CN-122002305-A - AI model management method and device and communication equipment

Abstract

The application discloses an AI model management method, an AI model management device and communication equipment, belonging to the technical field of communication, wherein the AI model management method comprises the steps that first communication equipment sends first information of a target AI model to second communication equipment, and the first information comprises at least one of a first identifier and a second identifier; and using a model identifier to indicate the target AI model, and performing model management of the target AI model, wherein the model identifier is associated with at least one of the first identifier and the second identifier, the first identifier is used for indicating the structure of the target AI model, and the second identifier is used for indicating the model parameters of the target AI model.

Inventors

• YANG ANG

Assignees

• 维沃移动通信有限公司

Dates

Publication Date

20260508

Application Date

20241106

Claims (20)

1. 1. An artificial intelligence AI model management method, comprising: the first communication device sends first information of a target AI model to the second communication device, wherein the first information comprises at least one of a first identifier and a second identifier; The first communication device indicates the target AI model using a model identification, performs model management of the target AI model, and is associated with at least one of the first identification and the second identification; Wherein the first identifier is used for indicating the structure of a target AI model, and the second identifier is used for indicating model parameters of the target AI model.
2. 2. The method of claim 1, wherein the model identification is transmitted by the second communication device to the first communication device for indicating at least the first identification or the second identification, or The model identification includes at least a first identification or a second identification.
3. 3. The method of claim 2, wherein the model identification is further used to indicate or the model identification further comprises at least one of: Version information of the model or model parameters, a time stamp, a cell identity, a cell group identity, a region identity, a public land mobile network PLMN identity, a provider identity of the first communication device, a model identity of the first communication device, a provider identity of the second communication device, a model identity of the second communication device.
4. 4. The method of claim 1, wherein the first identifier is a global identifier, or The first identity is obtained by at least one of: Protocol conventions; the manufacturer agrees offline; based on the second communication device allocation or indication determination.
5. 5. The method of claim 1, wherein the first identifiers comprise N1, and the first communication device supports activating, running, or using models corresponding to the N2 first identifiers; wherein N2 is greater than or equal to 1 and N1 is greater than or equal to N2.
6. 6. The method of claim 1, wherein the second identifier is a global identifier or a local identifier, or The second identity is obtained by at least one of: the manufacturer agrees offline; based on the second communication device allocation or indication determination.
7. 7. The method according to claim 1 or 6, wherein the second identity or the first information does not carry a target identity if the second identity is available at least in the current cell, in a cell group in which the current cell is located or in a region in which the current cell is located, or the second identity is a global identity; The target identification comprises at least one of a cell identification, a cell group identification and a region identification.
8. 8. The method according to claim 1 or 6, wherein if the second identifier is a local identifier, or the second identifier is used in a first area or a second area, the second identifier or the first information carries a target identifier, or the second identifier indicates a target identifier; The target identifier comprises at least one of a cell identifier, a cell group identifier and a region identifier, wherein the first region is a region corresponding to the target identifier, and the second region is a region with at least one of the same wireless environment characteristics, software configuration and hardware configuration as the first region.
9. 9. The method of claim 1, wherein the first identifiers corresponding to the different model parameter transmissions corresponding to the same model structure are different, or The first identifiers corresponding to the same model structure are the same, and the second identifiers are used for indicating all model parameter transmission or partial model parameter transmission, or The first information also comprises indication information for indicating all model parameter transmission or part of model parameter transmission, or The method further comprises the steps of: And determining that the target AI model is all model parameter transmission or part of model parameter transmission according to the received model parameters of the target AI model.
10. 10. The method of claim 9, wherein the determining that the target AI model is a full model parameter transmission or a partial model parameter transmission based on the received model parameters of the target AI model comprises: determining that the target AI model is all model parameter transmission or part of model parameter transmission according to the second information; Wherein the second information includes at least one of: the size of the model parameters, the cost of the model parameters, the load of the model parameters and the content contained in the model parameters.
11. 11. The method of claim 1, wherein only the second identifier is included in the first information if there is only one model structure of the target AI model.
12. 12. The method of any of claims 1-11, wherein the target AI model comprises at least one of: a first AI model for use by the first communication device or the second communication device; A reference AI model of the first AI model; A second AI model for use in testing by the first communication device, the second communication device, or the test device; a reference AI model of the second AI model; The first communication device, the second communication device, or the test device is configured to match a third AI model of a second AI model used in the test; and a reference AI model of the third AI model.
13. 13. The method of any of claims 1-12, wherein the target AI model is for achieving at least one of: processing a reference signal; Channel signal transmission; Receiving a channel signal; demodulating the channel signal; transmitting a channel signal; channel state information is obtained; beam management; channel prediction; Channel coding and decoding; encoding and decoding information sources; Interference suppression; positioning; Predicting high-level business or high-level parameters; management of high-level services or high-level parameters; and (5) analyzing the control signaling.
14. 14. An AI model management method, comprising: The second communication device receives first information of a target AI model sent by the first communication device, wherein the first information comprises at least one of a first identifier and/or a second identifier; the second communication device indicates the target AI model using a model identification, performs model management of the target AI model, and is associated with at least one of the first identification and the second identification; Wherein the first identifier is used for indicating the structure of a target AI model, and the second identifier is used for indicating model parameters of the target AI model.
15. 15. The method of claim 14, wherein the model identification is used to indicate at least a first identification or a second identification, or The model identification includes at least a first identification or a second identification.
16. 16. The method of claim 15, wherein in the case where the model identifier is used to indicate at least a first identifier or a second identifier, the method further comprises: The second communication device determines a model identifier of a target AI model according to the first information; the second communication device sends a model identification of the target AI model to the first communication device.
17. 17. The method of claim 15, wherein the model identification is further used to indicate or the model identification further comprises at least one of: Version information of the model or model parameters, a time stamp, a cell identity, a cell group identity, a region identity, a public land mobile network PLMN identity, a provider identity of the first communication device, a model identity of the first communication device, a provider identity of the second communication device, a model identity of the second communication device.
18. 18. The method of claim 14, wherein the first identifier is a global identifier, or The second identifier is a global identifier or a local identifier.
19. 19. The method of claim 14, wherein the first identifiers comprise N1, and the second communication device supports activating only models corresponding to the N2 first identifiers or only model parameters of the models corresponding to the N2 first identifiers to the first communication device; wherein N2 is greater than or equal to 1 and N1 is greater than or equal to N2.
20. 20. The method as recited in claim 14, further comprising: if the second identifier or the first information does not carry the target identifier, determining that the second identifier is at least available in the current cell, the cell group where the current cell is located or the region where the current cell is located, or determining that the second identifier is a global identifier; The target identification comprises at least one of a cell identification, a cell group identification and a region identification.

