Communication method and communication apparatus

By acquiring and analyzing channel state information, the performance monitoring differences of AI models are identified, solving the problem of poor performance of AI models in wireless communication systems and achieving efficient, accurate performance monitoring and flexible adaptation.

WO2026232920A1PCT designated stage Publication Date: 2026-11-12HONOR DEVICE CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/CN2026/080426
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-09
Filing Date
2026-02-27
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

In wireless communication systems, AI models often exhibit poor inference performance in certain scenarios due to a mismatch between training data and real-world scenarios. Therefore, it is necessary to monitor their performance to improve accuracy.

Method used

By acquiring the first channel state information (CSI) and determining the target value based on the target parameter, the difference between the first CSI and the predicted second CSI is used to monitor the performance of the AI ​​model, including determining precoding information from the spatial, frequency, and time domains, simplifying the calculation process and improving accuracy.

Benefits of technology

It enables accurate monitoring of AI model performance, improves the accuracy and flexibility of monitoring, reduces the power consumption of terminal devices, and is suitable for devices with different capabilities.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2026080426_12112026_PF_FP_ABST
    Figure CN2026080426_12112026_PF_FP_ABST
Patent Text Reader

Abstract

Provided in the present application are a communication method and a communication apparatus. Exemplarily, the method is applicable to a CSI prediction scenario. The method comprises: acquiring first channel state information (CSI), and determining first information on the basis of the first CSI and second CSI. The first CSI is obtained on the basis of measurement, and is determined on the basis of a target value of a target parameter, wherein the target value is one set from among candidate values of the target parameter, and the candidate values are one or more sets from among predefined values of the target parameter; and the second CSI is obtained on the basis of prediction, and the first information represents a difference between the first CSI and the second CSI. The method enables, by means of the first information, the monitoring over the accuracy of a second CSI result. The method is applied to a scenario in which CSI is predicted on the basis of an AI model, and can be used for monitoring the performance of the AI model. Moreover, a target value is one set from among candidate values, by means of which the selection range of the target value may be defined, thereby improving the accuracy of first CSI and further enhancing the accuracy of monitoring over the performance of the AI model.
Need to check novelty before this filing date? Find Prior Art

Description

Communication methods and communication devices

[0001] This application claims priority to Chinese patent application filed on May 9, 2025, with application number 202510601249.6 and entitled "Communication Method and Communication Device", the entire contents of which are incorporated herein by reference. Technical Field

[0002] This application relates to the field of communications, specifically to a communication method and a communication device. Background Technology

[0003] In recent years, the rapid development of artificial intelligence (AI) technology has brought new opportunities to the communications field. In some application scenarios, traditional signal processing algorithms are gradually being replaced by AI models to improve system performance and reduce computational complexity. For example, in wireless communication systems, AI models can be used to predict channel state information (CSI).

[0004] However, AI technology is inherently data-driven, and the generalization ability of AI models highly depends on the training data used to train them. In actual deployment, due to factors such as the time-varying nature of wireless channels, environmental interference, and user mobility, AI models may face the problem of mismatch between training data and real-world scenarios, resulting in poor inference performance in some situations. Therefore, it is necessary to monitor the performance of AI models. Summary of the Invention

[0005] This application provides a communication method and a communication device that enables the monitoring of the performance of AI models and improves the accuracy of the monitoring.

[0006] Firstly, a communication method is provided. This method can be executed by a terminal device, or by a component (such as a circuit, chip, or chip system) configured in the terminal device, or by a logic module or software capable of implementing all or part of the functions of the terminal device. This application does not limit this approach. The following description uses a terminal device as an example.

[0007] The method includes: acquiring first channel state information (CSI); the first CSI is based on measurement and determined according to the target value of a target parameter, wherein the target value is a set of candidate values ​​of the target parameter and the candidate values ​​are one or more sets of predefined values ​​of the target parameter; determining first information based on the first CSI and the second CSI; the second CSI is based on prediction, and the first information characterizes the difference between the first CSI and the second CSI.

