CN-116599614-B - Channel prediction model training method and device, electronic equipment and readable storage medium

CN116599614BCN 116599614 BCN116599614 BCN 116599614BCN-116599614-B

Abstract

The application discloses a channel prediction model training method, a device, electronic equipment and a readable storage medium, which are applied to the technical field of communication and comprise the steps of obtaining historical channel state information, historical channel data and historical base station priori information of at least one preset base station in a historical time period, determining a use environment scene of each preset base station, generating a plurality of training samples according to the historical channel state information, the historical channel data, the base station priori information and the use environment scene, and carrying out iterative optimization on a channel prediction model to be trained according to the plurality of training samples to obtain the channel prediction model. The application solves the technical problem of low generalization of the channel prediction model.

Inventors

WANG JINGWEI
WANG YU
HUANG HONGQIN
YOU KANMIN
LIN HOUHONG
YANG HANLI

Assignees

鹏城实验室

Dates

Publication Date: 20260505
Application Date: 20230531

Claims (9)

1. A channel prediction model training method, characterized in that the channel prediction model training method comprises: Acquiring historical channel state information, historical channel data and historical base station prior information of at least one preset base station in a historical time period, and determining a use environment scene of each preset base station, wherein the base station prior information comprises at least one of channel data time-frequency correlation information, channel data frequency domain information and channel data noise amplitude information; Generating a plurality of training samples according to the historical channel state information, the historical channel data, the historical base station prior information and the use environment scene; according to the training samples, carrying out iterative optimization on a channel prediction model to be trained to obtain a channel prediction model; The step of determining the usage environment scene of each preset base station comprises the following steps: taking any base station in the preset base stations as a target base station, and acquiring environment information of the environment where at least one classified base station is located, corresponding channel data and environment information of the environment where the target base station is located; Predicting a first probability of the target base station being in a classified environment scene corresponding to any base station in the classified base stations according to channel data corresponding to the classified base stations and historical channel data of the target base station; Predicting a second probability of the target base station in a classified environment scene corresponding to any base station in the classified base stations according to the environment information of the environment in which the classified base stations are located and the environment information of the environment in which the target base station is located; Taking any base station of the classified base stations as a classified selected base station, and calculating to obtain the total probability that the target base station is in the classified environment scene corresponding to the classified selected base station according to the first probability that the target base station is in the classified environment scene corresponding to the classified selected base station and the second probability that the target base station is in the classified environment scene corresponding to the classified selected base station; Selecting a target classified base station with the maximum total probability from the classified base stations, and taking the classified environment scene corresponding to the target classified base station as the use environment scene of the target base station.
2. The channel prediction model training method of claim 1, wherein the environmental information comprises at least one of environmental sound data, base station usage information, environmental weather information, and environmental location information, The step of predicting the second probability of the target base station being in the classified environmental scene corresponding to any base station in the classified base stations according to the environmental information of the environment in which the classified base stations are located and the environmental information of the environment in which the target base station is located includes: Acquiring an environmental scene prediction model obtained by iterative optimization of environmental information of the environment where the classified base station is located; and predicting the second probability of the target base station in the classified environment scene corresponding to any base station in the classified base stations according to the environment information of the environment in which the target base station is located by the environment scene prediction model.
3. The channel prediction model training method of claim 1, wherein the step of calculating a total probability that the target base station is in the classified environmental scene corresponding to the classified selected base station based on the first probability that the target base station is in the classified environmental scene corresponding to the classified selected base station and the second probability that the target base station is in the classified environmental scene corresponding to the classified selected base station comprises: determining a first weight corresponding to the first probability and a second weight corresponding to the second probability of the classified selected base station; Calculating a first product between the first probability and the first weight of the classified selected base station, and a second product between the second probability and the second weight, and taking the sum of the first product and the second product as the total probability that the target base station is in the classified environment scene corresponding to the classified selected base station.
4. The channel prediction model training method of claim 1, wherein the step of generating a plurality of training samples based on each of the historical channel state information, each of the historical channel data, each of the historical base station prior information, and each of the usage environment scenarios comprises: Generating a plurality of training samples corresponding to each usage environment scene according to the historical channel state information, the historical channel data and the historical base station prior information corresponding to each usage environment scene, wherein one training sample consists of input characteristic data and a real tag corresponding to the input characteristic data, the input characteristic data comprises the historical channel data and the historical base station prior information of a preset base station in the usage environment scene in the historical time period, and the real tag comprises the historical channel state information of the preset base station in the usage environment scene corresponding to the input characteristic data in the historical time period.
5. The method for training a channel prediction model according to claim 4, wherein the step of iteratively optimizing the channel prediction model to be trained based on the plurality of training samples to obtain the channel prediction model comprises: And respectively carrying out iterative optimization on the channel prediction models to be trained corresponding to the use environment scenes according to the plurality of training samples corresponding to the use environment scenes, so as to obtain the channel prediction models corresponding to the use environment scenes.
6. The method for training a channel prediction model according to any one of claims 1 to 5, further comprising, after the step of iteratively optimizing the channel prediction model to be trained based on the plurality of training samples, the step of: acquiring channel state information, channel data and base station prior information corresponding to each preset base station in preset time every preset time; Taking any time period in each preset time as a target time period, generating an adjustment sample according to channel state information, channel data, base station prior information and use environment scene of each preset base station, which correspond to each preset base station in the target time period, and carrying out iterative optimization on the number of layers of multi-layer perceptron of the multi-layer perceptron mixer in the channel prediction model according to the adjustment sample.
7. A channel prediction model training apparatus, characterized in that the channel prediction model training apparatus comprises: The acquisition module is used for acquiring historical channel state information, historical channel data and historical base station priori information of at least one preset base station in a historical time period and determining the use environment scene of each preset base station, wherein the base station priori information comprises at least one of channel data time-frequency correlation information, channel data frequency domain information and channel data noise amplitude information; The generation module is used for generating a plurality of training samples according to the historical channel state information, the historical channel data, the historical base station prior information and the use environment scenes; the iteration module is used for carrying out iteration optimization on the channel prediction model to be trained according to the training samples to obtain a channel prediction model; the acquisition module is further configured to: taking any base station in the preset base stations as a target base station, and acquiring environment information of the environment where at least one classified base station is located, corresponding channel data and environment information of the environment where the target base station is located; Predicting a first probability of the target base station being in a classified environment scene corresponding to any base station in the classified base stations according to channel data corresponding to the classified base stations and historical channel data of the target base station; Predicting a second probability of the target base station in a classified environment scene corresponding to any base station in the classified base stations according to the environment information of the environment in which the classified base stations are located and the environment information of the environment in which the target base station is located; Taking any base station of the classified base stations as a classified selected base station, and calculating to obtain the total probability that the target base station is in the classified environment scene corresponding to the classified selected base station according to the first probability that the target base station is in the classified environment scene corresponding to the classified selected base station and the second probability that the target base station is in the classified environment scene corresponding to the classified selected base station; Selecting a target classified base station with the maximum total probability from the classified base stations, and taking the classified environment scene corresponding to the target classified base station as the use environment scene of the target base station.
8. An electronic device, the electronic device comprising: at least one processor, and A memory communicatively coupled to the at least one processor, wherein, The memory stores instructions executable by the at least one processor to enable the at least one processor to perform the steps of the channel prediction model training method of any one of claims 1 to 6.
9. A computer-readable storage medium, characterized in that the computer-readable storage medium has stored thereon a program for realizing the channel prediction model training method, the program for realizing the channel prediction model training method being executed by a processor to realize the steps of the channel prediction model training method according to any one of claims 1 to 6.

