Communication method, and apparatus

Abstract

Provided in the present application are a communication method and an apparatus, which relate to the field of communications, and are used for improving the effect of model management. The method comprises: a network device sends to a terminal configuration information used for indicating a first opportunity; and at the first opportunity, the terminal sends first indication information used for determining the prediction performance of a first model in predicting network parameters. The network device configures for the terminal the configuration information used for indicating the first opportunity, and on the basis of the first opportunity, the terminal can send the first indication information used for determining the prediction performance of the first model in predicting network parameters, such that the network device can correctly execute model control, thereby ensuring the system performance.

Description

Communication methods and devices

[0001] This application claims priority to Chinese Patent Application No. 202410588286.3, filed with the State Intellectual Property Office of China on May 9, 2024, entitled "Communication Method and Apparatus", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of communications, and more particularly to communication methods and apparatus. Background Technology

[0003] In the R19 discussion, an AI-based measurement prediction method for radio resource management (RRM) was proposed, encompassing predictions in the time, spatial, and frequency domains. Specifically, the terminal can perform time-domain predictions using AI / machine learning (ML) models. The AI / ML model has two prediction capabilities: first, it can directly predict all measurement results within the prediction window and perform L3 filtering on these results; second, it can directly predict the results after L3 filtering. These measurement results can be refined to the beam level or summarized to the cell level. After performing time-domain prediction, the terminal can send the error between the predicted result and the corresponding true value to the network device. Using these errors, the network device can monitor the prediction performance of the model in real time and perform targeted model management and optimization accordingly.

[0004] However, the current model management is not very effective. Summary of the Invention

[0005] This application provides a communication method and apparatus for improving the effectiveness of model management.

[0006] To achieve the above objectives, this application adopts the following technical solution:

[0007] Firstly, a communication method is provided. This method is applied to a terminal. The executing entity of the method can be the terminal itself, a component or device applied to the terminal (e.g., a processor, chip, or chip system), or a logic module or software capable of implementing all or part of the terminal's functions. The communication method includes: receiving configuration information indicating a first timing; and sending first indication information at the first timing for determining the prediction performance of a first model prediction network parameter.

[0008] In the first aspect, the network device configures configuration information to the terminal indicating a first opportune moment. The terminal can then send first indication information based on this first opportune moment to determine the prediction performance of the first model when predicting network parameters. This enables the network device to correctly execute model control based on the first opportune moment and / or the content of the first indication information, ensuring system performance.

[0009] In one possible implementation, the configuration information includes a first threshold and a timer value. When the difference between the predicted value obtained by the first model predicting the network parameters and the actual value of the corresponding network parameters is greater than the first threshold, the timer is started, and the expiration time of the timer is the first timing. Alternatively, the configuration information includes a first period, and the timing indicated by the first period is the first timing. Alternatively, the configuration information indicates that a first indication information is sent when the predicted value obtained by the first model predicting the network parameters is sent, and the timing of sending the predicted value is the first timing.

[0010] In this implementation, several possible configuration information are designed. If the configuration information includes a first threshold and a timer value, the network device can determine the predicted value corresponding to the first indication information fed back by the terminal based on the content of the first indication information or the first timing. If the configuration information includes a first period or indicates that the first indication information is sent when the predicted value obtained from the first model predicts the network parameters, the network device can determine the predicted value corresponding to the first indication information fed back by the terminal based on the first timing.

[0011] In one possible implementation, the first indication information includes at least one index information and a first error corresponding to the index information, the first error being the difference between the predicted value obtained by the first model predicting the network parameters and the true value of the corresponding network parameters; or, the first indication information includes a second error, the second error being the result of filtering at least one first error.

[0012] In this implementation, when the configuration information includes a first threshold and a timer value, the index information can indicate the order in which the first errors are determined. The network device can determine the performance trend of the first model based on multiple first errors whose order is indicated by the index information, and thus manage the first model. When the first indication information includes a second error, since the second error is the result of filtering at least one first error, it reflects the model's performance over a period of time, avoiding false positives or false negatives caused by managing the model based on errors at a single moment.

[0013] In one possible implementation, the configuration information includes a first period, the timing indicated by the first period being a first timing; or, the configuration information indicates that a first indication information is sent when the predicted value obtained from the first model predicts the network parameters, the timing of sending the predicted value being a first timing; the first indication information includes a first true value of the network parameters, wherein the predicted value obtained from the first model predicts the network parameters corresponding to the first true value is the predicted value of a second timing, the correspondence between the true value and the predicted value of the second timing is determined according to the first parameter and the second period, the second period is the period for sending the predicted value, the first parameter is used to indicate the prediction window size of the first model, and the predicted value of the second timing is the predicted value sent before the first timing.

[0014] In this implementation, by configuring the two possible first timings mentioned above, the network device can determine the predicted value corresponding to the first indication information fed back by the terminal based on the second period and the first parameter when it receives the first indication information sent by the terminal at the first timing, thereby effectively managing the model.

[0015] In one possible implementation, the configuration information includes a first period, and the timing that conforms to the first period is a first timing; the method further includes: receiving a first duration from a network device; determining at least one prediction error within the first duration; and sending second indication information to the network device, the second indication information indicating at least one prediction error.

[0016] In this implementation, network devices can control the model based on the error over a period of time, thereby avoiding false positives or false negatives of the model, reducing unnecessary model control, and ensuring system performance.

[0017] In one possible implementation, the method further includes: receiving a second duration from the network device; and continuously transmitting predicted values ​​of the network parameters obtained from the first model prediction during the second duration.

[0018] In this implementation, the network device sends a second duration to the terminal, and the terminal continuously sends predicted values ​​during the second duration. This allows the network device to know the predicted values ​​for a future period, thereby enabling it to make better decisions such as handover decisions.

[0019] Secondly, a communication method is provided, which is applied to a network device. The execution subject of the method can be the network device, a component or device (e.g., a processor, chip, or chip system) applied to the network device, or a logic module or software capable of implementing all or part of the functions of the network device. The communication method includes: sending configuration information indicating a first timing for sending first indication information, the first indication information being used to determine the prediction performance when a first model predicts network parameters; and receiving the first indication information.

[0020] In the second aspect, the network device configures configuration information to the terminal indicating a first opportune moment. The terminal can then send first indication information based on this first opportune moment to determine the prediction performance of the first model's prediction network parameters. This enables the network device to correctly execute model control based on the first opportune moment and / or the content of the first indication information, ensuring system performance.

[0021] In one possible implementation, the configuration information includes a first threshold and a timer. When the difference between the predicted value obtained by the first model predicting the network parameters and the actual value of the corresponding network parameters is greater than the first threshold, the timer is started, and the expiration time of the timer is a first timing point. The time when the timer expires based on the timing start condition is also a first timing point. Alternatively, the configuration information includes a first period, and the timing indicated by the first period is the first timing point. Alternatively, the configuration information indicates that a first indication information is sent when the predicted value obtained by the first model predicting the network parameters is sent, and the timing of sending the predicted value is the first timing point.

[0022] In this implementation, several possible configuration information are designed. If the configuration information includes a first threshold and a timer value, the network device can determine the predicted value corresponding to the first indication information fed back by the terminal based on the content of the first indication information or the first timing. If the configuration information includes a first period or indicates that the first indication information is sent when the predicted value obtained from the first model predicts the network parameters, the network device can determine the predicted value corresponding to the first indication information fed back by the terminal based on the first timing.

[0023] In one possible implementation, the first indication information includes at least one index information and a first error corresponding to the index information, the first error being the difference between the predicted value obtained by the first model predicting the network parameters and the true value of the corresponding network parameters; or, the first indication information includes a second error, the second error being the result of filtering at least one first error.

[0024] In this implementation, when the configuration information includes a first threshold and a timer value, the index information can indicate the order in which the first errors are determined. The network device can determine the performance trend of the first model based on multiple first errors whose order is indicated by the index information, and thus manage the first model. When the first indication information includes a second error, since the second error is the result of filtering at least one first error, it reflects the model's performance over a period of time, avoiding false positives or false negatives caused by managing the model based on errors at a single moment.

[0025] In one possible implementation, the configuration information includes a first period, the timing indicated by the first period being a first timing; or, the configuration information indicates that a first indication information is sent when the predicted value obtained from the first model predicts the network parameters, the timing of sending the predicted value being a first timing; the first indication information includes a first true value of the network parameters, wherein the predicted value obtained from the first model predicts the network parameters corresponding to the first true value is the predicted value of a second timing, the correspondence between the true value and the predicted value of the second timing is determined according to the first parameter and the second period, the second period is the period for sending the predicted value, the first parameter is used to indicate the prediction window size of the first model, and the predicted value of the second timing is the predicted value sent before the first timing.

[0026] In this implementation, by configuring the two possible first timings mentioned above, the network device can determine the predicted value corresponding to the first indication information fed back by the terminal based on the second period and the first parameter when it receives the first indication information sent by the terminal at the first timing, thereby effectively managing the model.

[0027] In one possible implementation, the configuration information includes a first period, and the timing that conforms to the first period is a first timing; the method further includes: sending a first duration, the first duration being used to determine at least one prediction error; and receiving second indication information, the second indication information indicating at least one prediction error.

[0028] In this implementation, network devices can control the model based on the error over a period of time, thereby avoiding false positives or false negatives of the model, reducing unnecessary model control, and ensuring system performance.

[0029] In one possible implementation, the method further includes: sending a second duration, wherein the second duration is used to send the predicted values ​​obtained from the parameters of the first model prediction network; and receiving the predicted values ​​obtained from the parameters of the first model prediction network.

[0030] In this implementation, the network device sends a second duration to the terminal, and the terminal continuously sends predicted values ​​during the second duration. This allows the network device to know the predicted values ​​for a future period, thereby enabling it to make better decisions such as handover decisions.

[0031] Thirdly, a communication method is provided, which is applied to a terminal. The executing entity of the method can be the terminal, a component or device applied to the terminal (e.g., a processor, chip, or chip system), or a logic module or software capable of implementing all or part of the terminal's functions. The communication method includes: sending a first predicted value; sending first indication information at a first opportune moment, wherein the first opportune moment is the opportune moment when the error corresponding to the predicted value within the prediction window of each measured first model is obtained, and the prediction window is the prediction window for predicting the first predicted value; or, the first opportune moment is the opportune moment when the error corresponding to the first predicted value is obtained; or, the first opportune moment is the opportune moment when the first predicted value is sent, and the first indication information is used to determine the prediction performance of the first model when predicting network parameters.

[0032] In the third aspect, regarding the case where the first timing is the timing of the error corresponding to the predicted value within the prediction window of each measured first model, since the first timing is defined as the timing of the error corresponding to the predicted value within the prediction window of each measured first model, the terminal can provide the first indication information for a period of time corresponding to the prediction window multiple times. The network device can perform model management based on the first indication information for a period of time corresponding to the prediction window, thus avoiding false positives or false negatives caused by managing the model based on a single error.

[0033] When the first opportunity is the time when the error corresponding to the first predicted value is measured, the terminal can provide feedback on the first indication information for a period of time corresponding to the prediction window. The network device can use the first indication information for a period of time corresponding to the prediction window for model management, thus avoiding false positives or false negatives caused by managing the model based on a single error.

[0034] When the first timing is the timing of sending the first predicted value, the network device can determine the predicted value corresponding to the first indication information sent by the terminal when it receives the first indication information sent by the terminal at the first timing. That is, the predicted value corresponding to the first indication information is the predicted value before the first predicted value, so as to effectively manage the model.