[0008] The communication method provided in the first aspect of this application acquires a first CSI and determines first information based on the first and second CSIs to determine the difference between the measured CSI and the predicted CSI, thereby determining the accuracy of the predicted CSI. This, in turn, determines the performance of the AI ​​model used to predict the CSI, enabling performance monitoring of the AI ​​model. This facilitates network devices' decisions on whether to continue using the AI ​​model to predict CSI or revert to the traditional mode for CSI measurement, i.e., switching the CSI acquisition mode. Furthermore, the target parameter value used for the first CSI is determined to be a set of candidate values. Candidate values ​​are values ​​obtained after applying certain constraints to predefined values, thus ensuring higher accuracy of the first CSI determined based on the target value. The first CSI serves as a label for comparison with the second CSI to determine the first information. Therefore, improving the accuracy of the first CSI enhances the accuracy of the first information, enabling it to more accurately reflect the difference or similarity between the predicted CSI and the actual or ideal result, thereby improving the accuracy of AI model performance monitoring.

[0009] In one possible implementation, the first CSI includes first precoding information, and the target parameters include at least one of the following:

[0010] Parameters related to the first basis vector; the first basis vector is a basis vector related to the spatial domain;

[0011] The parameters are related to the second basis vector; the second basis vector is a frequency domain related basis vector.

[0012] Parameters related to the third basis vector; the third basis vector is a time-domain dependent basis vector;

[0013] The parameters related to the weighting coefficients of at least one of the first, second, and third basis vectors.

[0014] All of the above parameters are related to precoding information. The target parameters include one or more of the above parameters, and the target value used to determine the first precoding information is a set of candidate values. That is, the value of one or more of the parameters related to the precoding information is a value among the candidate values ​​obtained by restricting the value of the predefined value. Therefore, this method can improve the accuracy of the determined first precoding information and improve the accuracy of performance monitoring of AI models that predict precoding information.

[0015] In one possible implementation, the first precoding information is determined based on a first codebook type, which is used to indicate the precoding information determined according to a first base vector and a second base vector.

[0016] In one possible implementation, the target parameter includes one or more codebook parameters corresponding to the first codebook type, and the predefined value of the target parameter includes one or more codebook parameter configurations corresponding to the first codebook type.

[0017] In this implementation, based on the first codebook type, the first precoding information is determined according to the first base vector and the second base vector. That is, the first precoding information is determined from two dimensions: spatial domain and frequency domain. This simplifies the determination process of the first precoding information, thereby simplifying the calculation process of the first information, simplifying the performance monitoring process of the AI ​​model, and reducing the power consumption of the terminal device.

[0018] In one possible implementation, the first precoding information is determined based on a second codebook type, which is used to indicate the determination of precoding information according to a first base vector, a second base vector, and a third base vector; the target parameter includes one or more codebook parameters corresponding to the second codebook type, and the predefined value of the target parameter includes one or more codebook parameter configurations corresponding to the second codebook type.

[0019] In this implementation, the first precoding information is determined based on the second codebook type, according to the first, second, and third basis vectors. This means the first precoding information is determined from three dimensions: spatial, frequency, and time domains, further improving the accuracy of the obtained first precoding information, thereby enhancing the accuracy of the first information and ultimately improving the accuracy of AI model performance monitoring. Moreover, by determining the first precoding information based on the second codebook type, the codebook type used by the terminal device to acquire the first CSI is the same as the codebook type used to predict the second CSI. Therefore, this method only requires the terminal device to have the ability to determine CSI based on this single codebook type to achieve model performance monitoring, making this method highly applicable.

[0020] The method provided in this application embodiment can determine the first precoding information based on different types of codebooks, making the method applicable to devices with different capabilities and improving the flexibility of use.

[0021] In one possible implementation, the target parameters include the parameter combination index, the number L of the first basis vectors, and the parameter p related to the number of the second basis vectors. υ And at least one of the parameters β associated with the total number of weighting coefficients, wherein each value of the parameter combination index corresponds to a value of L, a value of p υ The value of and a value of β.

[0022] Parameter combination index, L, p υParameters such as β are all related to the accuracy of the precoded information. In this implementation, the target parameter includes at least one of the above parameters. By restricting the values ​​of these parameters in the first precoded information through candidate values, the accuracy of the first precoded information can be directly and effectively improved, thereby improving the accuracy of AI model performance monitoring.

