CN-122028055-A - Deployment method and device of artificial intelligence AI model

Abstract

The present disclosure relates to a deployment method and apparatus for an artificial intelligence AI model, the method comprising: the first node determines that the running state of the first node meets a preset condition. And the first node executes the AI deployment operation under the condition that the running state of the first node meets the preset condition. The AI deployment operation is one of operations of changing an AI model operated in a first node, wherein the first node is any one of a terminal, an access network device or a network element of a core network.

Inventors

• GU XIAOFEI
• LI HUIXIN
• ZHANG XIAOKANG
• SUN WANFEI

Assignees

• 大唐移动通信设备有限公司

Dates

Publication Date

20260512

Application Date

20241112

Claims (20)

1. 1. A method of deploying an artificial intelligence AI model, the method being applied to a first node in a mobile communication system, the method comprising: the first node determines that the running state of the first node meets a preset condition; The first node executes AI deployment operation under the condition that the running state of the first node meets the preset condition; The AI deployment operation is one of operations of changing an AI model running in the first node, wherein the first node is any one of a terminal, an access network device or a network element of a core network.
2. 2. The method of claim 1, wherein the step of determining the position of the substrate comprises, The first node determining that the running state of the first node meets a preset condition includes: the first node detects that the running state of the first node meets a preset condition; the first node executes an AI deployment operation under the condition that the running state of the first node meets the preset condition, including: And the first node executes AI deployment operation under the condition that the running state of the first node is detected to meet the preset condition.
3. 3. The method of claim 2, wherein the AI deployment operation comprises one or more of activating a first AI model, deactivating a second AI model, and switching an operating AI model to a third AI model; Wherein the first AI model, the second AI model, and the third AI model are AI models pre-deployed in the first node.
4. 4. The method of claim 2, wherein the first node performs an AI deployment operation if it detects that the operational state of the first node satisfies the preset condition, comprising: The first node sends request information to a second node when detecting that the running state of the first node meets the preset condition, wherein the request information is used for indicating the second node to send an AI model to the first node; The first node receives first indication information sent by the second node according to the request information, wherein the first indication information comprises a fourth AI model; And the first node executes AI deployment operation according to the first indication information, wherein the AI deployment operation comprises activating the fourth AI model or switching the running AI model into the fourth AI model.
5. 5. The method of claim 4, wherein the request information includes the operational status of the first node.
6. 6. The method of claim 1, wherein the step of determining the position of the substrate comprises, The first node determining that the running state of the first node meets a preset condition includes: the first node receives second indication information from a second node, wherein the second indication information is used for indicating the second node to detect that the running state of the first node meets a preset condition; the first node executes an AI deployment operation when the running state of the first node meets a preset condition, including: And the first node executes AI deployment operation after receiving the second indication information.
7. 7. The method of claim 6, wherein the AI deployment operation comprises one or more of activating a fifth AI model, deactivating a sixth AI model, and switching an operating AI model to a seventh AI model; Wherein the fifth AI model, the sixth AI model, and the seventh AI model are AI models that were pre-deployed in the first node.
8. 8. The method of claim 6, wherein the second indication information carries an eighth AI model, wherein the first node performs AI deployment operations after receiving the second indication information, comprising: and after receiving the second instruction information, the first node activates the eighth AI model carried in the second instruction information in the first node, or switches the running AI model to the eighth AI model carried in the second instruction information in the first node.
9. 9. The method according to any of claims 1-8, wherein the first node is a terminal; The preset conditions include one or more of the first node moving to an area where AI is restricted from being used, the first node being in a power saving mode, an operating system of the first node changing, a usage rate of a storage resource of the first node changing, a usage rate of a computing power of the first node changing, and an accuracy of an AI model running in the first node being below an accuracy threshold.
10. 10. The method according to any of claims 1-8, wherein the first node is a network element of a core network; the preset conditions include one or more of deployment of the first node to an area where AI is restricted from being used, a change in a load state of the first node, and an accuracy of an AI model running in the first node being below an accuracy threshold.
11. 11. The method according to any of claims 1-8, wherein the first node is an access network device; The preset conditions include one or more of a change in utilization rate of air interface resources of the first node, a change in load state of the first node, a change in utilization rate of computing power of the first node, and an accuracy of an AI model running in the first node being below an accuracy threshold.
12. 12. A deployment method of an artificial intelligence AI model, the method being applied to a mobile communication system including a first node and a second node, the method comprising: The second node acquires the running state of the first node, wherein the first node is any one of a terminal, access network equipment or network elements of a core network; After the second node obtains that the running state of the first node meets the preset condition, the second node sends first indication information to the first node, wherein the first indication information is used for indicating the first node to execute AI deployment operation; wherein the AI deployment operations include operations to change an AI model running in the first node.
13. 13. The method of claim 12, wherein the first indication information is specifically configured to instruct the first node to activate a first AI model, deactivate the second AI model, and switch an operating AI model to the third AI model; Wherein the first AI model, the second AI model, and the third AI model are AI models pre-deployed in the first node.
14. 14. The method of claim 12, wherein the first indication information includes a fourth AI model, and wherein the first indication information is specifically used to indicate that the first node activates the fourth AI model in the first node or switches an operating AI model to the fourth AI model in the first node.
15. 15. The method according to any of claims 12-14, wherein the second node obtaining the operational status of the first node comprises: the second node receives state information from the first node, wherein the state information comprises the running state of the first node.
16. 16. The method according to any of claims 12-14, wherein the second node obtaining the operational status of the first node comprises: the second node detects an operational state of the first node.
17. 17. The method of any one of claims 12-14, wherein the preset conditions include one or more of the first node moving to an area where AI is restricted to be used, the first node being in a power saving mode, an operating system of the first node changing, a use rate of a storage resource of the first node changing, a computationally efficient use rate of the first node changing, and an accuracy of an AI model running in the first node being below an accuracy threshold; or in the case that the first node is a network element of a core network, the preset conditions include one or more of deployment of the first node to an area where AI is restricted to be used, a change in a load state of the first node, and an accuracy of an AI model running in the first node being lower than an accuracy threshold; Or in the case that the first node is an access network device, the preset conditions include one or more of a change in the usage rate of an air interface resource of the first node, a change in a load state of the first node, a change in the usage rate of computing power of the first node, and an accuracy of an AI model running in the first node being lower than an accuracy threshold.
18. 18. A deployment method of an artificial intelligence AI model, the method being applied to a mobile communication system including a first node and a second node, the method comprising: The second node receives request information from the first node, wherein the request information is used for indicating to send an AI model to the first node, and the first node is any one of a terminal, access network equipment or a network element of a core network; The second node sends first indication information to the first node according to the request information, wherein the first indication information comprises the first AI model, and the first indication information is used for indicating the first node to activate the second AI model in the first node or switch the running AI model to the second AI model in the first node.
19. 19. The method of claim 18, wherein the request message includes an operational status of the first node, and wherein the second node sends a first indication message to the first node based on the request message, comprising: And the second node transmits first indication information to the first node after determining that the running state of the first node meets the preset condition according to the request information.
20. 20. The method of claim 18, further comprising the second node detecting an operational status of the first node; the second node sends first indication information to the first node according to the request information, and the first indication information comprises: And after determining that the running state of the first node meets the preset condition, the second node sends first indication information to the first node according to the request information.

