模型参数信息,以得到目标AI模型,而无需重复接收目标AI模型的模型结构信息,如此,可以节省传输资源。It can be seen that since the model structure information of the target AI model and the second AI model match, the first terminal can directly receive the model parameter information of the target AI model from the second terminal to obtain the target AI model without repeatedly receiving the target. The model structure information of the AI model can save transmission resources.
可选地,本申请实施例中,处理器810,还用于采用目标AI模型,执行第一操作。Optionally, in this embodiment of the present application, the processor 810 is also configured to use the target AI model to perform the first operation.
其中,上述第一操作包括以下至少一项:信号处理操作;信号传输操作;信号解调操作;信道状态信息获取操作;波束管理操作;信道预测操作;干扰抑制操作;定位操作;高层业务的预测和管理操作;高层参数的预测和管理操作;控制信令解析操作。Wherein, the above-mentioned first operation includes at least one of the following: signal processing operation; signal transmission operation; signal demodulation operation; channel state information acquisition operation; beam management operation; channel prediction operation; interference suppression operation; positioning operation; prediction of high-level services and management operations; high-level parameter prediction and management operations; control signaling parsing operations.
如此可知,由于第一终端可以采用目标AI模型,执行信号处理操作、信号传输操作、信道状态信息获取操作、波束管理操作、信道预测操作、干扰抑制操作、定位操作、高层业务的预测和管理操作、高层参数的预测和管理操作以及控制信令解析操作中的至少一个操作,因此,可以提升第一终端通信的吞吐量,且降低第一终端通信的时延。It can be seen that the first terminal can use the target AI model to perform signal processing operations, signal transmission operations, channel state information acquisition operations, beam management operations, channel prediction operations, interference suppression operations, positioning operations, and prediction and management operations of high-level services , at least one of a high-level parameter prediction and management operation and a control signaling parsing operation. Therefore, the throughput of the first terminal communication can be improved and the delay of the first terminal communication can be reduced.
本申请实施例还提供一种终端,该终端为第二终端,该第二终端包括处理器和通信接口,通信接口用于向第一终端发送目标AI模型的目标模型信息。该目标模型信息用于第一终端得到目标AI模型;该目标AI模型满足以下至少一项:目标AI模型的目标传输模式,是根据第一终端支持的AI模型传输模式的能力确定的;目标AI模型的目标应用时间,是
根据目标传输模式确定的;目标AI模型触发第一终端向第二终端发送目标反馈事件,该目标反馈事件用于反馈目标AI模型的应用情况。其中,上述目标传输模式包括以下任一项:第一传输模式、第二传输模式;在目标传输模式为第一传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;在目标传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息。该终端实施例与上述终端侧方法实施例对应,上述方法实施例的各个实施过程和实现方式均可适用于该终端实施例中,且能达到相同的技术效果。具体地,图10为实现本申请实施例的一种终端的硬件结构示意图。An embodiment of the present application also provides a terminal, which is a second terminal. The second terminal includes a processor and a communication interface. The communication interface is used to send target model information of the target AI model to the first terminal. The target model information is used by the first terminal to obtain the target AI model; the target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first terminal; the target AI The target application time of the model is Determined according to the target transmission mode; the target AI model triggers the first terminal to send a target feedback event to the second terminal, and the target feedback event is used to feed back the application status of the target AI model. Wherein, the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model. This terminal embodiment corresponds to the above-mentioned terminal-side method embodiment. Each implementation process and implementation manner of the above-mentioned method embodiment can be applied to this terminal embodiment, and can achieve the same technical effect. Specifically, FIG. 10 is a schematic diagram of the hardware structure of a terminal that implements an embodiment of the present application.
该终端800包括但不限于:射频单元801、网络模块802、音频输出单元803、输入单元804、传感器805、显示单元806、用户输入单元807、接口单元808、存储器809以及处理器810等中的至少部分部件。The terminal 800 includes but is not limited to: a radio frequency unit 801, a network module 802, an audio output unit 803, an input unit 804, a sensor 805, a display unit 806, a user input unit 807, an interface unit 808, a memory 809 and at least some of the components of a processor 810.
本领域技术人员可以理解,终端800还可以包括给各个部件供电的电源(比如电池),电源可以通过电源管理***与处理器810逻辑相连,从而通过电源管理***实现管理充电、放电、以及功耗管理等功能。图10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。Those skilled in the art can understand that the terminal 800 may also include a power supply (such as a battery) that supplies power to various components. The power supply may be logically connected to the processor 810 through a power management system, thereby managing charging, discharging, and power consumption through the power management system. Management and other functions. The terminal structure shown in FIG. 10 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.
应理解的是,本申请实施例中,输入单元804可以包括图形处理单元(Graphics Processing Unit,GPU)8041和麦克风8042,图形处理器8041对在视频捕获模式或图像捕获模式中由图像捕获装置(如摄像头)获得的静态图片或视频的图像数据进行处理。显示单元806可包括显示面板8061,可以采用液晶显示器、有机发光二极管等形式来配置显示面板8061。用户输入单元807包括触控面板8071以及其他输入设备8072中的至少一种。触控面板8071,也称为触摸屏。触控面板8071可包括触摸检测装置和触摸控制器两个部分。其他输入设备8072可以包括但不限于物理键盘、功能键(比如音量控制按键、开关按键等)、轨迹球、鼠标、操作杆,在此不再赘述。It should be understood that in the embodiment of the present application, the input unit 804 may include a graphics processing unit (Graphics Processing Unit, GPU) 8041 and a microphone 8042. The graphics processor 8041 is responsible for the image capture device (GPU) in the video capture mode or the image capture mode. Process the image data of still pictures or videos obtained by cameras (such as cameras). The display unit 806 may include a display panel 8061, which may be configured in the form of a liquid crystal display, an organic light emitting diode, or the like. The user input unit 807 includes a touch panel 8071 and at least one of other input devices 8072 . Touch panel 8071, also known as touch screen. The touch panel 8071 may include two parts: a touch detection device and a touch controller. Other input devices 8072 may include but are not limited to physical keyboards, function keys (such as volume control keys, switch keys, etc.), trackballs, mice, and joysticks, which will not be described again here.
本申请实施例中,射频单元801接收来自网络侧设备的下行数据后,可以传输给处理器810进行处理;另外,射频单元801可以向网络侧设备发送上行数据。通常,射频单元801包括但不限于天线、放大器、收发信机、耦合器、低噪声放大器、双工器等。In the embodiment of the present application, after receiving downlink data from the network side device, the radio frequency unit 801 can transmit the data to the processor 810 for processing; in addition, the radio frequency unit 801 can send uplink data to the network side device. Generally, the radio frequency unit 801 includes but is not limited to an antenna, an amplifier, a transceiver, a coupler, a low noise amplifier, a duplexer, etc.
存储器809可用于存储软件程序或指令以及各种数据。存储器809可主要包括存储程序或指令的第一存储区和存储数据的第二存储区,其中,第一存储区可存储操作***、至少一个功能所需的应用程序或指令(比如声音播放功能、图像播放功能等)等。此外,存储器809可以包括易失性存储器或非易失性存储器,或者,存储器809可以包括易失性和非易失性存储器两者。其中,非易失性存储器可以是只读存储器(Read-Only Memory,ROM)、可编程只读存储器(Programmable ROM,PROM)、可擦除可编程只读存储器(Erasable PROM,EPROM)、电可擦除可编程只读存储器(Electrically EPROM,EEPROM)或闪存。易失性存储器可以是随机存取存储器(Random Access Memory,RAM),静态随机存取存储器(Static RAM,SRAM)、动态随机存取存储器(Dynamic RAM,DRAM)、同步动态随机存取存储器(Synchronous DRAM,SDRAM)、双倍数据速率同步动态随机存取存储器(Double Data Rate SDRAM,DDRSDRAM)、增强型同步动态随机存取存储器(Enhanced SDRAM,ESDRAM)、同步连接动态随机存取存储器(Synch link DRAM,SLDRAM)和直接内存总线随机存取存储器(Direct Rambus RAM,DRRAM)。本申请实施例中的存储器809包括但不限于这些和任意其它适合类型的存储器。Memory 809 may be used to store software programs or instructions as well as various data. The memory 809 may mainly include a first storage area for storing programs or instructions and a second storage area for storing data, wherein the first 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. Additionally, memory 809 may include volatile memory or non-volatile memory, or memory 809 may include both volatile and non-volatile memory. Among them, 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). Memory 809 in embodiments of the present application includes, but is not limited to, these and any other suitable types of memory.