[0023] In one possible implementation, the target parameter includes a parameter combination index; the candidate values ​​for the parameter combination index are:

[0024] The predefined values ​​of the parameter combination index that are greater than the preset index threshold;

[0025] Alternatively, values ​​within the preset index range that are among the predefined values ​​of the parameter combination index;

[0026] Alternatively, the maximum value supported by the terminal device among the predefined values ​​of the parameter combination index.

[0027] In one possible implementation, the target value and / or candidate value of the target parameter are determined based on the first configuration information sent by the network device;

[0028] Alternatively, the target value of the target parameter and / or the candidate value of the target parameter can be a preset value.

[0029] Alternatively, the target value and / or candidate value of the target parameter may be determined based on a first value, which is the value of the relevant parameter used to determine the second CSI.

[0030] This implementation provides multiple methods for determining target and / or candidate values, facilitating flexible use and improving the accuracy of the first CSI in different ways, thereby enhancing the accuracy of AI model performance monitoring. Specifically, determining the target and / or candidate values ​​through first configuration information sent by the network device makes these values ​​controllable by the network device, which is more conducive to improving the accuracy and reliability of the first pre-encoded information. The target and / or candidate values ​​are preset values, making it convenient for terminal devices to obtain them and ensuring data stability, thus improving the stability of AI model performance monitoring. The target and / or candidate values ​​are determined based on the first value, resulting in higher consistency between the target parameter values ​​used to determine the first CSI and the relevant parameter values ​​used in the second CSI, improving the accuracy of the first information and thus enhancing the accuracy of AI model performance monitoring.

[0031] In one possible implementation, the candidate values ​​of the target parameter are those that satisfy the first condition from among the predefined values ​​of the target parameter.

[0032] In one possible implementation, the target value of the target parameter is the value among the candidate values ​​of the target parameter that satisfies the second condition.

[0033] In one possible implementation, if the target value of the target parameter is determined according to the first value, the second condition includes: the target value of the target parameter is greater than or equal to the first value used to determine the second precoding information.

[0034] In one possible implementation, the first precoding information is determined based on a first codebook type, the second precoding information is determined based on a second codebook type, and the second condition includes at least one of the following:

[0035] a. The target value of the number of first base vectors is greater than or equal to the value of the number of first base vectors used by the terminal device to determine the second precoding;

[0036] b. The target value of the number of second base vectors is greater than or equal to the value of the number of second base vectors used by the terminal device to determine the second precoding information;

[0037] c. The target value of the total number of weighting coefficients is greater than or equal to the value of the total number of weighting coefficients used by the terminal device to determine the second pre-compiled information;

[0038] d. The target value of the phase coefficient quantization precision is greater than or equal to the value of the phase coefficient quantization precision used by the terminal device to determine the second precoded information;

[0039] e. The target value of the amplitude coefficient quantization precision is greater than or equal to the value of the amplitude coefficient quantization precision used by the terminal device to determine the second precoded information.

[0040] In one possible implementation, the first precoding information is determined based on a second codebook type, and the second precoding information is determined based on a second codebook type, wherein the second condition includes at least one of the following:

[0041] a. The target value of the parameter combination index is greater than or equal to the value of the parameter combination index used by the terminal device to determine the second precoding;

[0042] b. The target value of the number of first base vectors is greater than or equal to the value of the number of first base vectors used by the terminal device to determine the second precoding;

[0043] c. The target value of the number of second base vectors is greater than or equal to the value of the number of second base vectors used by the terminal device to determine the second precoding information;

[0044] d. The target value of the number of third base vectors is greater than or equal to the value of the number of third base vectors used by the terminal device to determine the second precoding information;

[0045] e. The target value of the total number of weighting coefficients is greater than or equal to the value of the total number of weighting coefficients used by the terminal device to determine the second pre-compiled information;

[0046] f. The target value of the phase coefficient quantization precision is greater than or equal to the value of the phase coefficient quantization precision used by the terminal device to determine the second precoded information;

[0047] g. The target value of the amplitude coefficient quantization precision is greater than or equal to the value of the amplitude coefficient quantization precision used by the terminal device to determine the second precoded information.

