CN-122027504-A - Communication method and device

Abstract

The invention provides a communication method and a device, which are applied to the technical field of communication, wherein the method comprises the steps of acquiring a plurality of data, wherein the plurality of data comprises first data, second data and network feedback data, the first data is data input into a first model, the second data is data output by the first data through the first model, and the network feedback data is network quality related data obtained by evaluating the second data through a network node; training the first model according to the first data, the second data and the network feedback data. In the above scheme, by training the first model, the performance of the communication system can be improved.

Inventors

• WANG FEI
• PENG CHENGHUI
• Huang huanhuan
• SUN YUZE

Assignees

• 华为技术有限公司

Dates

Publication Date

20260512

Application Date

20241111

Claims (20)

1. 1. A method of communication, comprising: Acquiring a plurality of data, wherein the plurality of data comprises first data, second data and network feedback data, the first data is data input into a first model, the second data is data output by the first data through the first model, and the network feedback data is network quality related data obtained by evaluating the second data through a network node; And training the first model according to the first data, the second data and the network feedback data.
2. 2. The method of claim 1, wherein the method further comprises: Acquiring type information of each data in the plurality of data, wherein the type information comprises input data, output data or feedback data; and classifying the plurality of data according to the type information.
3. 3. The method of claim 1 or 2, wherein the training the first model based on the first data, the second data, and the network feedback data comprises: acquiring identification information corresponding to the first data, identification information corresponding to the second data and identification information corresponding to the network feedback data; and processing the first data, the second data and the network feedback data according to the identification information.
4. 4. The method of claim 3, wherein the identification information comprises a task identification, and wherein the processing the first data, the second data, and the network feedback data based on the identification information comprises: And determining the first data, the second data and the network feedback data corresponding to the same task according to the task identification.
5. 5. The method of claim 3, wherein the identification information includes a task identification and a time series, and wherein the processing the first data, the second data, and the network feedback data based on the identification information includes: According to the task identification, determining first data, second data and network feedback data corresponding to the same task; and numbering the first data, the second data and the network feedback data corresponding to the same task according to the time sequence.
6. 6. The method of claim 5, wherein the first data corresponding to the same task comprises first input data and second input data, the second data corresponding to the same task comprises first output data and second output data, the first input data corresponds to the first output data, the second input data corresponds to the second output data, the first output data has the same load as the second input data, the time sequence of the first output data precedes the time sequence of the second input data, the numbering the first data, the second data, and the network feedback data corresponding to the same task according to the time sequence comprises: and adding a first numerical value to the number corresponding to the first output data according to the time sequence of the first output data and the time sequence of the second input data to obtain the number corresponding to the second input data.
7. 7. The method of claim 5, wherein the first data corresponding to the same task comprises first input data, second input data, and third input data, the second data corresponding to the same task comprises first output data, second output data, and third output data, the first input data corresponds to the first output data, the second input data corresponds to the second output data, the third input data corresponds to the third output data, the first output data has a load that is the same as a load of the second input data and a load of the third input data, respectively, the time sequence of the first output data is the same as the time sequence of the second input data, the time sequence of the second input data is the same as the time sequence of the third input data, and the numbering the first data, the second data, and the network feedback data corresponding to the same task according to the time sequence comprises: adding a first numerical value to the number corresponding to the first output data according to the time sequence of the first output data and the time sequence of the second input data to obtain the number corresponding to the second input data; and determining that the number corresponding to the third input data is equal to the number corresponding to the second input data according to the time sequence of the second input data and the sequence of the time sequence of the third input data.
8. 8. The method of claim 5, wherein the first data corresponding to the same task comprises first input data, second input data, and third input data, the second data corresponding to the same task comprises first output data and second output data and third output data, the first input data corresponds to the first output data, the second input data corresponds to the second output data, the third input data corresponds to the third output data, the third input data comprises a load of the first output data and a load of the second output data, the third input data time-series follows the first output data time-series and the second output data time-series, and the numbering the first data, the second data, and the network feedback data corresponding to the same task according to the time-series comprises: Adding a first value to the number corresponding to the first output data according to the time sequence of the first output data, the time sequence of the second output data and the time sequence of the third input data to obtain the number corresponding to the third input data, wherein the time sequence of the first output data is after the time sequence of the second output data, or And adding a first numerical value to the number corresponding to the second output data according to the time sequence of the first output data, the time sequence of the second output data and the time sequence of the third input data to obtain the number corresponding to the third input data, wherein the time sequence of the second output data is after the time sequence of the first output data.