[0035] In one possible implementation, the first timing is the timing when the error corresponding to the predicted value within the prediction window of the first model is measured, or the first timing is the timing when the error corresponding to the first predicted value is measured; the first indication information includes the first error corresponding to the predicted value within the prediction window; or, the first indication information includes at least one index information and the first error corresponding to the index information, the first error being the difference between the predicted value obtained by the first model predicting network parameters and the true value of the corresponding network parameters; or, the first indication information includes a second error, the second error being the result after filtering at least one first error.

[0036] In this implementation, the first error includes the first error corresponding to the predicted value within the prediction window. Since the first timing is defined as the timing when the error corresponding to the predicted value within the prediction window of the first model is measured, or the first timing is the timing when the error corresponding to the first predicted value is measured, the terminal can provide feedback on the first error for a period of time corresponding to the prediction window multiple times or once. The network device can then manage the model based on the first error for that period of time, avoiding false positives or false negatives caused by managing the model based on a single error. Alternatively, the terminal can provide feedback on the second error corresponding to the prediction window once. Since the second error is the result of filtering at least one first error, it reflects the model's performance over a period of time, avoiding false positives or false negatives caused by managing the model based on the error at a single moment.

[0037] In one possible implementation, the first timing is the timing of sending the first predicted value; the method further includes: receiving third indication information, the third indication information being used to indicate that the first predicted value and the first indication information are sent at the first timing.

[0038] In this implementation, the network device informs the terminal to send a first predicted value and a first indication information at a first opportune moment, so that the network device can determine that the predicted value corresponding to the first indication information fed back by the terminal is the predicted value before the first predicted value, thereby correctly executing the model control and ensuring system performance.

[0039] In one possible implementation, the method further includes: sending a second predicted value at a second time before sending the first indication information at a first time; wherein the first indication information includes a first true value, and the true value corresponding to the second predicted value is the first true value.

[0040] In this implementation, the network device can determine that the predicted value corresponding to the first indication information fed back by the terminal is the predicted value before the first predicted value, i.e., the second predicted value, so as to correctly execute the model control and ensure system performance.

[0041] In one possible implementation, the method further includes: receiving a second duration; and continuously transmitting the predicted values ​​obtained from the parameters of the first model prediction network within the second duration.

[0042] In this implementation, the network device sends a second duration to the terminal, and the terminal continuously sends predicted values ​​during the second duration. This allows the network device to know the predicted values ​​for a future period, thereby enabling it to make better decisions such as handover decisions.

[0043] Fourthly, a communication method is provided, which is applied to a network device. The execution subject of the method can be the network device, or a component or device (e.g., processor, chip, or chip system) applied to the network device, or a logic module or software capable of implementing all or part of the functions of the network device. The communication method includes: receiving a first predicted value; receiving first indication information, wherein the first indication information is sent at a first opportune moment, the first opportune moment being the opportune moment when the error corresponding to the predicted value within the prediction window of each measured first model is obtained, and the prediction window being the prediction window for predicting the first predicted value; or, the first opportune moment being the opportune moment when the error corresponding to the first predicted value is measured; or, the first opportune moment being the opportune moment when the first predicted value is sent, and the first indication information is used to determine the prediction performance of the first model when predicting network parameters.

[0044] In the fourth aspect, for the case where the first timing is the timing of the error corresponding to the predicted value within the prediction window of each measured first model, since the first timing is defined as the timing of the error corresponding to the predicted value within the prediction window of each measured first model, the terminal can provide the first indication information for a period of time corresponding to the prediction window multiple times. The network device can perform model management based on the first indication information for a period of time corresponding to the prediction window, thus avoiding false positives or false negatives caused by managing the model based on a single error.

[0045] When the first opportunity is the time when the error corresponding to the first predicted value is measured, the terminal can provide feedback on the first indication information for a period of time corresponding to the prediction window. The network device can use the first indication information for a period of time corresponding to the prediction window for model management, thus avoiding false positives or false negatives caused by managing the model based on a single error.

[0046] When the first timing is the timing of sending the first predicted value, the network device can determine the predicted value corresponding to the first indication information sent by the terminal when it receives the first indication information sent by the terminal at the first timing. That is, the predicted value corresponding to the first indication information is the predicted value before the first predicted value, so as to effectively manage the model.

[0047] In one possible implementation, the first timing is the timing when the error corresponding to the predicted value within the prediction window of the first model is measured, or the first timing is the timing when the error corresponding to the first predicted value is measured; the first indication information includes the first error corresponding to the predicted value within the prediction window; or, the first indication information includes at least one index information and the first error corresponding to the index information, the first error being the difference between the predicted value obtained by the first model predicting network parameters and the true value of the corresponding network parameters; or, the first indication information includes a second error, the second error being the result after filtering at least one first error.

[0048] In this implementation, the first error includes the first error corresponding to the predicted value within the prediction window. Since the first timing is defined as the timing when the error corresponding to the predicted value within the prediction window of each measured first model is obtained, or the first timing is the timing when the error corresponding to the measured first predicted value is obtained, the terminal can provide feedback on the first error for a period of time corresponding to the prediction window multiple times or once. The network device can perform model management based on the first error for a period of time corresponding to the prediction window, avoiding false positives or false negatives caused by managing the model based on a single error. Alternatively, the terminal can provide feedback on the second error corresponding to the prediction window once. Since the second error is the result of filtering at least one first error, that is, the second error can reflect the performance of the model over a period of time, avoiding false positives or false negatives caused by managing the model based on the error at a single moment.

[0049] In this implementation, several possible first timing scenarios are designed, and multiple schemes are provided to indicate when the terminal should send the first indication information, so that the network device can correctly execute the model's control and ensure system performance.

[0050] In one possible implementation, the first timing is the timing of sending the first predicted value; the method further includes: sending third indication information, the third indication information being used to indicate that the first predicted value and the first indication information are sent at the first timing.

[0051] In this implementation, the network device informs the terminal to send a first predicted value and a first indication information at a first opportune moment, so that the network device can determine that the predicted value corresponding to the first indication information fed back by the terminal is the predicted value before the first predicted value, thereby correctly executing the model control and ensuring system performance.

[0052] In one possible implementation, the method further includes: receiving a second predicted value before receiving the first indication information; wherein the second predicted value is sent at a second time, the first indication information includes a first true value, and the true value corresponding to the second predicted value is the first true value.

[0053] In this implementation, the network device can determine that the predicted value corresponding to the first indication information fed back by the terminal is the predicted value before the first predicted value, i.e., the second predicted value, so as to correctly execute the model control and ensure system performance.

[0054] In one possible implementation, the method further includes: sending a second duration, wherein the second duration is used to send the predicted values ​​obtained from the parameters of the first model prediction network; and receiving the predicted values ​​obtained from the parameters of the first model prediction network.

[0055] In this implementation, the network device sends a second duration to the terminal, and the terminal continuously sends predicted values ​​during the second duration. This allows the network device to know the predicted values ​​for a future period, thereby enabling it to make better decisions such as handover decisions.

[0056] Fifthly, a communication device is provided for implementing the method described in any one of the first to fourth aspects. For example, the communication device may be a terminal as described in the first aspect, or a device included in a terminal, such as a chip or chip system; or, the communication device may be a network device as described in the second aspect, or a device included in a network device, such as a chip or chip system. When the device is a chip system, it may be composed of chips or may include chips and other discrete components.

[0057] The communication device includes modules, units, or means corresponding to the implementation method. These modules, units, or means can be implemented in hardware, software, or by hardware executing corresponding software. The hardware or software includes one or more modules or units corresponding to the functions.

[0058] In some possible designs, the communication device may include a processing module and a transceiver module. The processing module can be used to implement the processing functions in any of the above aspects and any possible implementations. The transceiver module, also called a transceiver unit, is used to implement the sending and / or receiving functions in any of the above aspects and any possible implementations. The transceiver module may consist of transceiver circuitry, a transceiver, a transceiver unit, or a communication interface.

[0059] In some possible designs, the transceiver module includes a sending module and / or a receiving module, which are used to implement the sending or receiving functions in any of the above aspects and any possible implementations.

[0060] A sixth aspect provides a communication device, comprising: a processor and a communication interface; the communication interface being used to communicate with a module outside the communication device; the processor being used to execute computer programs or instructions to cause the communication device to perform the methods described in any aspect. For example, the communication device may be a terminal as described in the first aspect, or a device included in a terminal, such as a chip or a chip system; or, the communication device may be a network device as described in the second aspect, or a device included in a network device, such as a chip or a chip system. When the device is a chip system, it may be composed of chips or may include chips and other discrete devices.

[0061] A seventh aspect provides a communication device comprising: at least one processor; said processor being configured to execute a computer program or instructions stored in a memory to cause the communication device to perform the method described in any aspect. The memory may be coupled to the processor, or the memory may exist independently of the processor; for example, the memory and the processor may be two separate modules. The memory may be located outside or within the communication device.

[0062] The communication device is used to implement the method described in any of the first to fourth aspects. For example, the communication device can be a terminal as described in the first aspect, or a device contained in a terminal, such as a chip or chip system; or, the communication device can be a network device as described in the second aspect, or a device contained in a network device, such as a chip or chip system. When the device is a chip system, it can be composed of chips or can include chips and other discrete components.

[0063] Eighthly, a computer-readable storage medium is provided that stores a computer program or instructions that, when executed on a communication device, enable the communication device to perform the method described in either aspect.

[0064] In a ninth aspect, a computer program product containing instructions is provided that, when run on a communication device, enables the communication device to perform the method described in either aspect.

[0065] In a tenth aspect, a communication device is provided, configured to cause the communication device to perform the method described in any one aspect.

[0066] It is understandable that when the communication device provided in any of the fifth to seventh aspects is a chip, the sending action / function of the communication device can be understood as outputting information, and the receiving action / function of the communication device can be understood as inputting information.

[0067] The technical effects of any of the design methods in aspects five through ten can be found in the technical effects of different design methods in aspects one through four, and will not be repeated here.

[0068] Eleventhly, a communication system is provided, which includes the network equipment and the terminal described above. Attached Figure Description

[0069] Figure 1 is a schematic diagram of a communication system provided in an embodiment of this application;

[0070] Figure 2 is a schematic diagram of a network performing lifecycle management of an AI model according to an embodiment of this application;

[0071] Figure 3 is a schematic diagram of a scenario for time-domain prediction using a terminal application model provided in an embodiment of this application;

[0072] Figure 4 is a schematic diagram of another communication system provided in an embodiment of this application;

[0073] Figure 5 is a schematic diagram of an O-RAN architecture provided in an embodiment of this application;

[0074] Figure 6 is a flowchart illustrating a communication method provided in an embodiment of this application;

[0075] Figure 7 is a flowchart illustrating another communication method provided in an embodiment of this application;

[0076] Figure 8 is a flowchart illustrating another communication method provided in an embodiment of this application;

[0077] Figure 9 is a flowchart illustrating another communication method provided in an embodiment of this application;

[0078] Figure 10 is a schematic diagram of a scenario where the terminal predicts the measurement result and reports the predicted value, according to an embodiment of this application.

[0079] Figure 11 is a schematic diagram of a scenario where the terminal predicts the result after L3 filtering of the measurement result, according to an embodiment of this application, and the predicted value is reported.