[0048] In one possible implementation, the method further includes sending first information to the network device.

[0049] In one possible implementation, sending first information to a network device includes: sending a first report to the network device, wherein the first report includes the first information.

[0050] In one possible implementation, the first information is used to generate a first report; the value of the first parameter corresponding to the first report is greater than or equal to a first predefined value; and / or, the value of the second parameter corresponding to the first report is greater than or equal to a second predefined value; the value of the first parameter corresponding to the first report or the value of the second parameter corresponding to the first report is used to determine the time required for the terminal device to process the first report.

[0051] The first predefined value is the value of the first parameter Z′ corresponding to the predicted CSI-related report, as predefined by relevant protocols or regulations. Alternatively, the first predefined value is the value of the first parameter corresponding to a report of the same or similar type as the first report, as predefined by protocols or regulations. For example, the first predefined value is the value of the first parameter corresponding to CSI report 1, which is a CSI report generated according to the second codebook type. As another example, the first predefined value is the value of the first parameter corresponding to CSI report 2, which is a CSI report predicted using a non-AI model and generated according to the second codebook type.

[0052] The second predefined value is a value of the second parameter Z corresponding to a CSI-related report, predefined by relevant protocols or regulations. Alternatively, the second predefined value T2 is a value of the second parameter Z corresponding to a report of the same or similar type as the first report, predefined by protocols or regulations. For example, the second predefined value is the value of the second parameter Z corresponding to CSI report 1. As another example, the second predefined value is the value of the second parameter Z corresponding to CSI report 2. CSI report 1 and CSI report 2 are as described above.

[0053] The values ​​of the first and second parameters can be used to determine the resource occupancy time. In this embodiment, the process of the terminal device generating the first report includes both the process of acquiring the first CSI and the process of generating first information based on the first and second CSIs. The process is relatively complex, and therefore the resource occupancy time required to generate the first report may be relatively long. In this implementation, the value of the first parameter corresponding to the first report is greater than or equal to a first predefined value, and / or the value of the second parameter corresponding to the first report is greater than a second predefined value, so that the resource occupancy time determined by the terminal device is longer. This allows the terminal device sufficient time to determine the first information, generate the first report, and transmit the first report, thereby achieving a match between the processing time of the first report and the capabilities of the terminal device, and improving the effectiveness of the first report reporting.

[0054] In one possible implementation, the value of the first parameter corresponding to the first report is related to at least one of the following: a first predefined value, a first offset value, and a first multiple; wherein the first offset value is a value greater than or equal to 0, and the first multiple is a value greater than or equal to 1.

[0055] In this implementation, the value of the first parameter is determined based on at least one of the first predefined value, the first offset value, and the first multiple. This not only allows the terminal device sufficient time to process the first report, but also references the first predefined value of the protocol or regulations, thereby preventing the waste of time and resources, matching the resource occupation time of the first report with the actual needs, and improving the effectiveness of the first report.

[0056] In one possible implementation, the value of the first parameter corresponding to the first report is: the product of a first predefined value and a first multiple; or, the sum of the first predefined value and a first offset value; or, the sum of the product of the first predefined value and the first multiple and the first offset value.

[0057] In this implementation, the value of the first parameter corresponding to the first report is determined simply and effectively by calculating products, sums, etc. The first multiplier and the first offset value can be set according to the capabilities of the terminal device, allowing for flexible setting of resource usage time while matching the capabilities of the terminal device, thus improving the effectiveness of the first report submission.

[0058] In one possible implementation, the value of the second parameter corresponding to the first report is related to at least one of the following: a second predefined value, the number of measurement opportunities included in the first CSI, the interval of prediction opportunities included in the second CSI, a second offset value, and a second multiple; the measurement opportunities included in the first CSI refer to the measurement opportunities corresponding to the CSI measurement results included in the first CSI, the prediction opportunities included in the second CSI refer to the prediction opportunities corresponding to the CSI prediction results included in the second CSI, the second offset value is a value greater than or equal to 0, and the second multiple is a value greater than or equal to 1.