Description

AI model management method and device and communication equipment Technical Field The application belongs to the technical field of communication, and particularly relates to an AI model management method, an AI model management device and communication equipment. Background Artificial intelligence (ARTIFICIAL INTELLIGENCE, AI) is widely applied in various fields at present, and the artificial intelligence is integrated into a wireless communication network, so that the technical indexes such as throughput, time delay, user capacity and the like are obviously improved, and the artificial intelligence is an important task of the future wireless communication network. There are various implementations of AI modules, such as neural networks, decision trees, support vector machines, bayesian classifiers, etc. Currently for model transfer (model transfer) of a known model structure, there are two identifiers, the first for indicating the known model structure and the second for indicating the model parameters of the transfer. Although the model transmission has two identifications, the specific using method and signaling flow of the two identifications are not reasonable and effective schemes, the relation between the model transmission and the identification of the AI model is not clear, and how to use the model structure identification and the model parameter identification for AI model management is a problem to be solved urgently. Disclosure of Invention The embodiment of the application provides an AI model management method, an AI model management device and communication equipment, which can realize the management of an AI model by using a model structure identifier and a model parameter identifier. In a first aspect, an AI model management method is provided, including: the first communication device sends first information of a target AI model to the second communication device, wherein the first information comprises at least one of a first identifier and a second identifier; The first communication device indicates the target AI model using a model identification, performs model management of the target AI model, and is associated with at least one of the first identification and the second identification; Wherein the first identifier is used for indicating the structure of a target AI model, and the second identifier is used for indicating model parameters of the target AI model. In a second aspect, an AI model management method is provided, including: The second communication device receives first information of a target AI model sent by the first communication device, wherein the first information comprises at least one of a first identifier and/or a second identifier; the second communication device indicates the target AI model using a model identification, performs model management of the target AI model, and is associated with at least one of the first identification and the second identification; Wherein the first identifier is used for indicating the structure of a target AI model, and the second identifier is used for indicating model parameters of the target AI model. In a third aspect, there is provided an AI model management apparatus applied to a first communication device, including: the first sending module is used for sending first information of the target AI model to the second communication equipment, wherein the first information comprises at least one of a first identifier and a second identifier; A first processing module for indicating the target AI model using a model identification, performing model management of the target AI model, the model identification being associated with at least one of the first identification and the second identification; Wherein the first identifier is used for indicating the structure of a target AI model, and the second identifier is used for indicating model parameters of the target AI model. In a fourth aspect, there is provided an AI model management apparatus applied to a second communication device, including: the first receiving module is used for receiving first information of a target AI model sent by first communication equipment, wherein the first information comprises at least one of a first identifier and/or a second identifier; A third processing module for indicating the target AI model using a model identification, performing model management of the target AI model, the model identification being associated with at least one of the first identification and the second identification; Wherein the first identifier is used for indicating the structure of a target AI model, and the second identifier is used for indicating model parameters of the target AI model. In a fifth aspect, there is provided an AI model management apparatus configured to perform or implement the steps of the method as set forth in the first aspect. In a sixth aspect, a communication device is provided, the communication device being a first communication device