[0080] Figure 12 is a flowchart illustrating another communication method provided in an embodiment of this application;

[0081] Figure 13 is a flowchart illustrating another communication method provided in an embodiment of this application;

[0082] Figure 14 is a flowchart illustrating another communication method provided in an embodiment of this application;

[0083] Figure 15 is a schematic diagram of the structure of a communication device provided in an embodiment of this application;

[0084] Figure 16 is a schematic diagram of another communication device provided in an embodiment of this application;

[0085] Figure 17 is a schematic diagram of another communication device provided in an embodiment of this application. Detailed Implementation

[0086] The network architecture and business scenarios described in the embodiments of this application are for the purpose of more clearly illustrating the technical solutions of the embodiments of this application, and do not constitute a limitation on the technical solutions provided in the embodiments of this application. As those skilled in the art will know, with the evolution of network architecture and the emergence of new business scenarios, the technical solutions provided in the embodiments of this application are also applicable to similar technical problems.

[0087] Before introducing the embodiments of this application, some technologies involved in the embodiments of this application will be briefly introduced.

[0088] Observation Window: An observation window is a time period used to collect and analyze historical data. Within this period, the system collects various measurement data related to network performance, user behavior, channel conditions, and more. This data serves as input to the AI ​​model for training and optimization, enabling predictions of future network conditions. The length of the observation window is typically determined based on the specific application scenario and requirements, and can range from a few minutes to several hours or longer.

[0089] Prediction Window: A prediction window is a time period used to predict the future state of the network. With the help of an AI model, the system predicts the network state over a future period based on historical data collected within the prediction window. The length of the prediction window depends on the specific application scenario and requirements, and can be a few seconds, a few minutes, or longer. Through the prediction window, the system can understand network trends in advance and take corresponding measures to optimize network performance and user experience.

[0090] AI, originating in the 1950s, aims to perform complex calculations by simulating the thought processes of the human brain. With the rapid development of data storage and computing power, the application areas of AI have become increasingly widespread. In 3GPP Release 17, the approval of the Study Item (SI) signifies that AI technology has been officially incorporated into the considerations of New Radio (NR), aiming to improve network performance and optimize user experience through intelligent data collection and analysis.

[0091] Figure 1 shows a schematic diagram of the communication system for AI application in NR. As shown in Figure 1, the communication system includes:

[0092] The data collection entity is responsible for integrating data inputs from base stations (gNB), base station central units (gNB-CU), base station distribution units (gNB-DU), user equipment (UE), and other management entities. This data serves as the foundational database for AI model training and data analysis inference. Examples include training data, monitoring data, and inference data.

[0093] Model training involves analyzing training data provided by data collection entities to determine and output the optimal AI model.

[0094] The model storage entity is responsible for storing these trained or updated models.

[0095] The Management entity is responsible for providing model transfer / delivery requests. The Model Storage entity transfers / delivers the model to the Model Inference entity based on these requests. The Model Inference entity then feeds back the received model's inference output to the Management entity. The Management entity, based on the inference data, provides performance feedback / retraining requests to the Model Training entity. It also instructs the Model Inference entity to select / (deactivate) / switch / roll back the model.

[0096] The Model Inference entity is responsible for using these AI models, based on data provided by the Data Collection entity, to make reasonable AI-based predictions about network operation or to guide the network in policy adjustments.

[0097] The network (NW) can perform lifecycle management (LCM) on the aforementioned AI model. For example, Figure 2 illustrates the flowchart of the NW performing LCM on the AI ​​model, as shown in Figure 2, which includes:

[0098] S1. UE reporting capabilities and auxiliary information:

[0099] The UE reports its supported capabilities and / or auxiliary information to the network (NW). Capability information details the features of the UE, such as supported use cases and AI functions; auxiliary information provides detailed information about the AI / ML models supported by the UE and their applicability, including applicable conditions and suitability in the current scenario.

[0100] S2. Make decisions and issue management instructions:

[0101] Based on the information reported by the UE, the NW makes decisions and sends corresponding management commands to the UE, such as activation, deactivation, handover, or rollback. For example, if the information reported by the UE indicates the existence of an applicable model / function, the NW will send a command to activate that model / function.

[0102] S3 and UE reporting performance and measurement results:

[0103] The UE reports the performance data it monitors or the actual measurement results to the NW, which will serve as an auxiliary basis for the NW's monitoring.

[0104] S4. Monitor and manage:

[0105] The NW monitors the performance based on the performance data or test results reported by the UE and performs subsequent management operations as needed. For example, if the NW detects that the current UE's model / function performance is poor, it will send a command to the UE to deactivate or switch the model / function.

[0106] As described in the background art, the terminal can apply the above model for time-domain prediction. For example, Figure 3 shows a schematic diagram of a scenario where the terminal applies the above model for time-domain prediction. As shown in Figure 3, the current time is t. n t n+1 and t n+2 The timeline shows that, as time increases, the time periods of interest for both the observation window and the prediction window also change. The current time is t. n At that time, the above model predicted the future t within the prediction window. n+1 and t n+2 The prediction results at the time are A p and B p At the current time t n+1 At that time, the terminal actually measured the corresponding predicted result A within the observation window. p The true value A t And at the current time t n+2 At that time, the terminal actually measured the corresponding predicted result B within the observation window. p The true value B t .

[0107] After the terminal obtains the predicted results and corresponding true values, it can send the error between the predicted results and the corresponding true values ​​to the network device. Using these errors, the network device can monitor the predictive performance of the model in real time and perform targeted model management and optimization accordingly. In another implementation, the terminal can send the true values ​​to the network device. The network device can calculate the error between the true values ​​and the predicted results, and using these errors, it can monitor the predictive performance of the model in real time and perform targeted model management and optimization accordingly.

[0108] However, the objective does not define how to report errors or true values. In scenarios where errors are reported, the network may manage the model based on single errors or errors with poor timeliness, leading to false positives or false negatives.

[0109] In scenarios where the true value is reported, since the true value is measured after the predicted value corresponding to it is predicted, the true value and the corresponding predicted value are not sent to the network device at the same time. During the period between reporting the predicted value and reporting the true value corresponding to it, the terminal may report other predicted values ​​to the network device. The network device cannot clearly determine which prediction result the true value corresponds to, resulting in poor model management.

[0110] To address the aforementioned technical problems, this application provides a communication method. For scenarios where a network-configured terminal periodically reports prediction results, the network device configures configuration information indicating a first opportune moment for the terminal. The terminal can then send first indication information based on this first opportune moment to determine the prediction performance of the first model when predicting network parameters. For scenarios where prediction results are reported based on events, several first opportune moments are set for the terminal to send the aforementioned first indication information. This enables the network device to correctly execute model control, ensuring system performance. Furthermore, the network device can also control the model based on first indication information over a period of time, thereby avoiding false positives or false negatives, reducing unnecessary model control, and ensuring system performance.

[0111] The method provided in the embodiments of this application will now be described with reference to the accompanying drawings.

[0112] The communication method provided in this application can be applied to various communication systems, such as: Long Term Evolution (LTE) systems, 5th Generation (5G) mobile communication systems, Wireless Fidelity (WiFi) systems, future communication systems, or systems integrating multiple communication systems, etc., and this application does not limit the application. 5G can also be referred to as New Radio (NR).

[0113] The communication method provided in this application can be applied to various communication scenarios, such as one or more of the following communication scenarios: enhanced mobile broadband (eMBB), ultra-reliable low latency communication (URLLC), machine type communication (MTC), massive machine type communications (mMTC), device to device (D2D), vehicle to everything (V2X), vehicle to vehicle (V2V), and Internet of Things (IoT).

[0114] The communication method provided in this application embodiment will be described below using the communication system shown in Figure 4 as an example.

[0115] Figure 4 is a schematic diagram of a communication system provided in an embodiment of this application. As shown in Figure 4, the communication system may include:

[0116] Network device 410 and terminal 420.

[0117] In a scenario where the network configuration terminal periodically reports prediction results, the network device 410 is used to send configuration information indicating the first timing to the terminal 420.

[0118] Terminal 420 is used to receive the above configuration information and, at a first opportune moment, send first indication information to network device 410 for determining the prediction performance when the first model predicts network parameters.

[0119] Network device 410 is used to receive the aforementioned first instruction information.

[0120] Among them, network devices can monitor the performance of the first model based on the first indication information.

[0121] In the scenario where the terminal reports the prediction result triggered by the event, the terminal 420 is used to send the first prediction value to the network device 410 after the event is triggered, and to send the first indication information to the network device 410 at the first opportune moment.

[0122] The first timing can be: the timing when the error corresponding to the predicted value within the prediction window of the first model is measured, where the prediction window is the prediction window for predicting the first predicted value; or, the timing when the error corresponding to the first predicted value is measured; or, the timing when the first predicted value is sent, where the first indication information is used to determine the prediction performance of the first model when predicting network parameters.

[0123] Network device 410 is used to receive the first predicted value and the first indication information.

[0124] The network device can monitor the performance of the first model based on the first predicted value and the first indication information.

[0125] It should be noted that Figure 4 is merely an exemplary framework diagram, and the number of nodes included in Figure 4, as well as the state of the terminals, are not limited. In addition to the functional nodes shown in Figure 4, other nodes may be included, such as core network devices, gateway devices, application servers, etc., without limitation. Network devices communicate with core network devices via wired or wireless means, such as through next-generation (NG) interfaces.

[0126] The network device is primarily used to implement at least one of the following functions: terminal resource scheduling, wireless resource management, and wireless resource control. Specifically, the network device may include any node among a base station, wireless access point, transceiver point (TRP), transmission point (TP), and some other access node. In this embodiment, the apparatus for implementing the functions of the network device can be the network device itself; it can also be an apparatus capable of supporting the network device in implementing these functions, such as a chip system, which can be installed in the network device or used in conjunction with the network device.

[0127] Terminal equipment can be user equipment (UE), mobile station (MS), or mobile terminal (MT), etc. Specifically, a terminal can be a mobile phone, tablet computer, or computer with wireless transceiver capabilities. It can also be a virtual reality (VR) terminal, an augmented reality (AR) terminal, a wireless terminal in industrial control, a wireless terminal in autonomous driving, a wireless terminal in telemedicine, a wireless terminal in a smart grid, a wireless terminal in a smart city, a smart home, or an in-vehicle terminal, etc. In the embodiments of this application, the device used to implement the terminal's functions can be the terminal itself, or it can be a device that supports the terminal in implementing those functions, such as a chip system. This device can be installed in the terminal or used in conjunction with the terminal.

[0128] The communication system shown in Figure 4 can be a closed or traditional radio access network (LOR) communication system, or it can be an open radio access network (O-RAN) communication system. In other words, the network equipment shown in Figure 4 can be network equipment in a closed or traditional radio access network, or it can be network equipment in an O-RAN.

[0129] For example, Figure 5 shows a schematic diagram of the O-RAN architecture. As shown in Figure 5, the O-RAN architecture includes:

[0130] RIC (Radio Intelligence Controller): As a core component of the O-RAN architecture, the RIC is responsible for collecting network information and performing critical optimization tasks. It communicates with the gNB-CU (Centralized Unit) and gNB-DU (Distributed Unit) via the E2 interface, enabling the O-RIC to directly control the gNB-DU.

[0131] gNB-CU: The central unit of the gNB supporting O-RAN functionality, responsible for handling higher-level protocol stack functions of radio signals, including core network functions such as user data transmission, mobility control, radio access network sharing, positioning, and session management. The gNB-CU communicates with the gNB-DU via the F1 interface. During interaction, the gNB-CU can send control signals and configuration information to the gNB-DU through the F1 interface to guide the gNB-DU in transmitting, receiving, and processing radio signals. Simultaneously, the gNB-DU can also use the F1 interface to provide feedback on radio signal status information and performance data to the gNB-CU, assisting the gNB-CU in network optimization and decision-making.