[0059] In this implementation, the value of the second parameter is determined based on at least one of the second predefined value, the number of measurement opportunities included in the first CSI, the interval of prediction opportunities included in the second CSI, the second offset value, and the second multiple. This not only allows the terminal device to have sufficient time to process the first report, but also references predefined values ​​such as protocols or regulations, and prevents the waste of time and resources. It also achieves a match between the resource occupation time of the first report and the actual needs, thereby improving the effectiveness of the first report.

[0060] In one possible implementation, the value of the second parameter corresponding to the first report is the sum of at least one of a first predefined value, a second offset value, and a second value, and the second predefined value; wherein the second value is determined based on the number of measurement opportunities included in the first CSI and the interval between the prediction opportunities included in the second CSI; or, the product of the second predefined value and a second multiple.

[0061] In this implementation, the value of the second parameter corresponding to the first report is determined simply and effectively by calculating the sum of at least one of the first predefined value, the second offset value, and the second value with the second predefined value. The second value is determined based on the number of measurement opportunities included in the first CSI and the interval between the prediction opportunities included in the second CSI, fully considering the time required to process the first and second CSI data. This ensures that the terminal device has sufficient time to determine and process the first report, thereby matching the processing time of the first report with the capabilities of the terminal device and improving the effectiveness of the first report submission.

[0062] In one possible implementation, the first information is used to generate a first report; the number of processing resources required to process the first report is: a predefined number of processing resources, or determined according to the number of measurement opportunities included in the first CSI; the measurement opportunities included in the first CSI refer to the prediction opportunities corresponding to the CSI measurement results included in the first CSI.

[0063] Optionally, the predefined number of processing resources can be a predefined number of resources required to process CSI-related reports, as specified in an agreement or regulation. Alternatively, the predefined number of processing resources can be a predefined number of resources required for reports of the same or similar type as the first report, as specified in an agreement or regulation. For example, the predefined number of processing resources could be a predefined number of resources required to process CSI Report 1 or CSI Report 2, as described above.

[0064] In this implementation, the number of processing resources required to process the first report is a predefined number of processing resources, or it is determined according to the number of measurement opportunities included in the first CSI. This ensures that the terminal device has sufficient processing resources to process the first report, is consistent with the protocol or regulations, prevents resource waste, matches the processing resources occupied by the first report with the capabilities of the terminal device, and improves the effectiveness of the first report.

[0065] In one possible implementation, at least one parameter in the parameters related to the first CSI and the parameters related to the second CSI has the same value. This ensures that the dimensions of the first and second CSIs are consistent, thereby making the calculated first information more accurate, and consequently, making the monitoring results of AI model performance more accurate.

[0066] In one possible implementation, at least one parameter includes at least one of the following: bandwidth, frequency domain transmission density, number of antenna ports, number of physical resources contained in a frequency domain subband, number of precoding subbands contained in a frequency domain subband, number of precoding subbands, number of layers of the precoding matrix, and number of measurement opportunities contained in the first CSI, wherein the measurement opportunities contained in the first CSI refer to the measurement opportunities corresponding to the CSI measurement results contained in the first CSI.

[0067] The above parameters are also called common parameters.

[0068] In this implementation, one or more of the aforementioned common parameters take the same value in determining the first CSI as they take the same value in determining the second CSI. This ensures that the dimensions of the first and second CSIs are consistent, making the calculated first information more accurate and thus making the monitoring results of the AI ​​model performance more accurate.

[0069] In one possible implementation, the first CSI includes CSI measurement results corresponding to X measurement opportunities, and the second CSI includes CSI prediction results corresponding to N4 prediction opportunities, where X is an integer greater than or equal to 1 and N4 is an integer greater than or equal to X; determining first information based on the first CSI and the second CSI includes: determining a third CSI based on the second CSI; the third CSI includes CSI prediction results corresponding to X prediction opportunities out of the N4 prediction opportunities; and determining the first information based on the first CSI and the third CSI.