Description

Deployment method and device of artificial intelligence AI model Technical Field The disclosure relates to the field of communication, and in particular relates to a deployment method and device of an artificial intelligence AI model. Background Currently, with the development of artificial intelligence (ARTIFICIAL INTELLIGENCE, AI) technology, attempts have been made to improve the operation efficiency and service quality of a network using the artificial intelligence technology in a mobile communication system. For example, in the fifth generation mobile communication technology (5th Generation Mobile Communication Technology,5G), an AI model is run in a Network data analysis function (Network DATA ANALYTIC function, NWDAF). When each node in the network needs intelligent analysis, the node can send a related request to NWDAF and carry corresponding parameters, NWDAF can obtain an intelligent analysis result according to the parameters carried by the request and an AI model after receiving the request, and feed back the intelligent analysis result to the node. It can be seen that in the above prior art, when a node in the network needs intelligent analysis, the node is required to send a request carrying a parameter to NWDAF, then NWDAF obtains an intelligent analysis result by using an AI model, and then feeds back the intelligent analysis result to the node, so that the whole process is relatively complex. Disclosure of Invention In order to solve the technical problems, the present disclosure provides a deployment method and device of an artificial intelligence AI model. In a first aspect, a deployment method of an artificial intelligence AI model is provided, and the method is applied to a first node in a mobile communication system. And the first node executes AI deployment operation under the condition that the running state of the first node meets the preset condition. Wherein the AI deployment operation is one of operations to change an AI model running in the first node. The first node is any one of a terminal, an access network device or a network element of a core network. In some implementations, the first node determining that the operational state of the first node meets a preset condition includes the first node detecting that the operational state of the first node meets the preset condition. The first node executes an AI deployment operation under the condition that the running state of the first node meets the preset condition, wherein the first node executes the AI deployment operation under the condition that the running state of the first node meets the preset condition. In some implementations, the AI deployment operations include one or more of activating a first AI model, deactivating a second AI model, and switching an operating AI model to a third AI model. Wherein the first AI model, the second AI model, and the third AI model are AI models pre-deployed in the first node. In some implementations, the first node executes an AI deployment operation when detecting that the running state of the first node meets the preset condition, wherein the first node sends request information to a second node when detecting that the running state of the first node meets the preset condition, the request information is used for indicating the second node to send an AI model to the first node, the first node receives first indication information sent by the second node according to the request information, the first indication information comprises a fourth AI model, the first node executes the AI deployment operation according to the first indication information, and the AI deployment operation comprises activating the fourth AI model or switching the running AI model into the fourth AI model. In some implementations, the operating state of the first node is included in the request information. In some implementations, the first node determining that the running state of the first node meets a preset condition includes that the first node receives second indication information from a second node, the second indication information is used for indicating that the running state of the first node meets the preset condition, and the first node executing AI deployment operation when the running state of the first node meets the preset condition includes that the first node executes AI deployment operation after receiving the second indication information. In some implementations, the AI deployment operation includes one or more of activating a fifth AI model, deactivating a sixth AI model, and switching an operating AI model to a seventh AI model. Wherein the fifth AI model, the sixth AI model, and the seventh AI model are AI models that were pre-deployed in the first node. In some implementations, the second indication information carries an eighth AI model, and the first node executes AI deployment operation after receiving the second indication information, including that the first node activates t