处理器810可包括一个或多个处理单元;可选地,处理器810集成应用处理器和调制解调处理器,其中,应用处理器主要处理涉及操作***、用户界面和应用程序等的操作,调制解调处理器主要处理无线通信信号,如基带处理器。可以理解的是,上述调制解调处理器也可以不集成到处理器810中。The processor 810 may include one or more processing units; optionally, the processor 810 integrates an application processor and a modem processor, where the application processor mainly handles operations related to the operating system, user interface, application programs, etc., Modem processors mainly process wireless communication signals, such as baseband processors. It can be understood that the above modem processor may not be integrated into the processor 810.
其中,射频单元801,用于向第一终端发送目标AI模型的目标模型信息。Among them, the radio frequency unit 801 is used to send target model information of the target AI model to the first terminal.
上述目标模型信息用于第一终端得到目标AI模型。The above target model information is used by the first terminal to obtain the target AI model.
上述目标AI模型满足以下至少一项:目标AI模型的目标传输模式,是根据第一终端支持的AI模型传输模式的能力确定的;目标AI模型的目标应用时间,是根据目标传输模式确定的;目标AI模型触发第一终端向第二终端发送目标反馈事件,该目标反馈事件用于反馈目标AI模型的应用情况。The above target AI model satisfies at least one of the following: the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first terminal; the target application time of the target AI model is determined based on the target transmission mode; The target AI model triggers the first terminal to send a target feedback event to the second terminal, and the target feedback event is used to feed back the application status of the target AI model.
其中,上述目标传输模式包括以下任一项:第一传输模式、第二传输模式;在目标传输模式为第一传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参
数信息;在目标传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息。Wherein, the above-mentioned target transmission mode includes any of the following: a first transmission mode and a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameters of the target AI model. data information; when the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
本申请实施例提供的终端,由于目标AI模型的目标传输模式,是根据第一终端支持的AI模型传输模式的能力确定的,因此,可以避免因第一终端不支持第一传输模式或第二传输模式,而导致无法根据目标模型信息得到目标AI模型的情况,进而导致无法应用目标AI模型的情况;和/或,由于目标AI模型的目标应用时间,是根据目标传输模式确定的,因此,可以避免因目标AI模型的应用时间过大(或过小),而导致无法应用目标AI模型、或浪费等待时间的情况;和/或,由于目标AI模型可以触发第一终端向第二终端发送用于反馈目标AI模型的应用情况的目标反馈事件,因此,第二终端可以直接根据目标反馈事件,向第一终端发送第一终端能够应用的AI模型,而无需第一终端与第二终端进行多次传输。从而可以减少在得到目标AI模型的过程中,第一终端与第二终端进行传输的次数,如此,可以节省传输资源。In the terminal provided by the embodiment of the present application, since the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first terminal, it is possible to avoid errors caused by the first terminal not supporting the first transmission mode or the second transmission mode. transmission mode, resulting in the inability to obtain the target AI model based on the target model information, resulting in the inability to apply the target AI model; and/or, because the target application time of the target AI model is determined based on the target transmission mode, therefore, It can avoid the situation where the target AI model cannot be applied or the waiting time is wasted because the application time of the target AI model is too long (or too small); and/or, because the target AI model can trigger the first terminal to send data to the second terminal. A target feedback event used to feed back the application status of the target AI model. Therefore, the second terminal can directly send the AI model that the first terminal can apply to the first terminal based on the target feedback event without the need for the first terminal to communicate with the second terminal. Multiple transfers. This can reduce the number of transmissions between the first terminal and the second terminal in the process of obtaining the target AI model, thus saving transmission resources.
在一种可能的实现方式中,上述目标AI模型满足目标AI模型的目标传输模式,是根据第一终端支持的AI模型传输模式的能力确定的。In a possible implementation manner, the target AI model satisfying the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first terminal.
射频单元801,还用于从第一终端接收目标信息,该目标信息用于指示第一终端支持的AI模型传输模式的能力。The radio frequency unit 801 is also configured to receive target information from the first terminal, where the target information is used to indicate the capability of the AI model transmission mode supported by the first terminal.
处理器810,用于根据目标信息,确定目标传输模式。The processor 810 is configured to determine a target transmission mode according to the target information.
射频单元801,还用于基于目标传输模式,向第一终端发送目标模型信息。The radio frequency unit 801 is also used to send target model information to the first terminal based on the target transmission mode.
如此可知,由于第二终端可以从第一终端接收目标信息,并根据目标信息,确定目标传输模式,以按照目标传输模式向第一终端发送目标模型信息,因此,可以避免因第一终端不支持第一传输模式或第二传输模式,而导致第一设备无法得到目标AI模型的情况,进而导致无法应用目标AI模型的情况。It can be seen that since the second terminal can receive the target information from the first terminal, determine the target transmission mode according to the target information, and send the target model information to the first terminal according to the target transmission mode, it can avoid the problem that the first terminal does not support The first transmission mode or the second transmission mode causes the first device to be unable to obtain the target AI model, which further leads to the situation that the target AI model cannot be applied.
可选地,本申请实施例中,上述目标传输模式为第一传输模式。Optionally, in this embodiment of the present application, the target transmission mode is the first transmission mode.
处理器810,还用于确定目标AI模型和第一终端中的预存AI模型是否相匹配。The processor 810 is also used to determine whether the target AI model matches the pre-stored AI model in the first terminal.
射频单元801,还用于若目标AI模型和预存AI模型相匹配,则按照第四传输模式,向第一终端发送目标模型信息。The radio frequency unit 801 is also used to send the target model information to the first terminal according to the fourth transmission mode if the target AI model matches the pre-stored AI model.
其中,上述第四传输模式为以下任一项:第二传输模式、第五传输模式;在第四传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息;在第四传输模式为第五传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息,该第五传输模式对应的AI模型应用时间小于第一传输模式对应的AI模型应用时间。Wherein, the above-mentioned fourth transmission mode is any of the following: the second transmission mode or the fifth transmission mode; when the fourth transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model; in When the fourth transmission mode is the fifth transmission mode, the target model information includes model structure information and model parameter information of the target AI model. The AI model application time corresponding to the fifth transmission mode is shorter than the AI model application time corresponding to the first transmission mode. time.
如此可知,由于在目标AI模型和第一终端中的预存AI模型相匹配的情况下,第二终端可以按照第二传输模式向第一终端发送目标模型信息,即向第一终端发送目标AI模型的模型参数信息,而无需重复发送目标AI模型的模型结构信息,因此,可以节省传输资源;或者,第二终端可以按照第五传输模式向第一终端发送目标模型信息,以使得第一终端可以将较小的应用时间,确定为目标应用时间,因此,可以减少应用目标AI模型的时延。It can be seen that, when the target AI model matches the pre-stored AI model in the first terminal, the second terminal can send the target model information to the first terminal according to the second transmission mode, that is, send the target AI model to the first terminal. model parameter information without repeatedly sending the model structure information of the target AI model, thus saving transmission resources; or, the second terminal can send the target model information to the first terminal according to the fifth transmission mode, so that the first terminal can The smaller application time is determined as the target application time, so the delay in applying the target AI model can be reduced.
可选地,本申请实施例中,射频单元801,还用于从第一终端接收目标反馈事件。Optionally, in this embodiment of the present application, the radio frequency unit 801 is also used to receive a target feedback event from the first terminal.
其中,上述目标反馈事件用于反馈目标AI模型的应用情况。Among them, the above-mentioned target feedback event is used to feedback the application status of the target AI model.
如此可知,由于第二终端可以从第一终端接收用于反馈目标AI模型的应用情况的目标反馈事件,因此,第二终端可以直接根据目标反馈事件,向第一终端发送第一终端能够应用的AI模型,而无需第一终端与第二终端进行多次传输。It can be seen that since the second terminal can receive a target feedback event for feedback on the application status of the target AI model from the first terminal, the second terminal can directly send to the first terminal the information that the first terminal can apply based on the target feedback event. AI model without the need for multiple transmissions between the first terminal and the second terminal.