Description

Channel prediction model training method and device, electronic equipment and readable storage medium Technical Field The present application relates to the field of communications technologies, and in particular, to a method and apparatus for training a channel prediction model, an electronic device, and a readable storage medium. Background With the rapid development of technology, the communication technology is also developed to be mature, and in order to improve the performance of the communication system, the requirement for the accuracy of estimating the channel state information of the base station is also higher. At present, a channel prediction model is obtained through training of historical channel state information of various base stations, the historical channel state information of various base stations is used as a decision basis for channel state information prediction of the channel prediction model, the prediction accuracy of the channel prediction model for the base stations under other conditions is low easily, and the generalization of the channel prediction model is low. Disclosure of Invention The application mainly aims to provide a channel prediction model training method, a device, electronic equipment and a readable storage medium, and aims to solve the technical problem that a channel prediction model in the prior art is low in generalization. In order to achieve the above object, the present application provides a channel prediction model training method, including: Acquiring historical channel state information, historical channel data and historical base station priori information of at least one preset base station in a historical time period, and determining the use environment scene of each preset base station; generating a plurality of training samples according to the historical channel state information, the historical channel data, the base station prior information and the use environment scene; And carrying out iterative optimization on the channel prediction model to be trained according to the training samples to obtain the channel prediction model. In order to achieve the above object, the present application further provides a channel prediction model training apparatus, including: The acquisition module is used for acquiring historical channel state information, historical channel data and historical base station priori information of at least one preset base station in a historical time period and determining the use environment scene of each preset base station; The generation module is used for generating a plurality of training samples according to the historical channel state information, the historical channel data, the base station prior information and the use environment scenes; and the iteration module is used for carrying out iteration optimization on the channel prediction model to be trained according to the training samples to obtain the channel prediction model. The application also provides an electronic device comprising a memory, a processor and a program of the channel prediction model training method stored on the memory and capable of running on the processor, wherein the program of the channel prediction model training method can realize the steps of the channel prediction model training method when being executed by the processor. The present application also provides a computer readable storage medium having stored thereon a program for implementing a channel prediction model training method, which when executed by a processor implements the steps of the channel prediction model training method as described above. The application also provides a computer program product comprising a computer program which, when executed by a processor, implements the steps of a channel prediction model training method as described above. The application provides a channel prediction model training method, a device, electronic equipment and a readable storage medium, which are used for obtaining historical channel state information of at least one preset base station in a historical time period, base station priori information corresponding to each preset base station and environment information of each preset base station, generating a plurality of training samples according to the historical channel state information, the base station priori information and the environment information, carrying out iterative optimization on a channel prediction model to be trained according to the plurality of training samples to obtain the channel prediction model, wherein the training samples are generated by the historical channel state information, the base station priori information and the environment information, and the training samples are used for carrying out iterative optimization on the channel prediction model to be trained, so that the channel prediction model obtained through optimization is based on the historical channel state information, the base stat