[0132] gNB-DU: gNB distributed unit that supports O-RAN functions and is responsible for handling the physical layer functions of wireless signals, such as signal modulation, demodulation, up-conversion and down-conversion.

[0133] The O-RAN architecture may also include a service unit (not shown in Figure 5). The service unit (SU) can combine artificial intelligence technology and big data analysis technology to implement non-real-time macro-control and intervention of O-RAN radio resources through the A1 interface.

[0134] The communication method provided in this application embodiment will be described below with reference to the communication system shown in Figures 3 and 4. Specifically, this application addresses the scenario where the network configuration terminal periodically reports prediction results (hereinafter referred to as prediction values). The network device configures configuration information to the terminal indicating a first opportune moment. The terminal can then send first indication information based on this first opportune moment to determine the prediction performance of the first model when predicting network parameters. For scenarios where prediction results are reported due to events, several first opportune moments are set for the terminal to send the aforementioned first indication information. This enables the network device to correctly execute model control and ensures system performance.

[0135] It should be noted that in the following embodiments of this application, the message names, parameter names, or information names between network elements are just examples. Other names may also be used in other embodiments. The communication method provided in this application does not specifically limit these names.

[0136] It is understood that in the embodiments of this application, each network element can execute some or all of the steps in the embodiments of this application. These steps or operations are merely examples, and the embodiments of this application can also perform other operations or variations of various operations. Furthermore, the steps can be executed in different orders as presented in the embodiments of this application, and it is not necessary to execute all the operations in the embodiments of this application.

[0137] Figure 6 illustrates an example of a communication method provided in an embodiment of this application. The method is described using the interaction between a network device and a terminal as an example. Of course, the entity executing the network device action in this method can also be a device / module within the network device, such as a chip, processor, or processing unit within the network device; similarly, the entity executing the terminal action in this method can also be a device / module within the terminal, such as a chip, processor, or processing unit within the terminal. This embodiment of the application does not specifically limit this. The processing performed by a single executing entity (e.g., a network device or a terminal) in this embodiment of the application can also be divided into multiple executing entities, which can be logically and / or physically separated. For example, as shown in Figure 6, the method may include the following steps:

[0138] In S610, the network device sends configuration information to the terminal, and the terminal receives the configuration information accordingly.

[0139] In the scenario where the network configuration terminal periodically reports predicted values, the network device configures configuration information to indicate the first opportune moment to the terminal. This configuration information is used to indicate the first opportune moment. The period length for the terminal to report predicted values ​​can be denoted as T1.

[0140] In this scenario, the configuration information can include the following:

[0141] 1. Configuration 1

[0142] In configuration 1, the configuration information includes a first threshold and a timer value. When the terminal determines that the difference between the predicted value obtained from the first model's prediction of the network parameters and the actual value of the corresponding network parameters is greater than the first threshold, it starts the timer. At this time, the timer's expiration time is the first opportunity. Before the timer expires, the terminal continuously records the difference between the predicted value obtained from the first model's prediction of the network parameters and the actual value of the corresponding network parameters. The aforementioned first threshold can be agreed upon by the protocol or set by the network device, and is not restricted.

[0143] In this embodiment of the application, the true value can be the measured result of network parameters such as reference signal receiving power (RSRP), reference signal received quality (RSRQ), and signal to interference plus noise ratio (SINR). Correspondingly, the predicted value can be the predicted result of network parameters such as RSRP, RSRQ, and SINR.

[0144] 2. Configuration 2

[0145] In configuration 2, the configuration information includes a first period, which indicates the first opportune moment. The first period indicates the start time and length of the period, and the length of the period in the first period can be denoted as T2. Preferably, T2 can be set to a positive integer multiple of T1, which can ensure that the actual value or error fed back by the terminal has a corresponding prediction result.

[0146] 3. Configuration 3

[0147] In configuration 3, the configuration information indicates that the first indication information is sent when the predicted value obtained from the first model prediction network parameters is sent, and the timing of sending the predicted value is the first timing. In other words, configuration 3 instructs the terminal to also report the first indication information the next time it reports the predicted value.

[0148] For configuration information 1, the network device can determine the predicted value corresponding to the first indication information fed back by the terminal based on the content of the first indication information or the first timing. For configuration information 2 or configuration 3, the network device can determine the predicted value corresponding to the first indication information fed back by the terminal based on the first timing. The above configuration information will be described in conjunction with the content of the first indication information in the following step S620.

[0149] S620: The terminal sends a first instruction message to the network device at the first opportune moment, and the network device receives the first instruction message accordingly.

[0150] The first indication information is used to determine the prediction performance of the first model when predicting network parameters. The first model is a model used for time-domain prediction. For example, the first model can be an AI model or an ML model, etc. The network device can make decisions based on the first indication information, such as instructing the terminal to switch, or instructing the terminal to fallback to a non-AI mode, or instructing the terminal to switch models / functions, or instructing the terminal to deactivate models / functions.

[0151] The first indication information may include the error between the true value or the predicted value obtained from the parameters of the first model prediction network and the corresponding true value.

[0152] The design of the first indication information will be introduced below, taking into account the different configurations of the above configuration information.

[0153] First, for the case where the configuration information is configured as configuration 1, the first instruction information may include either instruction content 1 or instruction content 2.

[0154] 1. Instructions 1

[0155] In instruction content 1, the first instruction information includes at least one index information and a first error corresponding to the index information. The first error is the difference between the predicted value obtained by the first model predicting the network parameters and the true value of the corresponding network parameters.

[0156] For example, indication content 1 can be a list containing at least one element. Each element includes two features: an index and a corresponding first error. The value of the index is time-related, and for example, the index values ​​can increment sequentially over time. For instance, taking RSRP prediction as an example, the first element in the list is {0000, first error #1}, indicating that the terminal detected a first error exceeding the first threshold configured in configuration 1 (i.e., the measurement that met the timer start condition). The second element is {0001, first error #2}, indicating the first error detected by the terminal after the timer was started, and so on, constituting indication content 1.

[0157] The index information can indicate the order in which the first error is determined. The network device can determine the performance change trend of the first model based on the multiple first errors in the first indication information that are related by order through the index information, and then manage the first model.

[0158] 2. Instruction Content 2

[0159] In indication content 2, the first indication information includes a second error, which is the result of filtering at least one first error in indication content 1. The terminal can be pre-configured with a filtering algorithm to filter the measured at least one first error, resulting in a single error, namely the second error. In other words, indication content 2 includes one error. The filtering algorithm can be set to calculate the mean, mode, etc.

[0160] In one example, the filtering algorithm can be expressed as: Fn = (1-a)*Fn-1 + a*Dn, where Dn represents the most recently calculated prediction error, Fn represents the filtered result, Fn-1 represents the previous filtering result, and a is the filtering coefficient or a parameter calculated based on the filtering coefficient. The filtering coefficient can be pre-configured by the protocol or configured by the network device. Optionally, if the filtering coefficient is configured by the network device, the network device can send the filtering coefficient to the terminal in S610.

[0161] As for the case where the first indication information includes the second error, since the second error is the result of filtering at least one of the first errors, the second error can reflect the performance of the model over a period of time, thus avoiding false positives or false negatives caused by managing the model based on the error at a single moment.

[0162] Secondly, for cases where the configuration information is configured as configuration 2 or configuration 3, the first indication information may include the following indication content 3:

[0163] 3. Instructions 3

[0164] In instruction content 3, the first instruction information includes the first true value of the network parameter, wherein the predicted value obtained by the first model predicting the network parameter corresponding to the first true value is the predicted value at the second time. The correspondence between the true value and the predicted value at the second time is determined according to the first parameter and the second period. The second period is the period for sending the predicted value (i.e., T1 mentioned above). The first parameter is used to indicate the prediction window size of the first model. The predicted value at the second time is the predicted value sent before the first time.

[0165] When the first instruction information includes instruction content 3, the network device can receive the first instruction information sent by the terminal at the first opportune moment, and determine the predicted value corresponding to the first instruction information fed back by the terminal according to the second period and the first parameter, so as to effectively manage the model.

[0166] Specifically, for indication content 3 in configuration 2, the true value corresponds to the predicted value reported by the Nth terminal from the end, where N = [prediction window size / period length of the reported predicted value] + 1, and [] indicates rounding down. For example, if the prediction window size is 160ms and T1 is 240ms, then the true value returned by the terminal corresponds to the predicted value reported by the [160 / 240] + 1 = 1st terminal from the end, which is the most recent reported result.

[0167] For instruction content 3 in configuration 3, the actual value corresponds to the predicted value reported by the Mth terminal from the end, where M = [prediction window size / period length of the reported predicted value] + 2, and [] indicates rounding down. For example, if the prediction window size is 320ms and T1 is 240ms, then the actual value returned by the terminal corresponds to the [320 / 240] + 2 = 3rd reported predicted value from the end. In another example, if the prediction window size is 160ms and T1 is 240ms, then the actual value returned by the terminal corresponds to the [160 / 240] + 2 = 2nd reported predicted value from the end.

[0168] As can be seen, in configuration 3 compared to configuration 2, the predicted values ​​corresponding to the actual values ​​reported by the terminal differ. This is because in configuration 3, the terminal reports the predicted value along with the actual value. Therefore, the last predicted value reported by the terminal is the predicted value reported along with the actual value. Consequently, the actual value corresponds to the predicted value reported by the Mth terminal from the end. Configuration 2 does not have this requirement. Therefore, when the prediction window size and T1 are the same, M is 1 more than N.

[0169] In this embodiment of the application, for a scenario where the network configuration terminal periodically reports predicted values, the network device configures configuration information to the terminal indicating a first opportune moment. The terminal can then send first indication information based on this first opportune moment to determine the prediction performance of the first model when predicting network parameters. This enables the network device to correctly execute model control based on the first opportune moment and / or the content of the first indication information, ensuring system performance.

[0170] In one embodiment, the configuration information is set to the above-described configuration 2, as shown in FIG7. After S620, the method may further include:

[0171] S630, the network device sends a first duration to the terminal, and correspondingly, the terminal receives the first duration from the network device.

[0172] Specifically, the network device can determine when to send the first duration to the terminal by referring to the same principle as the first timing defined in Configuration 1 above. Specifically, the network device can set (or adopt a protocol-defined) a second threshold similar to the first threshold. When the network device determines that the difference (error) between the predicted value obtained by the first model predicting network parameters and the corresponding actual value of the network parameters is greater than the second threshold, it considers that the performance of the first model is likely poor. At this time, the network device can instruct the terminal to send the error at a higher frequency than T1 above, hoping to determine whether the performance of the first model has indeed deteriorated based on the error received at a higher frequency, and to manage the first model in a timely manner. At this time, the terminal is instructed to send the first duration, during which the terminal can continuously determine the error.

[0173] Optionally, the network device may also send an indication message to the terminal to indicate the start time of the first duration.

[0174] S640, the terminal determines at least one error within the first duration.

[0175] After receiving the first duration, the terminal begins to determine at least one error within the first duration.

[0176] As an alternative method, if the terminal receives an indication of the start time of the first duration when it receives the first duration, the terminal may begin determining at least one error within the first duration after the start time indicated by the indication has been reached.

[0177] S650, the terminal sends a second indication message indicating at least one error to the network device, and the network device receives the second indication message accordingly.

[0178] Once the terminal determines at least one error within the first duration, it can send it to the network device. The network device can then promptly determine and manage the performance of the first model based on at least one error. The design of the second instruction information can refer to the description of instruction content 1 or instruction content 2 of the first instruction information, and will not be repeated here.

[0179] In this embodiment, the network device can control the model based on the error over a period of time, thereby avoiding false positives or false negatives of the model, reducing unnecessary model control, and ensuring system performance.