可选地,本申请实施例中,射频单元801,还用于从第一终端接收第二AI模型的第二模型信息,该第二模型信息是第一终端按照第三传输模式发送的。Optionally, in this embodiment of the present application, the radio frequency unit 801 is also configured to receive second model information of the second AI model from the first terminal, where the second model information is sent by the first terminal according to the third transmission mode.
处理器810,还用于将与第二AI模型的模型结构信息相匹配的模型结构信息,确定为目标AI模型的模型结构信息。The processor 810 is also configured to determine the model structure information that matches the model structure information of the second AI model as the model structure information of the target AI model.
其中,上述第三传输模式为第一传输模式或第四传输模式;在目标传输模式为第一传输模式的情况下,第二模型信息包括第二AI模型的模型结构信息和模型参数信息;在目标传输模式为第四传输模式的情况下,第二模型信息包括第二AI模型的模型结构信息。Wherein, the above-mentioned third transmission mode is the first transmission mode or the fourth transmission mode; when the target transmission mode is the first transmission mode, the second model information includes model structure information and model parameter information of the second AI model; in When the target transmission mode is the fourth transmission mode, the second model information includes model structure information of the second AI model.
如此可知,由于目标AI模型和第二AI模型的模型结构信息相匹配,因此,第一终端可以直接从第二终端接收目标AI模型的模型参数信息,以接收目标AI模型,而无需重复接收目标AI模型的模型结构信息,如此,可以节省传输资源。
It can be seen that since the model structure information of the target AI model and the second AI model match, the first terminal can directly receive the model parameter information of the target AI model from the second terminal to receive the target AI model without repeatedly receiving the target. The model structure information of the AI model can save transmission resources.
可选地,本申请实施例中,处理器810,具体用于以下任一项:Optionally, in this embodiment of the present application, the processor 810 is specifically used for any of the following:
将与第二AI模型的模型结构信息相同的模型结构信息,确定为目标AI模型的模型结构信息;Determine the model structure information that is the same as the model structure information of the second AI model as the model structure information of the target AI model;
将与第二AI模型的模型结构信息的相似度大于或等于预设门限的模型结构,确定为目标AI模型的模型结构信息。The model structure whose similarity to the model structure information of the second AI model is greater than or equal to the preset threshold is determined as the model structure information of the target AI model.
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,通信接口用于从第二设备接收目标AI模型的目标模型信息,处理器用于根据目标模型信息,得到目标AI模型;或者,用于向第一设备发送目标AI模型的目标模型信息;该目标模型信息用于第一设备得到目标AI模型;该目标AI模型满足以下至少一项:目标AI模型的目标传输模式,是根据第一设备支持的AI模型传输模式的能力确定的;目标AI模型的目标应用时间,是根据目标传输模式确定的;目标AI模型触发第一设备向第二设备发送目标反馈事件,该目标反馈事件用于反馈目标AI模型的应用情况。其中,上述目标传输模式包括以下任一项:第一传输模式、第二传输模式;在目标传输模式为第一传输模式的情况下,目标模型信息包括目标AI模型的模型结构信息和模型参数信息;在目标传输模式为第二传输模式的情况下,目标模型信息包括目标AI模型的模型参数信息。该网络侧设备实施例与上述网络侧设备方法实施例对应,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。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 target model information of the target AI model from the second device, and the processor is used to obtain the target AI model according to the target model information; or , used to send the target model information of the target AI model to the first device; the target model information is used by the first device to obtain the target AI model; the target AI model satisfies at least one of the following: the target transmission mode of the target AI model is based on The capability of the AI model transmission mode supported by the first device is determined; the target application time of the target AI model is determined based on the target transmission mode; the target AI model triggers the first device to send a target feedback event to the second device, and the target feedback event Used to provide feedback on the application of the target AI model. Wherein, the above-mentioned target transmission mode includes any of the following: a first transmission mode, a second transmission mode; when the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model. ; When the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model. This network-side device embodiment corresponds to the above-mentioned network-side device 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.
具体地,本申请实施例还提供了一种网络侧设备。如图11所示,该网络侧设备900包括:天线901、射频装置902、基带装置903、处理器904和存储器905。天线901与射频装置902连接。在上行方向上,射频装置902通过天线901接收信息,将接收的信息发送给基带装置903进行处理。在下行方向上,基带装置903对要发送的信息进行处理,并发送给射频装置902,射频装置902对收到的信息进行处理后经过天线901发送出去。Specifically, the embodiment of the present application also provides a network side device. As shown in Figure 11, the network side device 900 includes: an antenna 901, a radio frequency device 902, a baseband device 903, a processor 904 and a memory 905. The antenna 901 is connected to the radio frequency device 902 . In the uplink direction, the radio frequency device 902 receives information through the antenna 901 and sends the received information to the baseband device 903 for processing. In the downlink direction, the baseband device 903 processes the information to be sent and sends it to the radio frequency device 902. The radio frequency device 902 processes the received information and then sends it out through the antenna 901.
以上实施例中网络侧设备执行的方法可以在基带装置903中实现,该基带装置903包括基带处理器。The method performed by the network side device in the above embodiment can be implemented in the baseband device 903, which includes a baseband processor.
基带装置903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11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905连接,以调用存储器905中的程序,执行以上方法实施例中所示的网络设备操作。The baseband device 903 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.
该网络侧设备还可以包括网络接口906,该接口例如为通用公共无线接口(common public radio interface,CPRI)。The network side device may also include a network interface 906, which is, for example, a common public radio interface (CPRI).
具体地,本申请实施例的网络侧设备900还包括:存储在存储器905上并可在处理器904上运行的指令或程序,处理器904调用存储器905中的指令或程序执行图11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。Specifically, the network side device 900 in the embodiment of the present application also includes: instructions or programs stored in the memory 905 and executable on the processor 904. The processor 904 calls the instructions or programs in the memory 905 to execute each of the steps shown in Figure 11. The method of module execution and achieving the same technical effect will not be described in detail here to avoid duplication.
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AI模型传输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。An embodiment of the present application also provides a readable storage medium, on which a program or instruction is stored. When the program or instruction is executed by a processor, the various processes of the above-mentioned AI model transmission method embodiment are implemented, and the same technical effect can be achieved. To avoid repetition, it will not be repeated here.
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。Wherein, 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.
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AI模型传输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。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 embodiment of the AI model transmission method. Each process can achieve the same technical effect. To avoid repetition, we will not go into details here.
应理解,本申请实施例提到的芯片还可以称为***级芯片,***芯片,芯片***或片上***芯片等。It should be understood that the 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.
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AI模型传输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。Embodiments of the present application further provide a computer program/program product. The computer program/program product is stored in a storage medium. The computer program/program product is executed by at least one processor to implement the above-mentioned AI model transmission method. Each process in the example can achieve the same technical effect. To avoid repetition, we will not repeat it here.
本申请实施例还提供了一种AI模型传输***,包括:第一设备及第二设备,该第一设备可用于执行如上所述的AI模型传输方法的步骤,该第二设备可用于执行如上所述的AI模型传输方法的步骤。Embodiments of the present application also provide an AI model transmission system, including: a first device and a second device. The first device can be used to perform the steps of the AI model transmission method as described above. The second device can be used to perform the above steps. The steps of the AI model transmission method.
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。It should be noted that, in this document, the terms "comprising", "comprises" or any other variations thereof are intended to cover a non-exclusive inclusion, such that a process, method, article or device that includes a series of elements not only includes those elements, It also includes other elements not expressly listed or inherent in the process, method, article or apparatus. Without further limitation, an element defined by the statement "comprises a..." does not exclude the presence of additional identical elements in a process, method, article or apparatus that includes that element. In addition, it should be pointed out that the scope of the methods and devices in the embodiments of the present application is not limited to performing functions in the order shown or discussed, but may also include performing functions in a substantially simultaneous manner or in reverse order according to the functions involved. Functions may be performed, for example, the methods described may be performed in an order different from that described, and various steps may be added, omitted, or combined. Additionally, features described with reference to certain examples may be combined in other examples.
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。Through the above description of the embodiments, those skilled in the art can clearly understand that 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. Based on this understanding, 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.