9. 9. The method of any of claims 1-8, wherein the first data comprises at least one of user demand information, retrieval enhancement generation RAG information, or prompt for prompt information.
10. 10. The method of any of claims 1-9, wherein the second data comprises at least one of a task decomposition result, tool call information, or configuration information of the network node.
11. 11. The method of any of claims 1-10, wherein the network feedback data comprises at least one of resource utilization, network throughput, task completion accuracy, or task average latency.
12. 12. A communication device, comprising: The system comprises an acquisition module, a network feedback module and a network node, wherein the acquisition module is used for acquiring a plurality of data, the plurality of data comprises first data, second data and network feedback data, the first data is data input into a first model, the second data is data output by the first data through the first model, and the network feedback data is network quality related data obtained by evaluating the second data through the network node; and the processing module is used for training the first model according to the first data, the second data and the network feedback data.
13. 13. The apparatus of claim 12, wherein the device comprises a plurality of sensors, The acquisition module is further used for acquiring type information of each data in the plurality of data, wherein the type information comprises input data, output data or feedback data; the processing module is further used for classifying the plurality of data according to the type information.
14. 14. The apparatus of claim 12 or 13, wherein, The acquisition module is further configured to acquire identification information corresponding to the first data, identification information corresponding to the second data, and identification information corresponding to the network feedback data; The processing module is further configured to process the first data, the second data, and the network feedback data according to the identification information.
15. 15. The apparatus of claim 14, wherein the identification information comprises a task identification; and the processing module is also used for determining the first data, the second data and the network feedback data corresponding to the same task according to the task identification.
16. 16. The apparatus of claim 14, wherein the identification information comprises a task identification and a time sequence; the processing module is further used for determining first data, second data and network feedback data corresponding to the same task according to the task identification; The processing module is further configured to number, according to the time sequence, the first data, the second data, and the network feedback data corresponding to the same task.
17. 17. The apparatus of claim 16, wherein the first data corresponding to the same task comprises first input data and second input data, the second data corresponding to the same task comprises first output data and second output data, the first input data corresponds to the first output data, the second input data corresponds to the second output data, the first output data has the same load as the second input data, and the time sequence of the first output data precedes the time sequence of the second input data; the processing module is further configured to add a first numerical value to a number corresponding to the first output data according to a sequence of the time sequence of the first output data and the time sequence of the second input data, so as to obtain a number corresponding to the second input data.
18. 18. The apparatus of claim 16, wherein the first data corresponding to the same task comprises first input data, second input data, and third input data, the second data corresponding to the same task comprises first output data, second output data, and third output data, the first input data corresponds to the first output data, the second input data corresponds to the second output data, the third input data corresponds to the third output data, the first output data has a load that is the same as a load of the second input data and a load of the third input data, respectively, the first output data has a time sequence that is the same as a time sequence of the second input data before the second input data; The processing module is further configured to add a first numerical value to a number corresponding to the first output data according to a sequence of the time sequence of the first output data and the time sequence of the second input data, so as to obtain a number corresponding to the second input data; The processing module is further configured to determine, according to the time sequence of the second input data and the time sequence of the third input data, that a number corresponding to the third input data is equal to a number corresponding to the second input data.
19. 19. The apparatus of claim 16, wherein the first data corresponding to the same task comprises first input data, second input data, and third input data, the second data corresponding to the same task comprises first output data and second output data and third output data, the first input data corresponding to the first output data, the second input data corresponding to the second output data, the third input data corresponding to the third output data, the payload of the third input data comprising a payload of the first output data and a payload of the second output data, the time sequence of the third input data following the time sequence of the first output data and the time sequence of the second output data; The processing module is further configured to add a first value to a number corresponding to the first output data according to a time sequence of the first output data, a time sequence of the second output data, and a time sequence of the third input data, to obtain a number corresponding to the third input data, where the time sequence of the first output data is after the time sequence of the second output data, or The processing module is further configured to add a first value to a number corresponding to the second output data according to a time sequence of the first output data, a time sequence of the second output data, and a sequence of the time sequence of the third input data, so as to obtain a number corresponding to the third input data, where the time sequence of the second output data is after the time sequence of the first output data.
20. 20. The apparatus of any one of claims 12-19, wherein the first data comprises at least one of user demand information, retrieve enhancement generation RAG information, or prompt for prompt information.