[0180] In one embodiment, as shown in FIG8, after S620, the method may further include:

[0181] S660, the network device sends a second duration to the terminal, and correspondingly, the terminal receives the second duration from the network device.

[0182] In this process, the network device can determine the performance of the first model based on the first indication information received in S620. If the performance of the first model is found to be superior, it means that the prediction value of the first model is reliable. At this time, a second duration can be sent to the terminal, expecting the terminal to send the prediction value during the second duration. This allows the network device to know the prediction value for a future period of time, thereby making better handover decisions, etc.

[0183] Optionally, the timing for the network device to send the second duration can also be triggered by an event, such as events A2, A3, A4, etc., without restriction.

[0184] As an alternative, the second duration sent by the network device can be replaced by a third period, which defines the start time and length of the period. The third period is sent to the terminal, expecting the terminal to send the predicted value based on the third period. Alternatively, the second duration sent by the network device can be replaced by a second duration and a third period, sending both to the terminal, expecting the terminal to send the predicted value based on the third period within the second duration. In this case, the third period is shorter than T1 mentioned above.

[0185] S670, the terminal continuously sends the predicted values ​​of the network parameters obtained by the first model to the network device during the second duration, and the network device receives the predicted values ​​of the network parameters obtained by the first model.

[0186] After receiving the second duration, the terminal can send a prediction value to the network device once for each prediction made within the second duration, until the second duration ends.

[0187] As an alternative, if the terminal receives the third cycle, it can send a predicted value based on the third cycle. Specifically, starting from the start time of the third cycle, the predicted value is sent once every three cycles. Alternatively, if the terminal receives the second duration and the third cycle, it can send a predicted value based on the third cycle during the second duration until the second duration ends.

[0188] In this embodiment, the network device sends a second duration to the terminal, and the terminal continuously sends predicted values ​​during the second duration. This allows the network device to know the predicted values ​​for a future period, thereby enabling it to make better decisions such as handover decisions.

[0189] As can be seen from the above description, this application embodiment addresses the scenario where a network-configured terminal periodically reports prediction results. The network device configures configuration information to the terminal indicating a first opportune moment. Based on this first opportune moment, the terminal can send first indication information to determine the prediction performance of the first model when predicting network parameters. This enables the network device to correctly execute model control, ensuring system performance. Furthermore, the network device can also control the model based on the first indication information over a period of time, thereby avoiding false positives or false negatives, reducing unnecessary model control, and ensuring system performance.

[0190] The above describes the implementation of the communication method provided in this application embodiment for scenarios where network configuration terminals periodically report predicted values. This application embodiment also provides a communication method for scenarios where event-triggered terminals report predicted values. For example, as shown in Figure 9, this method may include the following steps:

[0191] S910, the terminal sends the first predicted value to the network device, and the network device receives the first predicted value accordingly.

[0192] When an event occurs, the terminal triggers the transmission of a first predicted value. This first predicted value is the result of a first model predicting network parameters; the first model is a model used for time-domain prediction. The aforementioned events are typically predefined conditions or thresholds in the network. When these conditions are met or the thresholds are exceeded, the transmission of the first predicted value is triggered. For example, these events could be events A2, A3, A4, etc.

[0193] A2 event: This is a quality event related to the serving cell. The A2 event is triggered when the radio signal quality of the serving cell falls below a preset threshold (such as RSRP, RSRQ, etc.). At this time, the terminal may begin searching for other possible cells and prepare for cell reselection or handover.

[0194] Event A3: This is an event related to neighboring cells, typically used to compare the signal quality of the serving cell and neighboring cells. The A3 event is triggered when the signal quality of a neighboring cell is better than that of the serving cell, and by a certain margin (exceeding a certain offset value). At this point, the terminal may initiate a handover request to that neighboring cell.

[0195] A4 event: This is a frequency- or radio access technology (RAT) based event. An A4 event is triggered when the terminal detects better signal quality on a specific frequency or RAT. Unlike the A3 event, the A4 event is typically associated with inter-frequency or inter-system handover.

[0196] When any of the above events is triggered, the terminal will not only perform the corresponding action (such as cell search, measurement report, handover request, etc.), but will also trigger the transmission of the first predicted value. The first predicted value is predicted by the first model and is used to predict network parameters over a future period of time.

[0197] In one example, for event A2, when the terminal's predicted serving cell RSRP is lower than a threshold, the terminal reports the predicted serving cell RSRP. As another example, for event A3, when the difference between the terminal's predicted neighboring cell RSRP and the predicted serving cell RSRP is higher than a threshold, the terminal reports both the predicted neighboring cell RSRP and the predicted serving cell RSRP.

[0198] The transmission of first-prediction values ​​is crucial for network optimization and resource management. By receiving these values, the network can anticipate terminal demands and changes in the network environment, enabling more rational resource allocation and scheduling decisions. For example, if network devices receive first-prediction values ​​from multiple terminals indicating a rapid increase in load on a particular cell, the network devices can proactively increase the cell's capacity or guide some terminals to switch over, thus avoiding congestion and performance degradation.

[0199] S920: The terminal sends a first instruction message to the network device at the first opportune moment, and the network device receives the first instruction message accordingly.

[0200] The first indication information is used to determine the prediction performance of the first model when predicting network parameters. The network device can make decisions based on the first indication information. For example, the decision may be to instruct the terminal to switch, or to instruct the terminal to fallback to a non-AI mode, or to instruct the terminal to switch models / functions, or to instruct the terminal to deactivate models / functions.

[0201] The first timing can be configured as follows:

[0202] 1. Configuration 4

[0203] The first opportunity is the time when the error corresponding to the predicted value within the prediction window of the first model is measured. The prediction window is the prediction window for predicting the first predicted value. In other words, after sending the first predicted value to the network device, the terminal feeds back the error sequentially. The time point at which the prediction difference corresponding to the predicted value within the prediction window for each measured prediction of the first predicted value is the first opportunity. For example, the error fed back for the first time corresponds to the first predicted value within the prediction window, the error fed back for the second time corresponds to the second predicted value within the prediction window, and so on. Optionally, the terminal also feeds back an index, indicating the position of the predicted value corresponding to the error within the prediction window.

[0204] When the first timing is configured as 4, the first indication information includes a first error corresponding to the predicted value within the prediction window. Optionally, the first indication information may also include index information indicating the position of the predicted value within the prediction window.

[0205] In the configuration 4 scenario, the terminal can provide multiple feedbacks of the first error for a period of time corresponding to the prediction window. The network device can then manage the model based on the first error for that period of time, avoiding false positives or false negatives caused by managing the model based on a single error.

[0206] 2. Configuration 5

[0207] The first opportune moment is when the error corresponding to the first predicted value is measured. In other words, after the terminal sends the first predicted value to the network device, it continues to determine the error until the error corresponding to the first predicted value is determined; this moment is considered the first opportune moment.

[0208] When the first timing is configuration 5, the first indication information includes at least one index information and a first error corresponding to the index information. The first error is the difference between the predicted value obtained by the first model predicting the network parameters and the true value of the corresponding network parameters.

[0209] For example, the first indication information can be a list, which includes at least one element. Each element includes two features: an index and a corresponding first error. The value of the index is associated with the position of the predicted value corresponding to the error within the prediction window for the first predicted value. For instance, taking the prediction of RSRP as an example, the first element in the list is {0001, first error #1}, indicating that the predicted value corresponding to first error #1 is the first in the prediction window; the second element is {0002, first error #2}, indicating that the predicted value corresponding to first error #2 is the second in the prediction window, and so on.

[0210] As an alternative, the first indication information may also include a second error. The second error is the result of filtering at least one first error, continuously determined by the terminal after sending the first predicted value to the network device; that is, only one prediction error value is fed back. The filtering algorithm used during filtering can be set to calculate the mean, mode, etc.

[0211] In one example, the filtering algorithm can be expressed as: Fn = (1-a)*Fn-1 + a*Dn, where Dn represents the most recently calculated prediction error, Fn represents the filtered result, Fn-1 represents the previous filtering result, and a is the filtering coefficient or a parameter calculated based on the filtering coefficient. The filtering coefficient can be pre-configured by the protocol or configured by the network device.

[0212] In the above-mentioned configuration schemes for the first timing, the terminal can predict either the measurement result or the result after L3 filtering.

[0213] Figure 10 illustrates a scenario where the terminal reports the predicted value based on the measurement result. As shown in Figure 10, at the current time t... n At that time, the future t was reported. n+1 t n+2 and t n+3 The predicted value, for the first timing when configuration 4, when time t... n+1 At that time, t was measured n+1 The actual value at that time, at this point, report t n+1 The true value at time and the corresponding t n The error between the predicted values ​​at time t, similarly, when time t arrives. n+2 At that time, t was measured n+2 The actual value at that time, at this point, report tn+2 The true value at time and the corresponding t n The error between the predicted values ​​and the predicted values ​​will be discussed further, and so on.

[0214] When the first timing is configuration 5, assuming the first prediction value is in the future t... n+3 The predicted value, then when time t arrives... n+3 At that time, t was measured n+1 t n+2 and t n+3 The actual value at that time, at this point, report t n+1 t n+2 and t n+3 The actual values ​​measured at time and their corresponding t values n The error between the predicted values ​​or the result after error filtering is reported.

[0215] Figure 11 illustrates a scenario where the terminal reports the predicted value when the measurement result is obtained after L3 filtering. As shown in Figure 11, at the current time t... n-2 At that time, the future t was reported. n+1 The predicted value; at the current time t n-1 At that time, the future t was reported. n+2 The predicted value; at the current time t n At that time, the future t was reported. n+3 The predicted value. For the first timing condition being configuration 4, when time t arrives... n+1 At that time, t was measured n+1 The actual value at that time, at this point, report t n+1 The true value at time and the corresponding t n-2 The error between the predicted values ​​at time t, similarly, when time t arrives. n+2 At that time, t was measured n+2 The actual value at that time, at this point, report t n+2 The true value at time and the corresponding t n-1 The error between the predicted values ​​and the predicted values ​​will be discussed further, and so on.

[0216] When the first timing is configuration 5, assuming the first prediction value is in the future t... n+3 The predicted value, then when time t arrives... n+3 At that time, t was measured n+1 t n+2 and t n+3 The actual value at that time, at this point, report t n+1 t n+2 and t n+3 The actual values ​​measured at time and their corresponding t values n-2 t n-1 t nThe error between the predicted values ​​or the result after error filtering is reported.

[0217] In scenario 5, the terminal can provide the first error for a given period of time corresponding to the prediction window. The network device can then manage the model based on this first error, avoiding false positives or false negatives caused by managing the model based on a single error. Alternatively, the terminal can provide the second error corresponding to the prediction window. Since the second error is the result of filtering at least one first error, it reflects the model's performance over a period of time, avoiding false positives or false negatives caused by managing the model based on errors at a single moment.

[0218] 3. Configuration 6

[0219] The first opportune moment is when the first predicted value is sent. In other words, the terminal sends the first predicted value and the first indication information simultaneously. At this time, S910 and S920 can be executed together, and the first predicted value and the first indication information are carried in the same message and sent to the network device.

[0220] In this scenario, the first indication information includes the first true value. It should be understood that the first predicted value sent simultaneously by the terminal does not correspond to the first true value; the predicted value corresponding to the first true value is the second predicted value sent by the terminal at a second time point prior to the first time point. At this time, the network device can determine that the predicted value corresponding to the first indication information fed back by the terminal is the predicted value preceding the first predicted value, thereby correctly executing the model's control and ensuring system performance.