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。
The embodiments of the present application have been described above in conjunction with the accompanying drawings. However, the present application is not limited to the above-mentioned specific implementations. The above-mentioned specific implementations are only illustrative and not restrictive. Those of ordinary skill in the art will Inspired by this application, many forms can be made without departing from the purpose of this application and the scope protected by the claims, all of which fall within the protection of this application.
Claims (39)
- 一种人工智能AI模型传输方法,其中,包括:An artificial intelligence AI model transmission method, which includes:第一设备从第二设备接收目标AI模型的目标模型信息;The first device receives target model information of the target AI model from the second device;所述第一设备根据所述目标模型信息,得到所述目标AI模型;The first device obtains the target AI model based on the target model information;所述目标AI模型满足以下至少一项:The target AI model meets at least one of the following:所述目标AI模型的目标传输模式,是根据所述第一设备支持的AI模型传输模式的能力确定的;The target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first device;所述目标AI模型的目标应用时间,是根据所述目标传输模式确定的;The target application time of the target AI model is determined based on the target transmission mode;所述目标AI模型触发所述第一设备向所述第二设备发送目标反馈事件,所述目标反馈事件用于反馈所述目标AI模型的应用情况;The target AI model triggers the first device to send a target feedback event to the second device, and the target feedback event is used to feed back the application status of the target AI model;其中,所述目标传输模式包括以下任一项:第一传输模式、第二传输模式;Wherein, the target transmission mode includes any one of the following: a first transmission mode, a second transmission mode;在所述目标传输模式为所述第一传输模式的情况下,所述目标模型信息包括所述目标AI模型的模型结构信息和模型参数信息;在所述目标传输模式为所述第二传输模式的情况下,所述目标模型信息包括所述目标AI模型的模型参数信息。When the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model; when the target transmission mode is the second transmission mode In the case of , the target model information includes model parameter information of the target AI model.
- 根据权利要求1所述的方法,其中,所述目标AI模型满足所述目标AI模型的目标传输模式,是根据所述第一设备支持的AI模型传输模式的能力确定的;所述方法还包括:The method according to claim 1, wherein the target AI model satisfying the target transmission mode of the target AI model is determined according to the capability of the AI model transmission mode supported by the first device; the method further includes :所述第一设备向所述第二设备发送目标信息,所述目标信息用于指示所述第一设备支持的AI模型传输模式的能力。The first device sends target information to the second device, where the target information is used to indicate the capability of the AI model transmission mode supported by the first device.
- 根据权利要求1或2所述的方法,其中,所述第一设备支持的AI模型传输模式的能力包括以下任一项:The method according to claim 1 or 2, wherein the capability of the AI model transmission mode supported by the first device includes any of the following:支持所述第一传输模式;Support the first transmission mode;支持所述第二传输模式;Support the second transmission mode;支持所述第一传输模式和所述第二传输模式。The first transmission mode and the second transmission mode are supported.
- 根据权利要求1所述的方法,其中,所述目标AI模型满足所述目标AI模型的目标应用时间,是根据所述目标传输模式确定的;所述方法还包括:The method according to claim 1, wherein the target application time of the target AI model that satisfies the target AI model is determined according to the target transmission mode; the method further includes:所述第一设备根据所述目标传输模式,确定所述目标应用时间。The first device determines the target application time according to the target transmission mode.
- 根据权利要求4所述的方法,其中,所述第一设备根据所述目标传输模式,确定所述目标应用时间,包括:The method of claim 4, wherein the first device determines the target application time according to the target transmission mode, including:所述第一设备将所述目标传输模式对应的应用时间,确定为所述目标应用时间;The first device determines the application time corresponding to the target transmission mode as the target application time;第一应用时间和第二应用时间之间满足以下至少一项:At least one of the following conditions is satisfied between the first application time and the second application time:所述第一应用时间大于所述第二应用时间;The first application time is greater than the second application time;所述第一应用时间大于或等于第三应用时间,所述第三应用时间是根据所述第二应用时间和AI模型的编译时间确定的;The first application time is greater than or equal to the third application time, and the third application time is determined based on the second application time and the compilation time of the AI model;其中,所述第一应用时间为:所述第一传输模式对应的AI模型应用时间;所述第二应用时间为:所述第二传输模式对应的AI模型应用时间。Wherein, the first application time is: the AI model application time corresponding to the first transmission mode; the second application time is: the AI model application time corresponding to the second transmission mode.
- 根据权利要求4所述的方法,其中,所述第一设备根据所述目标传输模式,确定所述目标应用时间,包括:The method of claim 4, wherein the first device determines the target application time according to the target transmission mode, including:在所述目标传输模式为所述第一传输模式的情况下,若所述目标AI模型和所述第一设备中的预存AI模型相匹配,则所述第一设备将第四应用时间确定为所述目标应用时间;When the target transmission mode is the first transmission mode, if the target AI model matches the pre-stored AI model in the first device, the first device determines the fourth application time as The target application time;其中,所述第四应用时间小于第一应用时间;所述第一应用时间为:所述第一传输模式对应的AI模型应用时间。Wherein, the fourth application time is less than the first application time; the first application time is: the AI model application time corresponding to the first transmission mode.
- 根据权利要求1所述的方法,其中,所述目标AI模型满足所述目标AI模型触发所述第一设备向所述第二设备发送目标反馈事件;The method according to claim 1, wherein the target AI model satisfies the target AI model to trigger the first device to send a target feedback event to the second device;所述目标反馈事件为以下任一项:第一反馈事件、第二反馈事件;The target feedback event is any one of the following: a first feedback event, a second feedback event;其中,所述第一反馈事件用于反馈未应用所述目标AI模型;所述第二反馈事件用于反馈已应用所述目标AI模型。The first feedback event is used to feedback that the target AI model has not been applied; the second feedback event is used to feedback that the target AI model has been applied.
- 根据权利要求7所述的方法,其中,所述第一反馈事件为以下任一项:The method of claim 7, wherein the first feedback event is any of the following:所述第一设备未应用所述目标AI模型;The first device does not apply the target AI model;所述第一设备未应用所述目标AI模型,且所述第一设备仍应用第一AI模型;The first device does not apply the target AI model, and the first device still applies the first AI model;所述第一设备未应用所述目标AI模型,且所述第一设备不应用AI模型; The first device does not apply the target AI model, and the first device does not apply the AI model;所述第一设备不支持所述第二设备通过所述目标传输模式发送的所述目标模型信息;The first device does not support the target model information sent by the second device through the target transmission mode;第三反馈事件;Third feedback event;所述第一设备不支持所述目标AI模型的第一模型信息;The first device does not support the first model information of the target AI model;其中,所述第一AI模型为:所述第一设备接收所述目标AI模型之前应用的AI模型;Wherein, the first AI model is: the AI model applied before the first device receives the target AI model;所述第三反馈事件为:所述第一设备不支持所述第二设备通过所述目标传输模式发送的所述目标模型信息;所述第三反馈事件携带第一信息,所述第一信息为所述目标AI模型中的所述第一设备不支持的模型结构的信息;The third feedback event is: the first device does not support the target model information sent by the second device through the target transmission mode; the third feedback event carries first information, and the first information Information about the model structure that is not supported by the first device in the target AI model;所述第一模型信息包括以下至少一项:模型大小、模型复杂度、模型操作数。The first model information includes at least one of the following: model size, model complexity, and model operation number.
- 根据权利要求7所述的方法,其中,所述第二反馈事件为以下任一项:The method of claim 7, wherein the second feedback event is any of the following:所述第一设备已应用所述目标AI模型;The first device has applied the target AI model;第四反馈事件;The fourth feedback event;所述第一设备已应用所述目标AI模型,且已替换第一AI模型;The first device has applied the target AI model and replaced the first AI model;所述第一设备支持所述第二设备通过所述目标传输模式发送的所述目标模型信息;The first device supports the target model information sent by the second device through the target transmission mode;所述第一设备已编译所述目标AI模型;The first device has compiled the target AI model;其中,所述第四反馈事件为:所述第一设备已应用所述目标AI模型;所述第四反馈事件携带第二信息,所述第二信息为所述目标AI模型的模型标识信息;Wherein, the fourth feedback event is: the first device has applied the target AI model; the fourth feedback event carries second information, and the second information is model identification information of the target AI model;所述第一AI模型为:所述第一设备接收所述目标AI模型之前应用的AI模型。The first AI model is: an AI model applied before the first device receives the target AI model.