[0070] X can be equal to 1, or greater than 1 and less than or equal to N4. That is, the third CSI can include the CSI prediction result corresponding to one prediction opportunity in the second CSI, or it can include CSI prediction results corresponding to multiple prediction opportunities. When X = 1, the third CSI includes the CSI prediction result corresponding to one of the N4 prediction opportunities in the second CSI. The terminal device only needs to determine the first information based on the CSI measurement result of one measurement opportunity and the CSI prediction result of one prediction opportunity, resulting in less computation and reduced power consumption of the terminal device.

[0071] When 1 < X ≤ N4, the first CSI includes CSI measurement results corresponding to X measurement times, and the third CSI includes CSI prediction results corresponding to X prediction times. In other words, the terminal device determines the first information based on the CSI measurement results from multiple measurement times and the CSI prediction results from multiple prediction times. Thus, the first information can characterize the differences between the CSI measurement results and the CSI prediction results at multiple times, reflecting the inference performance of the AI ​​model at different times, resulting in a more comprehensive and accurate monitoring of the AI ​​model's performance.

[0072] Optionally, the second CSI can be the CSI from the predicted CSI report. That is, the third CSI can be obtained directly from the predicted CSI report, which allows for the reuse of the predicted CSI report, simplifies the process of determining the first piece of information, and simplifies the model performance monitoring process.

[0073] In one possible implementation, when X is 1:

[0074] X prediction opportunities are the first prediction opportunities among N4 prediction opportunities;

[0075] Alternatively, the X prediction opportunities are the N4th prediction opportunity out of the N4 prediction opportunities;

[0076] Alternatively, the X prediction opportunities are the nth prediction opportunity among the N4 prediction opportunities, where n is an integer greater than 1 and less than N4;

[0077] Alternatively, the X prediction timings are the prediction timings closest to the reception timing of a first reference signal among the N4 prediction timings, and the first reference signal is used to acquire the first CSI.

[0078] This implementation provides multiple methods for determining the prediction timing. Generally, the AI ​​model's prediction accuracy is higher at earlier prediction timings than at later ones. In one method, the prediction timing is the first of N4 prediction timings; that is, the first piece of information is determined based on the CSI prediction result of the earliest prediction timing to determine the AI ​​model's performance. If the CSI prediction result of the first prediction timing has poor accuracy, it indicates that the AI ​​model's performance is poor. This method can accurately and easily monitor the performance of AI models.

[0079] In another approach, when the prediction timing is the N4th prediction timing out of N4 prediction timings, the first information is determined by the CSI prediction result of the last prediction timing to determine the performance of the AI ​​model. If the accuracy of the CSI prediction result of the last prediction timing is good, it indicates that the performance of the AI ​​model is good. This method can accurately and simply monitor the performance of the AI ​​model.

[0080] In another approach, if the first or N4th prediction timing out of the N4 prediction timings cannot match the measurement timings included in the first CSI, other prediction timings besides the first and N4th prediction timings (i.e., the nth prediction timing) can be selected to determine the first information, thereby improving the reliability of AI model performance monitoring.

[0081] In another approach, the prediction timing is the one closest to the reception timing of a first reference signal (i.e., the first RS) among N4 prediction timings. In this approach, the first information is calculated based on the CSI prediction result corresponding to the prediction timing closest to the reception timing of the first RS, so that the matching degree between the prediction timing and the measurement timing is the highest, thereby making the calculated first information more accurate, and thus making the performance monitoring results of the AI ​​model more accurate.

[0082] In one possible implementation, the first CSI includes first pre-encoded information, the third CSI includes second pre-encoded information, and the first information includes one or more squared generalized cosine similarity (SGCS) mean values; determining the first information based on the first CSI and the third CSI includes: determining multiple SGCS values ​​based on the first pre-encoded information and the second pre-encoded information; and determining one or more SGCS mean values ​​based on the multiple SGCS values.

[0083] In this implementation, one or more SGCS averages are calculated based on multiple SGCS values. These SGCS averages are used to generate the first report and then fed back to the network device. This eliminates the need to include every SGCS value in the first report, reducing the overhead of the terminal device feeding back the first report and also reducing the overhead of the network device receiving the first report, thus saving network resources.