Description

Communication method and device Technical Field The present application relates to the field of communications technologies, and in particular, to a communications method and apparatus. Background With the continued development of wireless communication technology, future communication systems may be fused with artificial intelligence (ARTIFICIAL INTELLIGENCE, AI) technology, for example, a fusion of future mobile communication systems with AI models. There are a large number of agents (agents) in future communication networks based on AI models, which can implement future network services through data interactions. Because the AI model is a pre-trained model, the AI model is also required to be optimized according to the actual business effect brought by the reasoning decision when the AI model is applied in the network, so as to adjust the generation strategy of the model. However, when the AI model is used in a network, it is difficult to predict whether the generated policy is good or bad, the policy can only be judged from the actual operation result of the service after the policy is executed, and in the network operation process, the decision of judging the AI model cannot be intervened in real time, which may result in poor performance of the communication system. Disclosure of Invention The embodiment of the application provides a communication method and a communication device, which can improve the performance of a communication system by training a first model. In a first aspect, an embodiment of the present application provides a communication method, where the method may be performed by a first network element, may also be performed by a module (for example, a processor, a chip, or a chip system) applied to the first network element, and may also be implemented by a logic node, a logic module, or software capable of implementing all or part of the functions of the first network element, where the method includes: The method comprises the steps of obtaining a plurality of data, wherein the plurality of data comprise first data, second data and network feedback data, the first data are data input into a first model, the second data are data output by the first data through the first model, and the network feedback data are network quality related data obtained by evaluating the second data through a network node. And training the first model according to the first data, the second data and the network feedback data. By acquiring a plurality of data, the first network element can train the first model according to the first data, the second data and the network feedback data in the plurality of data, which is beneficial to realizing optimization of decision of the agent by utilizing the network feedback, thereby improving the performance of the communication system. In one possible design, type information of each of the plurality of data is obtained, the type information including input data, output data, or feedback data, and the plurality of data is classified according to the type information. The first network element classifies the plurality of data by acquiring the type information to obtain the first data, the second data and the network feedback data, which is favorable for standardizing the data format of the network feedback subsequently, thereby realizing the optimization of the decision of the agent by utilizing the network feedback and improving the performance of the communication system. In another possible design, the identification information corresponding to the first data, the identification information corresponding to the second data and the identification information corresponding to the network feedback data are obtained, and the first data, the second data and the network feedback data are processed according to the identification information. By acquiring the identification information, the first network element can process the first data, the second data and the network feedback data according to the identification information, so that the standardization of the data format of the network feedback is realized, the follow-up optimization of the decision of the intelligent agent by utilizing the network feedback is facilitated, and the performance of the communication system is improved. In another possible design, the identification information comprises a task identification, and the first data, the second data and the network feedback data corresponding to the same task are determined according to the task identification. The first data, the second data and the network feedback data are processed through the task identifier, so that the standardization of the data format of the network feedback can be realized, the decision of an intelligent agent can be optimized by utilizing the network feedback subsequently, and the performance of a communication system is improved. In another possible design, the identification information includes a task identification and a time