[0221] In this embodiment of the application, for the case where the first timing is the timing of the error corresponding to the predicted value within the prediction window of each measured first model, since the first timing is defined as the timing of the error corresponding to the predicted value within the prediction window of each measured first model, the terminal can provide the first indication information for a period of time corresponding to the prediction window multiple times. The network device can perform model management based on the first indication information for a period of time corresponding to the prediction window, thus avoiding false positives or false negatives caused by managing the model based on a single error.

[0222] When the first opportunity is the time when the error corresponding to the first predicted value is measured, the terminal can provide feedback on the first indication information for a period of time corresponding to the prediction window. The network device can use the first indication information for a period of time corresponding to the prediction window for model management, thus avoiding false positives or false negatives caused by managing the model based on a single error.

[0223] When the first timing is the timing of sending the first predicted value, the network device can determine the predicted value corresponding to the first indication information sent by the terminal when it receives the first indication information sent by the terminal at the first timing. That is, the predicted value corresponding to the first indication information is the predicted value before the first predicted value, so as to effectively manage the model.

[0224] In one embodiment, the first timing is the timing when the error corresponding to the predicted value within the prediction window of the first model is measured, or the first timing is the timing when the error corresponding to the first predicted value is measured.

[0225] In one embodiment, for the scenario where the first timing is the above-described configuration 6, that is, the first timing is the timing of sending the first predicted value; as shown in FIG12, before step S910, the method may further include:

[0226] S930: The network device sends a third instruction message to the terminal, and the terminal receives the third instruction message accordingly.

[0227] In this context, the first timing, configuration 6, can be indicated to the terminal by the network device via third indication information, for example, similar to the design when the configuration information is configuration 3 in the embodiment shown in Figure 6. The network device instructs the terminal to send the first indication information simultaneously with the predicted value obtained from the first model prediction network parameters, via the third indication information.

[0228] In one embodiment, as shown in FIG13, after S920, the method may further include:

[0229] S940, the network device sends a second duration to the terminal, and correspondingly, the terminal receives the second duration from the network device.

[0230] In this system, the network device can determine the performance of the first model based on the first indication information received in S920. If the performance of the first model is found to be superior, it means that the prediction value of the first model is reliable. At this time, a second duration can be sent to the terminal, expecting the terminal to send the prediction value during the second duration. This allows the network device to know the prediction value for a future period of time, thereby making better handover decisions, etc.

[0231] Optionally, the timing for the network device to send the second duration can also be triggered by an event, such as events A2, A3, A4, etc., without restriction.

[0232] As an alternative, the second duration sent by the network device can be replaced with a third period, which defines the start time and length of the period. The third period is sent to the terminal in the expectation that the terminal will send the predicted value based on the third period. Alternatively, the second duration sent by the network device can be replaced with both a second duration and a third period. Both the second duration and the third period are sent to the terminal in the expectation that the terminal will send the predicted value based on the third period within the second duration.

[0233] S950, the terminal continuously sends the predicted values ​​of the network parameters obtained by the first model to the network device during the second duration, and the network device receives the predicted values ​​of the network parameters obtained by the first model.

[0234] After receiving the second duration, the terminal can send a prediction value to the network device once for each prediction made within the second duration, until the second duration ends.

[0235] As an alternative, if the terminal receives the third cycle, it can send a predicted value based on the third cycle. Specifically, starting from the start time of the third cycle, the predicted value is sent once every three cycles. Alternatively, if the terminal receives the second duration and the third cycle, it can send a predicted value based on the third cycle during the second duration until the second duration ends.

[0236] In this embodiment, the network device sends a second duration to the terminal, and the terminal continuously sends predicted values ​​during the second duration. This allows the network device to know the predicted values ​​for a future period, thereby enabling it to make better decisions such as handover decisions.

[0237] Based on the above description, it is clear that this application sets several first opportunities for the terminal to send the aforementioned first instruction information in scenarios where prediction results are reported based on events. This enables network devices to correctly execute model control and ensure system performance. Furthermore, network devices can also control the model based on the first instruction information over a period of time, thereby avoiding false positives or false negatives, reducing unnecessary model control, and ensuring system performance.

[0238] The above mainly describes the solution provided by the embodiments of this application from the perspective of the execution logic of each step. To aid understanding, the communication method provided by this application will be further described below in a specific implementation scenario with reference to the communication system shown in Figure 1 of the embodiments of this application. In the following description, the explanation of each step and its beneficial effects can be found in the description of the corresponding steps in Figures 6-13 of the embodiments of this application, which will not be repeated here. Instead, the focus will be on explaining the specific implementation of the process to help better understand this application.

[0239] In one embodiment, the communication method provided in this application can be applied to the O-RAN architecture shown in Figure 5. In this architecture, network devices may include gNB-CU network elements, gNB-DU network elements, and a management node, where the management node can be a RIC or SU. Figure 14 shows a flowchart of the communication method under this architecture. As shown in Figure 14, the method may include the following steps:

[0240] S141, the terminal sends the first instruction information to the control node through the gNB-DU network element and the gNB-CU network element.

[0241] The first indication information may include an error or a true value, a description of the first indication information and its sending timing, for example, the three first timings indicated by configuration information as described in the embodiment shown in Figure 6, or the three first timings as described in the embodiment shown in Figure 9. Please refer to the description of the corresponding steps in the embodiments of Figures 6-13, which will not be repeated here.

[0242] S142, the control node generates a decision instruction based on the first instruction information and sends the decision instruction to the terminal through the gNB-DU network element and the gNB-CU network element.

[0243] The first indication information can be used to determine the prediction performance of the first model when predicting network parameters. The control node can generate decision instructions based on the first indication information. For example, the decision instructions can be to instruct the terminal to switch, or to instruct the terminal to fallback to non-AI mode, or to instruct the terminal to switch models / functions, or to instruct the terminal to deactivate models / functions.

[0244] In one implementation, the terminal reports an error. Upon receiving the error from the terminal, the gNB-DU forwards it to the gNB-CU, which then forwards it to the control node. Various methods by which the terminal sends the error to the gNB-DU, and various methods by which the control node maps the error to the predicted value, can be found in the descriptions of the embodiments shown in Figure 6 or Figure 9, and will not be repeated here.

[0245] In another implementation, the terminal feeds back the actual value. After receiving the actual value from the terminal, the gNB-DU forwards it to the gNB-CU or directly to the control node. The various ways in which the terminal sends the actual value to the gNB-DU, as well as the correspondence between the actual value and the predicted value, can be referred to the relevant descriptions in the embodiments shown in Figure 6 or Figure 9, and will not be repeated here.

[0246] When gNB-DU forwards the real value to gNB-CU, gNB-CU has the following processing options after receiving the real value:

[0247] Method 1: gNB-CU directly sends the actual value to the control node, which then calculates the error.

[0248] In Method 1, for cases where the terminal periodically reports predicted values ​​based on the first period, the terminal needs to first inform the control node of the first period for reporting predicted values. This allows the control node to determine which predicted value corresponds to the actual value based on the first period and its own stored prediction window size. Informing the control node of the first period can be achieved through the following optional steps:

[0249] S143, the terminal sends the first cycle of the reported predicted value to the control node through gNB-DU and gNB-CU.

[0250] Method 2: The gNB-CU calculates the error based on the actual value reported by the terminal and then sends the error to the control node. For cases where the terminal periodically reports the predicted value based on the first cycle, the control node needs to inform the gNB-CU of the prediction window size. This allows the gNB-CU to determine the predicted value corresponding to the actual value based on the prediction window size and the first cycle received in S142. Informing the prediction window size can be achieved through the following optional steps:

[0251] S144, the control node sends the prediction window size to gNB-CU.

[0252] Optionally, in the case where the control node configures the terminal with a first cycle, i.e., the case shown in the embodiment of Figure 7, when the error exceeds the second threshold, the control node can send a first duration to the terminal via gNB-DU and / or gNB-CU, and the terminal can determine at least one error within the first duration. For a detailed explanation, please refer to the description of the embodiment shown in Figure 7, which will not be repeated here. In this case, the method may include the following optional steps:

[0253] S145, the control node sends the first duration to the terminal via gNB-DU and / or gNB-CU.

[0254] S146, the terminal determines at least one error within the first duration.

[0255] After receiving the first duration, the terminal begins to determine at least one error within the first duration.

[0256] S147, the terminal sends a second indication message indicating at least one error to the control node via gNB-DU and / or gNB-CU.

[0257] Once the terminal determines at least one error within the first duration, it can send it to the control node. The control node can then determine the performance of the first model in a timely manner based on at least one error and manage it accordingly.

[0258] In this optional implementation, the control node can control the model based on the error over a period of time, thereby avoiding false positives or false negatives of the model, reducing unnecessary model control, and ensuring system performance.

[0259] As can be seen from the above description, the communication method provided in this application embodiment is applicable to the O-RAN architecture. Specifically, the terminal can send first indication information for determining the prediction performance of the first model prediction network parameters based on a first opportune moment. This enables the control node to correctly execute model control and ensure system performance. Furthermore, the control node can also control the model based on the first indication information over a period of time, thereby avoiding false positives or false negatives of the model, reducing unnecessary model control, and ensuring system performance.

[0260] The above mainly describes the solution provided by the embodiments of this application from the perspective of the execution logic of each step. It is understood that each node, such as a terminal, includes corresponding hardware structures and / or software modules to execute each function in order to achieve the above functions. Those skilled in the art should readily recognize that, in conjunction with the algorithm steps of the examples described in the embodiments disclosed herein, the method of the embodiments of this application can be implemented in hardware, software, or a combination of hardware and computer software. Whether a function is executed in a hardware or computer software-driven hardware manner depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0261] This application embodiment can divide the terminal into functional modules according to the above method example. For example, each function can be divided into its own functional module, or two or more functions can be integrated into one processing module. The integrated module can be implemented in hardware or as a software functional module. It should be noted that the module division in this application embodiment is illustrative and only represents one logical functional division. In actual implementation, there may be other division methods.

[0262] In specific implementations, the network elements shown in this application, such as terminals, can adopt the composition structure shown in Figure 15 or include the components shown in Figure 15. Figure 15 is a schematic diagram of the structure of a communication device provided in an embodiment of this application. When the communication device has the functions of the terminal described in the embodiment of this application, the communication device can be a terminal or a chip or system-on-a-chip in the terminal. When the communication device has the functions of the network device described in the embodiment of this application, the communication device can be a network device or a chip or system-on-a-chip in the network device.

[0263] For example, Figure 15 illustrates a possible structural schematic of a communication device. It is understood that the communication device 110 includes means of the necessary form, such as modules, units, elements, circuits, or interfaces, to be appropriately configured together to execute this solution. The communication device 110 may be a terminal or network device as described in the above method embodiments, or it may be a component (e.g., a chip) in these devices used to implement the methods described in the above method embodiments. The communication device 110 includes one or more processors 111. The processor 111 may be a general-purpose processor or a dedicated processor, for example, a baseband processor or a central processing unit. The baseband processor can be used to process communication protocols and communication data, while the central processing unit can be used to control the communication device, execute software programs, and process data from the software programs.

[0264] Optionally, in one design, the processor 111 may include a program 113 (sometimes also referred to as code or instructions) that can be executed on the processor 111 to cause the communication device 110 to perform the methods described in the embodiments below. In yet another possible design, the communication device 110 includes circuitry (not shown in FIG15) for implementing the signal processing functions described in the above embodiments.

[0265] Optionally, the communication device 110 may include one or more memories 112 storing a program 114 (sometimes referred to as code or instructions), which can be run on the processor 111 to cause the communication device 110 to perform the methods described in the above method embodiments.