[0084] In one possible implementation, the multiple SGCS values ​​are v×N3 SGCS values, where v represents the number of layers in the precoding matrix and N3 represents the number of precoding subbands; based on the multiple SGCS values, one or more SGCS averages are determined, including: calculating the average of the N3 SGCS values ​​corresponding to each layer in the v×N3 SGCS values ​​respectively, to obtain v SGCS averages.

[0085] In this implementation, the v SGCS mean values ​​correspond to the similarity between the second and first precoding matrices of the v layers, respectively. This approach helps network devices understand the accuracy of the CSI predictions for different layers by the AI ​​model.

[0086] In one possible implementation, the multiple SGCS values ​​are v×N3×X SGCS values, where X represents the number of measurement opportunities included in the first CSI; based on the multiple SGCS values, one or more SGCS averages are determined, including: calculating the average of the v×N3 SGCS values ​​corresponding to each prediction opportunity among the v×N3×X SGCS values ​​to obtain X SGCS averages.

[0087] In this implementation, the X SGCS mean values ​​correspond to the similarity between the second and first precoding matrices at X prediction times. This approach allows network devices to understand the accuracy of the CSI predictions made by the AI ​​model at different prediction times.

[0088] In one possible implementation, there are multiple SGCS values, namely v×N3×X SGCS values. Based on the multiple SGCS values, one or more SGCS averages are determined, including: calculating the average of the N3 SGCS values ​​corresponding to each layer at each prediction time among the v×N3×X SGCS values, to obtain the v×X SGCS averages.

[0089] In this implementation, the v×X SGCS mean values ​​correspond to the similarity between the second and first precoding matrices of v layers at X prediction times. This approach allows network devices to understand the accuracy of the AI ​​model's predictions of CSI for different layers at different prediction times.

[0090] Secondly, a communication method is provided, which can be executed by a network device, or by a component (such as a circuit, chip, or chip system) configured in the network device, or by a logic module or software capable of implementing all or part of the functions of the network device. This application does not limit this. The following description uses a network device (such as a satellite) as an example.

[0091] The method includes: receiving first information sent by a terminal device; the first information characterizes the difference between a first CSI and a second CSI, wherein the first CSI is obtained based on measurement and the second CSI is obtained based on prediction, and the first CSI is determined according to a target value of a target parameter, wherein the target value is a set of candidate values ​​of the target parameter and the candidate values ​​are a set or more sets determined from predefined values ​​of the target parameter.

[0092] In one possible implementation, before receiving the first information sent by the terminal device, the method further includes: sending first configuration information to the terminal device, the first configuration information including the target value of the target parameter and / or the candidate value of the target parameter.

[0093] In one possible implementation, the method further includes: sending a first reference signal to a terminal device, the first reference signal being used to measure a first CSI.

[0094] The second aspect is the implementation on the network device side, which corresponds to the first aspect. The explanations, supplements, and descriptions of the beneficial effects of the first aspect also apply to the second aspect, and will not be repeated here.

[0095] Thirdly, a communication device is provided, comprising a processing module and a transceiver module. The processing module is used to acquire first channel state information (CSI) and determine first information based on the first CSI and a second CSI; the first CSI is based on measurement and determined according to a target value of a target parameter, wherein the target value is a set of candidate values ​​for the target parameter, and the candidate values ​​are one or more sets of predefined values ​​for the target parameter; the second CSI is based on prediction, and the first information characterizes the difference between the first CSI and the second CSI.

[0096] Fourthly, a communication device is provided, comprising a transceiver module. The transceiver module is used to receive first information sent by a terminal device; the first information characterizes the difference between a first CSI and a second CSI, wherein the first CSI is based on measurement and the second CSI is based on prediction, and the first CSI is determined according to a target value of a target parameter, wherein the target value is a set of candidate values ​​for the target parameter, and the candidate values ​​are one or more sets determined from predefined values ​​of the target parameter.

[0097] The third and fourth aspects are the implementation on the device side corresponding to the first and second aspects. The explanations, supplements, and descriptions of the beneficial effects of the first and second aspects also apply to the third and fourth aspects, and will not be repeated here.