- 根据权利要求1所述的方法,其中,所述方法还包括:The method of claim 1, further comprising:在所述目标AI模型与所述第一设备的能力不匹配的情况下,所述第一设备不应用所述目标AI模型。In the case where the target AI model does not match the capabilities of the first device, the first device does not apply the target AI model.
- 根据权利要求1所述的方法,其中,所述方法还包括:The method of claim 1, further comprising:所述第一设备按照第三传输模式,向所述第二设备发送第二AI模型的第二模型信息;The first device sends the second model information of the second AI model to the second device according to the third transmission mode;其中,所述第三传输模式为所述第一传输模式或第四传输模式;Wherein, the third transmission mode is the first transmission mode or the fourth transmission mode;在所述目标传输模式为所述第一传输模式的情况下,所述第二模型信息包括所述第二AI模型的模型结构信息和模型参数信息;在所述目标传输模式为所述第四传输模式的情况下,所述第二模型信息包括所述第二AI模型的模型结构信息;When the target transmission mode is the first transmission mode, the second model information includes model structure information and model parameter information of the second AI model; when the target transmission mode is the fourth In the case of transmission mode, the second model information includes model structure information of the second AI model;所述目标AI模型与所述第二AI模型的模型结构信息相匹配。The target AI model matches the model structure information of the second AI model.
- 根据权利要求1所述的方法,其中,所述方法还包括:The method of claim 1, further comprising:所述第一设备采用所述目标AI模型,执行第一操作;The first device uses the target AI model to perform the first operation;其中,所述第一操作包括以下至少一项:Wherein, the first operation includes at least one of the following:信号处理操作;signal processing operations;信号传输操作;signal transmission operations;信号解调操作;signal demodulation operation;信道状态信息获取操作;Channel status information acquisition operation;波束管理操作;beam management operations;信道预测操作;Channel prediction operation;干扰抑制操作;interference suppression operations;定位操作;Positioning operations;高层业务的预测和管理操作;High-level business forecasting and management operations;高层参数的预测和管理操作;Prediction and management operations of high-level parameters;控制信令解析操作。Control signaling parsing operations.
- 一种AI模型传输方法,其中,包括:An AI model transmission method, including:第二设备向第一设备发送目标AI模型的目标模型信息;The second device sends the target model information of the target AI model to the first device;所述目标模型信息用于所述第一设备得到所述目标AI模型;The target model information is used by the first device to obtain the target AI model;所述目标AI模型满足以下至少一项:The target AI model meets at least one of the following:所述目标AI模型的目标传输模式,是根据所述第一设备支持的AI模型传输模式的能力确定的;The target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first device;所述目标AI模型的目标应用时间,是根据所述目标传输模式确定的;The target application time of the target AI model is determined based on the target transmission mode;所述目标AI模型触发所述第一设备向所述第二设备发送目标反馈事件,所述目标反馈事件用于反馈所述目标AI模型的应用情况;The target AI model triggers the first device to send a target feedback event to the second device, and the target feedback event is used to feed back the application status of the target AI model;其中,所述目标传输模式包括以下任一项:第一传输模式、第二传输模式; Wherein, the target transmission mode includes any one of the following: a first transmission mode, a second transmission mode;在所述目标传输模式为所述第一传输模式的情况下,所述目标模型信息包括所述目标AI模型的模型结构信息和模型参数信息;在所述目标传输模式为所述第二传输模式的情况下,所述目标模型信息包括所述目标AI模型的模型参数信息。When the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model; when the target transmission mode is the second transmission mode In the case of , the target model information includes model parameter information of the target AI model.
- 根据权利要求13所述的方法,其中,所述目标AI模型满足所述目标AI模型的目标传输模式,是根据所述第一设备支持的AI模型传输模式的能力确定的;所述方法还包括:The method according to claim 13, wherein the target AI model satisfying the target transmission mode of the target AI model is determined according to the capability of the AI model transmission mode supported by the first device; the method further includes :所述第二设备从所述第一设备接收目标信息,所述目标信息用于指示所述第一设备支持的AI模型传输模式的能力;The second device receives target information from the first device, the target information being used to indicate the capabilities of the AI model transmission mode supported by the first device;所述第二设备根据所述目标信息,确定目标传输模式;The second device determines a target transmission mode according to the target information;所述第二设备基于所述目标传输模式,向所述第一设备发送所述目标模型信息。The second device sends the target model information to the first device based on the target transmission mode.
- 根据权利要求14所述的方法,其中,所述目标传输模式为所述第一传输模式;所述方法还包括:The method according to claim 14, wherein the target transmission mode is the first transmission mode; the method further includes:所述第二设备确定所述目标AI模型和所述第一设备中的预存AI模型是否相匹配;The second device determines whether the target AI model matches the pre-stored AI model in the first device;若所述目标AI模型和所述预存AI模型相匹配,则所述第二设备按照第四传输模式,向所述第一设备发送所述目标模型信息;If the target AI model matches the pre-stored AI model, the second device sends the target model information to the first device according to the fourth transmission mode;其中,所述第四传输模式为以下任一项:第二传输模式、第五传输模式;Wherein, the fourth transmission mode is any one of the following: a second transmission mode, a fifth transmission mode;在所述第四传输模式为所述第二传输模式的情况下,所述目标模型信息包括所述目标AI模型的模型参数信息;在所述第四传输模式为所述第五传输模式的情况下,所述目标模型信息包括所述目标AI模型的模型结构信息和模型参数信息;When the fourth transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model; when the fourth transmission mode is the fifth transmission mode Next, the target model information includes model structure information and model parameter information of the target AI model;所述第五传输模式对应的AI模型应用时间小于所述第一传输模式对应的AI模型应用时间。The AI model application time corresponding to the fifth transmission mode is less than the AI model application time corresponding to the first transmission mode.
- 根据权利要求14所述的方法,其中,所述方法还包括:The method of claim 14, wherein the method further includes:所述第二设备从所述第一设备接收目标反馈事件;The second device receives a target feedback event from the first device;其中,所述目标反馈事件用于反馈所述目标AI模型的应用情况。Wherein, the target feedback event is used to feedback the application status of the target AI model.
- 根据权利要求14所述的方法,其中,所述方法还包括:The method of claim 14, further comprising:所述第二设备从所述第一设备接收第二AI模型的第二模型信息,所述第二模型信息是所述第一设备按照第三传输模式发送的;The second device receives second model information of the second AI model from the first device, where the second model information is sent by the first device according to a third transmission mode;所述第二设备将与所述第二AI模型的模型结构信息相匹配的模型结构信息,确定为所述目标AI模型的模型结构信息;The second device determines the model structure information that matches the model structure information of the second AI model as the model structure information of the target AI model;其中,所述第三传输模式为所述第一传输模式或第四传输模式;Wherein, the third transmission mode is the first transmission mode or the fourth transmission mode;在所述目标传输模式为所述第一传输模式的情况下,所述第二模型信息包括所述第二AI模型的模型结构信息和模型参数信息;在所述目标传输模式为所述第四传输模式的情况下,所述第二模型信息包括所述第二AI模型的模型结构信息。When the target transmission mode is the first transmission mode, the second model information includes model structure information and model parameter information of the second AI model; when the target transmission mode is the fourth In the case of transmission mode, the second model information includes model structure information of the second AI model.
- 根据权利要求17所述的方法,其中,所述第二设备将与所述第二AI模型的模型结构信息相匹配的模型结构信息,确定为所述目标AI模型的模型结构信息,包括以下任一项:The method according to claim 17, wherein the second device determines model structure information matching the model structure information of the second AI model as the model structure information of the target AI model, including any of the following: One item:所述第二设备将与所述第二AI模型的模型结构信息相同的模型结构信息,确定为所述目标AI模型的模型结构信息;The second device determines the model structure information that is the same as the model structure information of the second AI model as the model structure information of the target AI model;所述第二设备将与所述第二AI模型的模型结构信息的相似度大于或等于预设门限的模型结构信息,确定为所述目标AI模型的模型结构信息。The second device determines the model structure information whose similarity to the model structure information of the second AI model is greater than or equal to a preset threshold as the model structure information of the target AI model.