[0266] Optionally, the processor 111 and / or memory 112 may include AI modules 117 and 118, which are used to implement AI-related functions. The AI ​​modules can be implemented through software, hardware, or a combination of both. For example, the AI ​​module may include a RIC module. For example, the AI ​​module may be a near real-time RIC or a non-real-time RIC.

[0267] Optionally, the processor 111 and / or memory 112 may also store data. The processor and memory may be configured separately or integrated together.

[0268] Optionally, the communication device 110 may further include a transceiver 115 and / or an antenna 116. The processor 111, sometimes referred to as a processing unit, controls the communication device. The transceiver 115, sometimes referred to as a transceiver unit, transceiver, transceiver circuit, or transceiver, is used to realize the transmission and reception functions of the communication device through the antenna 116.

[0269] Figure 16 shows a structural diagram of a communication device 16 applied to a terminal. Each module in the device shown in Figure 16 performs the functions corresponding to the steps in Figures 6-9 and achieves the corresponding technical effects. The beneficial effects of each module performing the steps can be referred to the descriptions of the corresponding steps in Figures 6-9, and will not be repeated here. The functions can be implemented by hardware or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions. This communication device can be a terminal or a chip or system-on-a-chip in the terminal. For example, the communication device includes:

[0270] The transceiver module 161 is configured to receive configuration information, wherein the configuration information is used to indicate a first timing; and to send first indication information at the first timing, wherein the first indication information is used to determine the prediction performance when the first model predicts the network parameters.

[0271] In one embodiment, the configuration information includes a first threshold and a timer value. When the difference between the predicted value obtained by the first model predicting the network parameters and the actual value of the corresponding network parameters is greater than the first threshold, the timer is started, and the expiration time of the timer is a first timing.

[0272] Alternatively, the configuration information may include a first cycle, where the timing indicated by the first cycle is the first timing.

[0273] Alternatively, the configuration information indicates that a first indication information is sent when the predicted value obtained from the first model prediction network parameters is sent, and the timing of sending the predicted value is the first timing.

[0274] In one embodiment, the first indication information includes at least one index information and a first error corresponding to the index information, wherein the first error is the difference between the predicted value obtained by the first model predicting the network parameters and the true value of the corresponding network parameters.

[0275] Alternatively, the first indication information may include a second error, which is the result of filtering at least one first error.

[0276] In one embodiment, the configuration information includes a first period, the timing indicated by the first period being a first timing; or, the configuration information indicates that a first indication information is sent when the predicted value obtained from the first model prediction network parameters is sent, the timing of sending the predicted value being a first timing.

[0277] The first indication information includes the first true value of the network parameters, wherein the predicted value obtained by the first model predicting the network parameters corresponding to the first true value is the predicted value at the second time. The correspondence between the true value and the predicted value at the second time is determined according to the first parameter and the second period. The second period is the period for sending the predicted value. The first parameter is used to indicate the prediction window size of the first model. The predicted value at the second time is the predicted value sent before the first time.

[0278] In one embodiment, the configuration information includes a first period, and the timing that conforms to the first period is a first timing; a transceiver module 161 is used to receive a first duration from the network device; a processing module 162 is used to determine at least one prediction error within the first duration; and the transceiver module 161 is used to send second indication information to the network device, the second indication information indicating at least one prediction error.

[0279] In one embodiment, the transceiver module 161 is further configured to receive a second duration from the network device; and continuously transmit the predicted values ​​obtained from the first model predicting network parameters within the second duration.

[0280] In another embodiment, the transceiver module 161 is used to send a first predicted value; and to send first indication information at a first timing, wherein the first timing is the timing when the error corresponding to the predicted value within the prediction window of each first model is measured, and the prediction window is the prediction window for predicting the first predicted value; or, the first timing is the timing when the error corresponding to the first predicted value is measured; or, the first timing is the timing when the first predicted value is sent, and the first indication information is used to determine the prediction performance of the first model when predicting network parameters.

[0281] In one implementation, the first timing is the timing when the error corresponding to the predicted value within the prediction window of the first model is measured, or the first timing is the timing when the error corresponding to the first predicted value is measured.

[0282] The first indication information includes a first error corresponding to the predicted value within the prediction window; or, the first indication information includes at least one index information and a first error corresponding to the index information, wherein the first error is the difference between the predicted value obtained by the first model predicting the network parameters and the true value of the corresponding network parameters; or, the first indication information includes a second error, wherein the second error is the result of filtering at least one first error.

[0283] In one implementation, the first timing is the timing for sending the first predicted value; the transceiver module 161 is used to receive third indication information, which is used to indicate that the first predicted value and the first indication information are sent at the first timing.

[0284] In one embodiment, the transceiver module 161 is configured to send a second predicted value at a second time before sending the first indication information at a first time; wherein the first indication information includes a first true value, and the true value corresponding to the second predicted value is the first true value.

[0285] In one embodiment, the transceiver module 161 is configured to receive a second duration and continuously transmit the predicted values ​​obtained from the parameters of the first model prediction network within the second duration.

[0286] Figure 17 shows a structural diagram of a communication device 17, which is applied to a network device. Each module in the device shown in Figure 17 performs the functions corresponding to the steps in Figures 6-9 and achieves the corresponding technical effects. The beneficial effects of each module performing the steps can be referred to the descriptions of the corresponding steps in Figures 6-9, and will not be repeated here. The functions can be implemented by hardware or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions. This communication device can be a network device or a chip or system-on-a-chip in a network device. For example, the communication device includes:

[0287] The transceiver module 171 is used to send configuration information, wherein the configuration information is used to indicate the first timing for sending the first indication information, and the first indication information is used to determine the prediction performance when the first model predicts the network parameters; and to receive the first indication information.

[0288] In one embodiment, the configuration information includes a first threshold and a timer. When the difference between the predicted value obtained by the first model predicts the network parameters and the actual value of the corresponding network parameters is greater than the first threshold, the timer is started, and the expiration time of the timer is a first opportunity. The time when the timer expires based on the timer start condition is the first opportunity. Alternatively, the configuration information includes a first period, and the opportunity indicated by the first period is the first opportunity. Alternatively, the configuration information indicates that a first indication information is sent when the predicted value obtained by the first model predicts the network parameters is sent, and the opportunity when the predicted value is sent is the first opportunity.

[0289] In one embodiment, the first indication information includes at least one index information and a first error corresponding to the index information, wherein the first error is the difference between the predicted value obtained by the first model predicting the network parameters and the true value of the corresponding network parameters; or, the first indication information includes a second error, wherein the second error is the result of filtering at least one first error.

[0290] In one embodiment, the configuration information includes a first period, the timing indicated by the first period being a first timing; or, the configuration information indicates that a first indication information is sent when the predicted value obtained by the first model predicts the network parameters is sent, the timing of sending the predicted value is the first timing; the first indication information includes a first true value of the network parameters, wherein the predicted value obtained by the first model predicts the network parameters corresponding to the first true value is the predicted value of the second timing, the correspondence between the true value and the predicted value of the second timing is determined according to the first parameter and the second period, the second period is the period for sending the predicted value, the first parameter is used to indicate the prediction window size of the first model, and the predicted value of the second timing is the predicted value sent before the first timing.

[0291] In one embodiment, the configuration information includes a first period, and the timing that conforms to the first period is a first timing; a transceiver module 171 is used to send a first duration, the first duration being used to determine at least one prediction error; and to receive second indication information, the second indication information indicating at least one prediction error.

[0292] In one embodiment, the transceiver module 171 is configured to send a second duration, wherein the second duration is used to send the predicted value obtained from the parameters of the first model prediction network; and to receive the predicted value obtained from the parameters of the first model prediction network.

[0293] In another embodiment, the transceiver module 171 is configured to receive a first predicted value; receive first indication information, wherein the first indication information is sent at a first timing, the first timing being the timing when the error corresponding to the predicted value within the prediction window of each measured first model is obtained, the prediction window being the prediction window for predicting the first predicted value; or, the first timing being the timing when the error corresponding to the first predicted value is obtained; or, the first timing being the timing when the first predicted value is sent, the first indication information being used to determine the prediction performance of the first model when predicting network parameters.

[0294] In one implementation, the first timing is the timing when the error corresponding to the predicted value within the prediction window of the first model is measured, or the first timing is the timing when the error corresponding to the first predicted value is measured; the first indication information includes the first error corresponding to the predicted value within the prediction window; or, the first indication information includes at least one index information and the first error corresponding to the index information, the first error being the difference between the predicted value obtained by the first model predicting network parameters and the true value of the corresponding network parameters; or, the first indication information includes a second error, the second error being the result after filtering at least one first error.

[0295] In one implementation, the first timing is the timing for sending the first predicted value; the transceiver module 171 is used to send third indication information, which is used to indicate that the first predicted value and the first indication information are sent at the first timing.

[0296] In one embodiment, the transceiver module 171 is configured to receive a second predicted value before receiving the first indication information; wherein the second predicted value is sent at a second time, the first indication information includes a first true value, and the true value corresponding to the second predicted value is the first true value.

[0297] In one embodiment, the transceiver module 171 is used to send a second duration, wherein the second duration is used to send the predicted value obtained by the first model prediction network parameters; and to receive the predicted value obtained by the first model prediction network parameters.

[0298] This application embodiment also provides a communication system, which is a communication system corresponding to a model management scenario. The communication system includes a terminal and a network device, wherein the terminal may have the functions of the communication device shown in FIG16, and the network device may have the functions of the communication device shown in FIG17.

[0299] This application also provides a computer-readable storage medium. All or part of the processes in the above method embodiments can be implemented by a computer program instructing related hardware. This program can be stored in the computer-readable storage medium, and when executed, it can include the processes of the above method embodiments. The computer-readable storage medium can be a terminal device of any of the foregoing embodiments, such as an internal storage unit including a data sending end and / or a data receiving end, such as a hard disk or memory of the terminal device. The computer-readable storage medium can also be an external storage device of the terminal device, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc., equipped on the terminal device. Further, the computer-readable storage medium can include both the internal storage unit and the external storage device of the terminal device. The computer-readable storage medium is used to store the computer program and other programs and data required by the terminal device. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0300] This application also provides computer instructions. All or part of the processes in the above method embodiments can be executed by computer instructions to instruct related hardware (such as computers, processors, network devices, and terminals). The program can be stored in the aforementioned computer-readable storage medium.

[0301] This application also provides a chip system. The chip system may consist of chips or include chips and other discrete devices, without limitation. The chip system includes a processor and a transceiver. All or part of the processes in the above method embodiments can be completed by this chip system, such as the chip system being used to implement the functions performed by the terminal or network device in the above method embodiments.

[0302] In one possible design, the chip system further includes a memory for storing program instructions and / or data. When the chip system is running, the processor executes the program instructions stored in the memory to enable the chip system to perform the functions performed by the terminal or network device in the above method embodiments.

[0303] In the embodiments of this application, the processor may be a general-purpose processor, a digital signal processor, an application-specific integrated circuit (ASIC), a field-programmable gate array (FPGA), or other programmable logic devices, discrete gate or transistor logic devices, or discrete hardware components, capable of implementing or executing the methods, steps, and logic block diagrams disclosed in the embodiments of this application. The general-purpose processor may be a microprocessor or any conventional processor. The steps of the methods disclosed in the embodiments of this application can be directly manifested as being executed by a hardware processor, or executed by a combination of hardware and software modules within the processor.

[0304] In the embodiments of this application, the memory can be non-volatile memory, such as a hard disk drive (HDD) or a solid-state drive (SSD), or it can be volatile memory, such as random-access memory (RAM). Memory is any other medium capable of carrying or storing desired program code in the form of instructions or data structures, and accessible by a computer, but is not limited thereto. The memory in the embodiments of this application can also be a circuit or any other device capable of implementing storage functions, used to store instructions and / or data.

[0305] It should be noted that the terms "first" and "second," etc., in the specification, claims, and drawings of this application are used to distinguish different objects, not to describe a specific order. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products, or apparatuses.

[0306] It should be understood that in the embodiments of this application, "at least one (item)" refers to one or more, "more than one" refers to two or more, "at least two (items)" refers to two or three or more, and "and / or" is used to describe the association relationship of related objects, indicating that there can be three relationships. For example, "A and / or B" can represent: only A exists, only B exists, and A and B exist simultaneously, where A and B can be singular or plural. The character " / " generally indicates that the related objects before and after are in an "or" relationship. "At least one (item) of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one (item) of a, b, or c can represent: a, b, c, "a and b", "a and c", "b and c", or "a and b and c", where a, b, and c can be single or multiple. It should be understood that in the embodiments of this application, "B corresponding to A" means that B is associated with A. For example, B can be determined based on A. It should also be understood that determining B based on A does not mean determining B solely based on A; B can also be determined based on A and / or other information. Furthermore, the term "connection" in the embodiments of this application refers to various connection methods, such as direct or indirect connections, to achieve communication between devices; the embodiments of this application do not impose any limitations on this.

[0307] Unless otherwise specified, the term "transmission" in the embodiments of this application refers to bidirectional transmission, encompassing the actions of sending and / or receiving. Specifically, "transmission" in the embodiments of this application includes sending data, receiving data, or both sending and receiving data. In other words, data transmission here includes uplink and / or downlink data transmission. Data may include channels and / or signals; uplink data transmission refers to uplink channel and / or uplink signal transmission, and downlink data transmission refers to downlink channel and / or downlink signal transmission. The terms "network" and "system" in the embodiments of this application refer to the same concept; a communication system is a communication network.

[0308] Through the above description of the embodiments, those skilled in the art can clearly understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0309] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another device, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between devices or units may be electrical, mechanical, or other forms.

[0310] The units described as separate components may or may not be physically separate. A component shown as a unit can be one or more physical units; that is, it can be located in one place or distributed in multiple different locations. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0311] Furthermore, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or as a software functional unit. If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a readable storage medium. Based on this understanding, the technical solution of the embodiments of this application, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This software product is stored in a storage medium and includes several instructions to cause a device, such as a microcontroller, chip, or processor, to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0312] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions within the technical scope disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

1. A communication method, characterized in that, include: Receive configuration information, wherein the configuration information is used to indicate a first timing; At the first opportune moment, a first indication message is sent, wherein the first indication message is used to determine the prediction performance of the first model when predicting network parameters, and the first model is a model used for time-domain prediction.

2. The communication method according to claim 1, characterized in that, The configuration information includes a first threshold and a timer value. When the difference between the predicted value obtained by the first model predicting the network parameters and the actual value of the corresponding network parameters is greater than the first threshold, the timer is activated, and the expiration time of the timer is the first timing. Alternatively, the configuration information may include a first period, wherein the timing indicated by the first period is the first timing. Alternatively, the configuration information indicates that a first indication information is sent when the predicted value obtained from the first model prediction network parameters is sent, and the timing of sending the predicted value is the first timing.

3. The communication method according to claim 1 or 2, characterized in that, The first indication information includes at least one index information and a first error corresponding to the index information. The first error is the difference between the predicted value obtained by the first model predicting the network parameters and the true value of the corresponding network parameters. Alternatively, the first indication information may include a second error, which is the result of filtering the at least one first error.

4. The communication method according to any one of claims 1-3, characterized in that, The configuration information includes a first period, the timing indicated by the first period being the first timing; or, the configuration information indicates that a first indication information is sent when the predicted value obtained from the first model prediction network parameters is sent, the timing of sending the predicted value being the first timing. The first indication information includes the first true value of the network parameters, wherein the predicted value obtained by the first model predicting the network parameters corresponding to the first true value is the predicted value at the second time. The correspondence between the true value and the predicted value at the second time is determined according to the first parameter and the second period. The second period is the period for sending the predicted value. The first parameter is used to indicate the prediction window size of the first model. The predicted value at the second time is the predicted value sent before the first time.

5. The communication method according to any one of claims 1-4, characterized in that, The configuration information includes a first period, and the timing that matches the first period is the first timing; the method further includes: The first duration of receiving data from the network device; At least one prediction error is determined within the first duration; Send a second indication message to the network device, the second indication message indicating the at least one prediction error.

6. The communication method according to any one of claims 1-5, characterized in that, The method further includes: Receive a second duration from the network device; During the second duration, the predicted values ​​of the network parameters obtained by the first model are continuously sent.

7. A communication method, characterized in that, include: Send configuration information, wherein the configuration information is used to indicate the first timing of sending the first indication information, the first indication information is used to determine the prediction performance when the first model predicts the network parameters, and the first model is a model for time-domain prediction; Receive the first instruction information.

8. The communication method according to claim 7, characterized in that, The configuration information includes a first threshold and a timer. When the difference between the predicted value obtained by the first model predicting the network parameters and the actual value of the corresponding network parameters is greater than the first threshold, the timer is activated, and the expiration time of the timer is the first timing. The first timing is the time when the timer expires based on the timing start condition; Alternatively, the configuration information may include a first period, wherein the timing indicated by the first period is the first timing. Alternatively, the configuration information indicates that a first indication information is sent when the predicted value obtained from the first model prediction network parameters is sent, and the timing of sending the predicted value is the first timing.

9. The communication method according to claim 7 or 8, characterized in that, The first indication information includes at least one index information and a first error corresponding to the index information. The first error is the difference between the predicted value obtained by the first model predicting the network parameters and the true value of the corresponding network parameters. Alternatively, the first indication information may include a second error, which is the result of filtering the at least one first error.

10. The communication method according to any one of claims 7-9, characterized in that, The configuration information includes a first period, the timing indicated by the first period being the first timing; or, the configuration information indicates that a first indication information is sent when the predicted value obtained from the first model prediction network parameters is sent, the timing of sending the predicted value being the first timing. The first indication information includes the first true value of the network parameters, wherein the predicted value obtained by the first model predicting the network parameters corresponding to the first true value is the predicted value at the second time. The correspondence between the true value and the predicted value at the second time is determined according to the first parameter and the second period. The second period is the period for sending the predicted value. The first parameter is used to indicate the prediction window size of the first model. The predicted value at the second time is the predicted value sent before the first time.

11. The communication method according to any one of claims 7-10, characterized in that, The configuration information includes a first period, and the timing that matches the first period is the first timing; the method further includes: A first duration is sent, the first duration being used to determine at least one prediction error; Receive a second indication message, which indicates the at least one prediction error.

12. The communication method according to any one of claims 7-11, characterized in that, The method further includes: Send a second duration, wherein the second duration is used to send the predicted values ​​obtained by the first model predicting the network parameters; Receive the predicted values ​​obtained by the first model from predicting the network parameters.

13. A communication method, characterized in that, include: Send a first predicted value, wherein the first predicted value is a predicted value obtained by the first model predicting the network parameters, and the first model is a model used for time-domain prediction; A first indication message is sent at a first opportune moment, wherein the first opportune moment is the opportune moment when the error corresponding to the predicted value within the prediction window of each first model is measured, and the prediction window is the prediction window for predicting the first predicted value; or, the first opportune moment is the opportune moment when the error corresponding to the first predicted value is measured; or, the first opportune moment is the opportune moment when the first predicted value is sent, and the first indication message is used to determine the prediction performance of the first model when predicting network parameters.

14. The communication method according to claim 13, characterized in that, The first timing is the timing when the error corresponding to the predicted value within the prediction window of the first model is measured, or the first timing is the timing when the error corresponding to the first predicted value is measured. The first indication information includes a first error corresponding to the predicted value within the prediction window; or, the first indication information includes at least one index information and a first error corresponding to the index information, wherein the first error is the difference between the predicted value obtained by the first model predicting the network parameters and the true value of the corresponding network parameters; or, the first indication information includes a second error, wherein the second error is the result after filtering the at least one first error.

15. The communication method according to claim 13, characterized in that, The first timing is the timing of sending the first predicted value; the method further includes: Receive third indication information, the third indication information being used to indicate that the first predicted value and the first indication information are sent at the first timing.

16. The communication method according to claim 15, characterized in that, The method further includes: Before sending the first indication information at a first opportune moment, a second predicted value is sent at a second opportune moment; wherein, the first indication information includes a first true value, and the true value corresponding to the second predicted value is the first true value.

17. The communication method according to any one of claims 13-16, characterized in that, The method further includes: Receive the second duration; During the second duration, the predicted values ​​of the network parameters obtained by the first model are continuously sent.

18. A communication method, characterized in that, include: Receive a first predicted value, wherein the first predicted value is a predicted value obtained by predicting network parameters by a first model, and the first model is a model used for time-domain prediction; The system receives first indication information, wherein the first indication information is sent at a first timing, the first timing being the timing when the error corresponding to the predicted value within the prediction window of each measured first model is obtained, the prediction window being the prediction window for predicting the first predicted value; or, the first timing is the timing when the error corresponding to the first predicted value is obtained; or, the first timing is the timing when the first predicted value is sent, the first indication information being used to determine the prediction performance of the first model when predicting network parameters.

19. The communication method according to claim 18, characterized in that, The first timing is the timing when the error corresponding to the predicted value within the prediction window of the first model is measured, or the first timing is the timing when the error corresponding to the first predicted value is measured. The first indication information includes a first error corresponding to the predicted value within the prediction window; or, the first indication information includes at least one index information and a first error corresponding to the index information, wherein the first error is the difference between the predicted value obtained by the first model predicting the network parameters and the true value of the corresponding network parameters; or, the first indication information includes a second error, wherein the second error is the result after filtering the at least one first error.

20. The communication method according to claim 18, characterized in that, The first timing is the timing of sending the first predicted value; the method further includes: Send a third indication message, which is used to indicate that the first predicted value and the first indication message are sent at the first timing.

21. The communication method according to claim 20, characterized in that, The method further includes: Before receiving the first indication information, a second predicted value is received; wherein the second predicted value is sent at a second time, the first indication information includes a first true value, and the true value corresponding to the second predicted value is the first true value.

22. The communication method according to any one of claims 18-21, characterized in that, The method further includes: Send a second duration, wherein the second duration is used to send the predicted values ​​obtained by the first model from predicting the network parameters; Receive the predicted values ​​obtained by the first model from predicting the network parameters.

23. A communication device, characterized in that, It includes a module that performs the method as described in any one of claims 1-6; or, it includes a module that performs the method as described in any one of claims 7-12; or, it includes a module that performs the method as described in any one of claims 13-17; or, it includes a module that performs the method as described in any one of claims 18-22.

24. A communication device, characterized in that, The communication device includes a processor and a transceiver, the processor and the transceiver being configured to support the communication device in performing the method as described in any one of claims 1-22.

25. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed, perform the method as described in any one of claims 1-22.

26. A communication system, characterized in that, The communication system includes: a terminal and a network device, wherein the terminal is configured to perform the method as described in any one of claims 1-6, and the network device is configured to perform the method as described in any one of claims 7-12; Alternatively, the terminal may be used to perform the method as described in any one of claims 13-17, and the network device may be used to perform the method as described in any one of claims 18-22.

Images