- 一种AI模型传输装置,其中,所述AI模型传输装置为第一AI模型传输装置,所述第一AI模型传输装置包括:接收模块和处理模块;An AI model transmission device, wherein the AI model transmission device is a first AI model transmission device, and the first AI model transmission device includes: a receiving module and a processing module;所述接收模块,用于从第二AI模型传输装置接收目标AI模型的目标模型信息;The receiving module is configured to receive target model information of the target AI model from the second AI model transmission device;所述处理模块,用于根据所述接收模块接收的所述目标模型信息,得到所述目标AI模型;The processing module is used to obtain the target AI model according to the target model information received by the receiving module;所述目标AI模型满足以下至少一项:The target AI model meets at least one of the following:所述目标AI模型的目标传输模式,是根据所述第一AI模型传输装置支持的AI模型传输模式的能力确定的;The target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device;所述目标AI模型的目标应用时间,是根据所述目标传输模式确定的;The target application time of the target AI model is determined based on the target transmission mode;所述目标AI模型触发所述第一AI模型传输装置向所述第二AI模型传输装置发送目标反馈事件,所述目标反馈事件用于反馈所述目标AI模型的应用情况;The target AI model triggers the first AI model transmission device to send a target feedback event to the second AI model transmission device, and the target feedback event is used to feed back the application status of the target AI model;其中,所述目标传输模式包括以下任一项:第一传输模式、第二传输模式;Wherein, the target transmission mode includes any one of the following: a first transmission mode, a second transmission mode;在所述目标传输模式为所述第一传输模式的情况下,所述目标模型信息包括所述目标 AI模型的模型结构信息和模型参数信息;在所述目标传输模式为所述第二传输模式的情况下,所述目标模型信息包括所述目标AI模型的模型参数信息。When the target transmission mode is the first transmission mode, the target model information includes the target Model structure information and model parameter information of the AI model; when the target transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model.
- 根据权利要求19所述的AI模型传输装置,其中,所述目标AI模型满足所述目标AI模型的目标传输模式,是根据所述第一AI模型传输装置支持的AI模型传输模式的能力确定的;所述第一AI模型传输装置还包括:发送模块;The AI model transmission device according to claim 19, wherein the target AI model satisfying the target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device. ;The first AI model transmission device further includes: a sending module;所述发送模块,用于向所述第二AI模型传输装置发送目标信息,所述目标信息用于指示所述第一AI模型传输装置支持的AI模型传输模式的能力。The sending module is configured to send target information to the second AI model transmission device, where the target information is used to indicate the capabilities of the AI model transmission mode supported by the first AI model transmission device.
- 根据权利要求19或20所述的AI模型传输装置,其中,所述第一AI模型传输装置支持的AI模型传输模式的能力包括以下任一项:The AI model transmission device according to claim 19 or 20, wherein the capabilities of the AI model transmission mode supported by the first AI model transmission device include any of the following:支持所述第一传输模式;Support the first transmission mode;支持所述第二传输模式;Support the second transmission mode;支持所述第一传输模式和所述第二传输模式。The first transmission mode and the second transmission mode are supported.
- 根据权利要求19所述的AI模型传输装置,其中,所述目标AI模型满足所述目标AI模型的目标应用时间,是根据所述目标传输模式确定的;所述第一AI模型传输装置还包括:确定模块;The AI model transmission device according to claim 19, wherein the target AI model satisfying the target application time of the target AI model is determined according to the target transmission mode; the first AI model transmission device further includes : Determine the module;所述确定模块,用于根据所述目标传输模式,确定所述目标应用时间。The determining module is configured to determine the target application time according to the target transmission mode.
- 根据权利要求22所述的AI模型传输装置,其中,The AI model transmission device according to claim 22, wherein,所述确定模块,具体用于将所述目标传输模式对应的应用时间,确定为所述目标应用时间;The determination module is specifically configured to determine the application time corresponding to the target transmission mode as the target application time;第一应用时间和第二应用时间之间满足以下至少一项:At least one of the following conditions is satisfied between the first application time and the second application time:所述第一应用时间大于所述第二应用时间;The first application time is greater than the second application time;所述第一应用时间大于或等于第三应用时间,所述第三应用时间是根据所述第二应用时间和AI模型的编译时间确定的;The first application time is greater than or equal to a third application time, and the third application time is determined according to the second application time and the compilation time of the AI model;其中,所述第一应用时间为:所述第一传输模式对应的AI模型应用时间;所述第二应用时间为:所述第二传输模式对应的AI模型应用时间。Wherein, the first application time is: the AI model application time corresponding to the first transmission mode; the second application time is: the AI model application time corresponding to the second transmission mode.
- 根据权利要求22所述的AI模型传输装置,其中,The AI model transmission device according to claim 22, wherein,所述确定模块,具体用于在所述目标传输模式为所述第一传输模式的情况下,若所述目标AI模型和所述第一AI模型传输装置中的预存AI模型相匹配,则将第四应用时间确定为所述目标应用时间;The determination module is specifically configured to, when the target transmission mode is the first transmission mode, if the target AI model matches the pre-stored AI model in the first AI model transmission device, then The fourth application time is determined as the target application time;其中,所述第四应用时间小于第一应用时间;所述第一应用时间为:所述第一传输模式对应的AI模型应用时间。Among them, the fourth application time is less than the first application time; the first application time is: the AI model application time corresponding to the first transmission mode.
- 根据权利要求19所述的AI模型传输装置,其中,所述目标AI模型满足所述目标AI模型触发所述第一AI模型传输装置向所述第二AI模型传输装置发送目标反馈事件;The AI model transmission device according to claim 19, wherein the target AI model satisfies the target AI model to trigger the first AI model transmission device to send a target feedback event to the second AI model transmission device;所述目标反馈事件为以下任一项:第一反馈事件、第二反馈事件;The target feedback event is any one of the following: a first feedback event, a second feedback event;其中,所述第一反馈事件用于反馈未应用所述目标AI模型;所述第二反馈事件用于反馈已应用所述目标AI模型。The first feedback event is used to feedback that the target AI model has not been applied; the second feedback event is used to feedback that the target AI model has been applied.
- 根据权利要求25所述的AI模型传输装置,其中,所述第一反馈事件为以下任一项:The AI model transmission device according to claim 25, wherein the first feedback event is any one of the following:所述第一AI模型传输装置未应用所述目标AI模型;The first AI model transmission device does not apply the target AI model;所述第一AI模型传输装置未应用所述目标AI模型,且所述第一AI模型传输装置仍应用第一AI模型;The first AI model transmission device does not apply the target AI model, and the first AI model transmission device still applies the first AI model;所述第一AI模型传输装置未应用所述目标AI模型,且所述第一AI模型传输装置不应用AI模型;The first AI model transmission device does not apply the target AI model, and the first AI model transmission device does not apply the AI model;所述第一AI模型传输装置不支持所述第二AI模型传输装置通过所述目标传输模式发送的所述目标模型信息;The first AI model transmission device does not support the target model information sent by the second AI model transmission device through the target transmission mode;第三反馈事件;Third feedback event;所述第一AI模型传输装置不支持所述目标AI模型的第一模型信息;The first AI model transmission device does not support the first model information of the target AI model;其中,所述第一AI模型为:所述第一AI模型传输装置接收所述目标AI模型之前应用的AI模型;Wherein, the first AI model is: the AI model applied before the first AI model transmission device receives the target AI model;所述第三反馈事件为:所述第一AI模型传输装置不支持所述第二AI模型传输装置通过所述目标传输模式发送的所述目标模型信息;所述第三反馈事件携带第一信息,所述第一信息为所述目标AI模型中的所述第一AI模型传输装置不支持的模型结构的信息; The third feedback event is: the first AI model transmission device does not support the target model information sent by the second AI model transmission device through the target transmission mode; the third feedback event carries the first information , the first information is information about a model structure that is not supported by the first AI model transmission device in the target AI model;所述第一模型信息包括以下至少一项:模型大小、模型复杂度、模型操作数。The first model information includes at least one of the following: model size, model complexity, and model operation number.
- 根据权利要求25所述的AI模型传输装置,其中,所述第二反馈事件为以下任一项:The AI model transmission device according to claim 25, wherein the second feedback event is any one of the following:所述第一AI模型传输装置已应用所述目标AI模型;The first AI model transmission device has applied the target AI model;第四反馈事件;The fourth feedback event;所述第一AI模型传输装置已应用所述目标AI模型,且已替换第一AI模型;The first AI model transmission device has applied the target AI model and replaced the first AI model;所述第一AI模型传输装置支持所述第二AI模型传输装置通过所述目标传输模式发送的所述目标模型信息;The first AI model transmission device supports the target model information sent by the second AI model transmission device through the target transmission mode;所述第一AI模型传输装置已编译所述目标AI模型;The first AI model transmission device has compiled the target AI model;其中,所述第四反馈事件为:所述第一AI模型传输装置已应用所述目标AI模型;所述第四反馈事件携带第二信息,所述第二信息为所述目标AI模型的模型标识信息;Wherein, the fourth feedback event is: the first AI model transmission device has applied the target AI model; the fourth feedback event carries second information, and the second information is the model of the target AI model. identification information;所述第一AI模型为:所述第一AI模型传输装置接收所述目标AI模型之前应用的AI模型。The first AI model is: the AI model applied before the first AI model transmission device receives the target AI model.
- 根据权利要求19所述的AI模型传输装置,其中,所述处理模块,还用于在所述目标AI模型与所述第一AI模型传输装置的能力不匹配的情况下,不应用所述目标AI模型。The AI model transmission device according to claim 19, wherein the processing module is further configured to not apply the target AI model when the capabilities of the target AI model do not match the capabilities of the first AI model transmission device. AI model.
- 根据权利要求19所述的AI模型传输装置,其中,所述第一AI模型传输装置还包括:发送模块;The AI model transmission device according to claim 19, wherein the first AI model transmission device further comprises: a sending module;所述发送模块,用于按照第三传输模式,向所述第二AI模型传输装置发送第二AI模型的第二模型信息;The sending module is configured to send the second model information of the second AI model to the second AI model transmission device according to the third transmission mode;其中,所述第三传输模式为所述第一传输模式或第四传输模式;Wherein, the third transmission mode is the first transmission mode or the fourth transmission mode;在所述目标传输模式为所述第一传输模式的情况下,所述第二模型信息包括所述第二AI模型的模型结构信息和模型参数信息;在所述目标传输模式为所述第四传输模式的情况下,所述第二模型信息包括所述第二AI模型的模型结构信息;When the target transmission mode is the first transmission mode, the second model information includes model structure information and model parameter information of the second AI model; when the target transmission mode is the fourth In the case of transmission mode, the second model information includes model structure information of the second AI model;所述目标AI模型与所述第二AI模型的模型结构信息相匹配。The target AI model matches the model structure information of the second AI model.
- 根据权利要求19所述的AI模型传输装置,其中,所述处理模块,还用于采用所述目标AI模型,执行第一操作;The AI model transmission device according to claim 19, wherein the processing module is further used to use the target AI model to perform a first operation;其中,所述第一操作包括以下至少一项:Wherein, the first operation includes at least one of the following:信号处理操作;signal processing operations;信号传输操作;signal transmission operations;信号解调操作;signal demodulation operation;信道状态信息获取操作;Channel state information acquisition operation;波束管理操作;beam management operations;信道预测操作;Channel prediction operations;干扰抑制操作;interference suppression operations;定位操作;Positioning operations;高层业务的预测和管理操作;High-level business forecasting and management operations;高层参数的预测和管理操作;Prediction and management operations of high-level parameters;控制信令解析操作。Control signaling parsing operations.
- 一种AI模型传输装置,其中,所述AI模型传输装置为第二AI模型传输装置,所述第二AI模型传输装置包括:发送模块;An AI model transmission device, wherein the AI model transmission device is a second AI model transmission device, and the second AI model transmission device includes: a sending module;所述发送模块,用于向第一AI模型传输装置发送目标AI模型的目标模型信息;The sending module is used to send the target model information of the target AI model to the first AI model transmission device;所述目标模型信息用于所述第一AI模型传输装置得到所述目标AI模型;The target model information is used by the first AI model transmission device to obtain the target AI model;所述目标AI模型满足以下至少一项:The target AI model meets at least one of the following:所述目标AI模型的目标传输模式,是根据所述第一AI模型传输装置支持的AI模型传输模式的能力确定的;The target transmission mode of the target AI model is determined based on the capability of the AI model transmission mode supported by the first AI model transmission device;所述目标AI模型的目标应用时间,是根据所述目标传输模式确定的;The target application time of the target AI model is determined based on the target transmission mode;所述目标AI模型触发所述第一AI模型传输装置向所述第二AI模型传输装置发送目标反馈事件,所述目标反馈事件用于反馈所述目标AI模型的应用情况;The target AI model triggers the first AI model transmission device to send a target feedback event to the second AI model transmission device, and the target feedback event is used to feed back the application status of the target AI model;其中,所述目标传输模式包括以下任一项:第一传输模式、第二传输模式;Wherein, the target transmission mode includes any one of the following: a first transmission mode, a second transmission mode;在所述目标传输模式为所述第一传输模式的情况下,所述目标模型信息包括所述目标AI模型的模型结构信息和模型参数信息;在所述目标传输模式为所述第二传输模式的情况下,所述目标模型信息包括所述目标AI模型的模型参数信息。 When the target transmission mode is the first transmission mode, the target model information includes model structure information and model parameter information of the target AI model; when the target transmission mode is the second transmission mode In the case of , the target model information includes model parameter information of the target AI model.
- 根据权利要求31所述的AI模型传输装置,其中,所述目标AI模型满足所述目标AI模型的目标传输模式,是根据所述第一AI模型传输装置支持的AI模型传输模式的能力确定的;所述第二AI模型传输装置还包括:接收模块和确定模块;The AI model transmission device according to claim 31, wherein the target AI model satisfying the target transmission mode of the target AI model is determined according to the capability of the AI model transmission mode supported by the first AI model transmission device. ;The second AI model transmission device also includes: a receiving module and a determining module;所述接收模块,用于从所述第一AI模型传输装置接收目标信息,所述目标信息用于指示所述第一AI模型传输装置支持的AI模型传输模式的能力;The receiving module is configured to receive target information from the first AI model transmission device, where the target information is used to indicate the capabilities of the AI model transmission mode supported by the first AI model transmission device;所述确定模块,用于根据所述接收模块接收的所述目标信息,确定目标传输模式;The determining module is configured to determine the target transmission mode according to the target information received by the receiving module;所述发送模块,还用于基于所述确定模块确定的所述目标传输模式,向所述第一AI模型传输装置发送所述目标模型信息。The sending module is further configured to send the target model information to the first AI model transmission device based on the target transmission mode determined by the determining module.
- 根据权利要求32所述的AI模型传输装置,其中,所述目标传输模式为第一传输模式;The AI model transmission device according to claim 32, wherein the target transmission mode is a first transmission mode;所述确定模块,还用于确定所述目标AI模型和所述第一AI模型传输装置中的预存AI模型是否相匹配;The determination module is also used to determine whether the target AI model matches the pre-stored AI model in the first AI model transmission device;所述发送模块,还用于若所述确定模块确定所述目标AI模型和所述预存AI模型相匹配,则按照第四传输模式,向所述第一AI模型传输装置发送所述目标AI模型;The sending module is also configured to send the target AI model to the first AI model transmission device according to the fourth transmission mode if the determination module determines that the target AI model matches the pre-stored AI model. ;其中,所述第四传输模式为以下任一项:第二传输模式、第五传输模式;Wherein, the fourth transmission mode is any one of the following: a second transmission mode, a fifth transmission mode;在所述第四传输模式为所述第二传输模式的情况下,所述目标模型信息包括所述目标AI模型的模型参数信息;在所述第四传输模式为所述第五传输模式的情况下,所述目标模型信息包括所述目标AI模型的模型结构信息和模型参数信息;When the fourth transmission mode is the second transmission mode, the target model information includes model parameter information of the target AI model; when the fourth transmission mode is the fifth transmission mode Next, the target model information includes model structure information and model parameter information of the target AI model;所述第五传输模式对应的AI模型应用时间小于所述第一传输模式对应的AI模型应用时间。The AI model application time corresponding to the fifth transmission mode is less than the AI model application time corresponding to the first transmission mode.
- 根据权利要求32所述的AI模型传输装置,其中,The AI model transmission device according to claim 32, wherein,所述接收模块,还用于从所述第一AI模型传输装置接收目标反馈事件;The receiving module is also configured to receive target feedback events from the first AI model transmission device;其中,所述目标反馈事件用于反馈所述目标AI模型的应用情况。Wherein, the target feedback event is used to feedback the application status of the target AI model.
- 根据权利要求32所述的AI模型传输装置,其中,The AI model transmission device according to claim 32, wherein,所述接收模块,还用于从所述第一AI模型传输装置接收第二AI模型的第二模型信息,所述第二模型信息是所述第一AI模型传输装置按照第三传输模式发送的;The receiving module is also configured to receive second model information of the second AI model from the first AI model transmission device, where the second model information is sent by the first AI model transmission device according to the third transmission mode. ;所述确定模块,还用于将与所述接收模块接收的所述第二AI模型的模型结构信息相匹配的模型结构信息,确定为所述目标AI模型的模型结构信息;The determining module is also configured to determine the model structure information that matches the model structure information of the second AI model received by the receiving module as the model structure information of the target AI model;其中,所述第三传输模式为所述第一传输模式或第四传输模式;Wherein, the third transmission mode is the first transmission mode or the fourth transmission mode;在所述目标传输模式为所述第一传输模式的情况下,所述第二模型信息包括所述第二AI模型的模型结构信息和模型参数信息;在所述目标传输模式为所述第四传输模式的情况下,所述第二模型信息包括所述第二AI模型的模型结构信息。When the target transmission mode is the first transmission mode, the second model information includes model structure information and model parameter information of the second AI model; when the target transmission mode is the fourth In the case of transmission mode, the second model information includes model structure information of the second AI model.
- 根据权利要求35所述的AI模型传输装置,其中,The AI model transmission device according to claim 35, wherein,所述确定模块,具体用于以下任一项:The determination module is specifically used for any of the following:将与所述第二AI模型的模型结构信息相同的模型结构信息,确定为所述目标AI模型的模型结构信息;Determine the model structure information that is the same as the model structure information of the second AI model as the model structure information of the target AI model;将与所述第二AI模型的模型结构信息的相似度大于或等于预设门限的模型结构信息,确定为所述目标AI模型的模型结构信息。Model structure information whose similarity with the model structure information of the second AI model is greater than or equal to a preset threshold is determined as the model structure information of the target AI model.
- 一种终端,其中,包括处理器和存储器,所述存储器存储可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如权利要求1至18中任一项所述的AI模型传输方法的步骤。A terminal, which includes a processor and a memory. The memory stores programs or instructions that can be run on the processor. When the program or instructions are executed by the processor, any one of claims 1 to 18 is implemented. The steps of the AI model transmission method described in one item.
- 一种网络侧设备,其中,包括处理器和存储器,所述存储器存储可在所述处理器上运行的程序或指令,所述程序或指令被所述处理器执行时实现如权利要求1至18中任一项所述的AI模型传输方法的步骤。A network side device, which includes a processor and a memory. The memory stores programs or instructions that can be run on the processor. When the program or instructions are executed by the processor, the implementation of claims 1 to 18 is achieved. The steps of the AI model transmission method described in any one of the above.
- 一种可读存储介质,其中,所述可读存储介质上存储程序或指令,所述程序或指令被处理器执行时实现如权利要求1至18中任一项所述的AI模型传输方法的步骤。 A readable storage medium, wherein programs or instructions are stored on the readable storage medium, and when the programs or instructions are executed by a processor, the AI model transmission method as described in any one of claims 1 to 18 is implemented. step.
Applications Claiming Priority (2)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202211167997.0A CN117835405A (en) | 2022-09-23 | 2022-09-23 | Artificial intelligent AI model transmission method, device, terminal and medium |
CN202211167997.0 | 2022-09-23 |
Publications (1)
Publication Number | Publication Date |
---|---|
WO2024061287A1 true WO2024061287A1 (en) | 2024-03-28 |
Family
ID=90453866
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
PCT/CN2023/120138 WO2024061287A1 (en) | 2022-09-23 | 2023-09-20 | Artificial intelligence (ai) model transmission method and apparatus, and terminal and medium |
Country Status (2)
Country | Link |
---|---|
CN (1) | CN117835405A (en) |
WO (1) | WO2024061287A1 (en) |
Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021057807A1 (en) * | 2019-09-23 | 2021-04-01 | Oppo广东移动通信有限公司 | Deep learning model generation method and apparatus, device, and storage medium |
WO2021120177A1 (en) * | 2019-12-20 | 2021-06-24 | 华为技术有限公司 | Method and apparatus for compiling neural network model |
WO2022041285A1 (en) * | 2020-08-31 | 2022-03-03 | 华为技术有限公司 | Model data transmission method and communication apparatus |
CN114399019A (en) * | 2021-12-30 | 2022-04-26 | 南京风兴科技有限公司 | Neural network compiling method, system, computer device and storage medium |
-
2022
- 2022-09-23 CN CN202211167997.0A patent/CN117835405A/en active Pending
-
2023
- 2023-09-20 WO PCT/CN2023/120138 patent/WO2024061287A1/en unknown
Patent Citations (4)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
WO2021057807A1 (en) * | 2019-09-23 | 2021-04-01 | Oppo广东移动通信有限公司 | Deep learning model generation method and apparatus, device, and storage medium |
WO2021120177A1 (en) * | 2019-12-20 | 2021-06-24 | 华为技术有限公司 | Method and apparatus for compiling neural network model |
WO2022041285A1 (en) * | 2020-08-31 | 2022-03-03 | 华为技术有限公司 | Model data transmission method and communication apparatus |
CN114399019A (en) * | 2021-12-30 | 2022-04-26 | 南京风兴科技有限公司 | Neural network compiling method, system, computer device and storage medium |
Also Published As
Publication number | Publication date |
---|---|
CN117835405A (en) | 2024-04-05 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
WO2023040887A1 (en) | Information reporting method and apparatus, terminal and readable storage medium | |
WO2024061287A1 (en) | Artificial intelligence (ai) model transmission method and apparatus, and terminal and medium | |
EP4236428A1 (en) | Communication data processing method and apparatus, and communication device | |
WO2023160547A1 (en) | Information response method, information sending method, terminal, and network side device | |
WO2023198094A1 (en) | Model input determination method and communication device | |
WO2023213270A1 (en) | Model training processing methods, apparatus, terminal and network side device | |
WO2023174253A1 (en) | Ai model processing method and device | |
WO2024067281A1 (en) | Ai model processing method and apparatus, and communication device | |
WO2023186018A1 (en) | Information transmission method and apparatus, terminal, and readable storage medium | |
WO2024041421A1 (en) | Measurement feedback processing method and apparatus, terminal, and network side device | |
WO2024055993A1 (en) | Cqi transmission method and apparatus, and terminal and network-side device | |
WO2023198167A1 (en) | Model fine adjustment method and apparatus, and device | |
WO2024007949A1 (en) | Ai model processing method and apparatus, and terminal and network-side device | |
WO2024041420A1 (en) | Measurement feedback processing method and apparatus, and terminal and network-side device | |
WO2024149156A1 (en) | Information transmission method and apparatus, and terminal and network-side device | |
WO2023098535A1 (en) | Information interaction method and apparatus, and communication device | |
WO2023179653A1 (en) | Beam processing method and apparatus, and device | |
WO2023174325A1 (en) | Ai model processing method and device | |
WO2023179664A1 (en) | Channel information transmission method and apparatus, and communication device | |
WO2024032694A1 (en) | Csi prediction processing method and apparatus, communication device, and readable storage medium | |
WO2023179651A1 (en) | Beam processing method and apparatus and device | |
WO2023198062A1 (en) | Csi measurement and reporting method, apparatus, device and system, and storage medium | |
WO2024041419A1 (en) | Information transmission method and apparatus, terminal, and network side device | |
WO2024001981A1 (en) | Precoding matrix indication method, terminal, and network side device | |
EP4404613A1 (en) | Parameter selection method, parameter configuration method, terminal and network side device |
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: 23867573 Country of ref document: EP Kind code of ref document: A1 |