[0098] Fifthly, a communication device is provided, including a processor. The processor is coupled to a memory and can be used to execute instructions or data in the memory to implement the method in any possible implementation of the first aspect described above. Optionally, the communication device further includes a memory. Optionally, the communication device further includes a communication interface, and the processor is coupled to the communication interface.

[0099] In one implementation, the communication interface can be a transceiver, or an input / output interface.

[0100] In another implementation, the communication device is a chip configured in a terminal device. When the communication device is a chip configured in a terminal device, the communication interface can be an input / output interface.

[0101] In a sixth aspect, a communication device is provided, including a processor. The processor is coupled to a memory and can be used to execute instructions or data in the memory to implement the method in any possible implementation of the second aspect described above. Optionally, the communication device further includes a memory. Optionally, the communication device further includes a communication interface, and the processor is coupled to the communication interface.

[0102] In one implementation, the communication interface can be a transceiver, or an input / output interface.

[0103] In another implementation, the communication device is a chip configured in a satellite. When the communication device is a chip configured in a satellite, the communication interface can be an input / output interface.

[0104] In a seventh aspect, a processor is provided, comprising: an input circuit, an output circuit, and a processing circuit. The processing circuit is used to receive signals through the input circuit and to transmit signals through the output circuit, causing the processor to execute a method in any possible implementation of any aspect.

[0105] In specific implementation, the processor can be one or more chips, the input circuit can be input pins, the output circuit can be output pins, and the processing circuit can be transistors, gate circuits, flip-flops, and various logic circuits. The input signal received by the input circuit can be received and input by, for example, but not limited to, a receiver, and the signal output by the output circuit can be, for example, but not limited to, output to and transmitted by a transmitter. Furthermore, the input circuit and the output circuit can be the same circuit, which is used as both the input circuit and the output circuit at different times. This application does not limit the specific implementation of the processor and various circuits.

[0106] Eighthly, a communication device is provided, including a processor and a memory. The processor is used to read instructions stored in the memory, receive signals via a receiver, and transmit signals via a transmitter to execute the method in any possible implementation of any of the preceding aspects.

[0107] Optionally, there may be one or more processors and one or more memories.

[0108] Ninthly, a computer program product is provided, comprising: a computer program (also referred to as code or instructions) that, when executed, causes a computer to perform a method in any possible implementation of any of the above aspects.

[0109] In a tenth aspect, a computer-readable storage medium is provided that stores a computer program (also referred to as code or instructions) that, when executed on a computer, causes the computer to perform the methods in any possible implementation of any of the above aspects.

[0110] Eleventhly, embodiments of this application provide a chip system including one or more processors for calling and executing instructions stored in memory, causing the methods in any of the above aspects or possible implementations to be executed. The chip system may be composed of chips or may include chips and other discrete devices.

[0111] The chip system may include input circuits or interfaces for transmitting information or data, and output circuits or interfaces for receiving information or data.

[0112] In a twelfth aspect, a communication system is provided, including the aforementioned terminal device and network device. Optionally, the communication system may further include other devices that communicate with the terminal device and / or network device. Attached Figure Description

[0113] Figure 1 is a schematic diagram of a communication system provided in an embodiment of this application;

[0114] Figure 2 is a schematic diagram of an example of CSI prediction principle provided in an embodiment of this application;

[0115] Figure 3 is a schematic diagram illustrating the principle of determining CPU usage time according to an embodiment of this application.

[0116] Figure 4 is a flowchart illustrating an example of a communication method provided in an embodiment of this application;

[0117] Figure 5 is a schematic diagram of an example of timing provided in an embodiment of this application;

[0118] Figure 6 is a schematic diagram of a communication device provided in an embodiment of this application;

[0119] Figure 7 is a schematic diagram of another communication device provided in an embodiment of this application;

[0120] Figure 8 is a schematic diagram of another example of a communication device provided in the embodiments of this application. Detailed Implementation

[0121] The technical solutions in the embodiments of this application will now be described with reference to the accompanying drawings.

[0122] The technical solutions provided in this application can be applied to various communication systems, such as: Global System for Mobile Communications (GSM) systems, General Packet Radio Service (GPRS), Wireless Local Area Network (WLAN), Long Term Evolution (LTE) systems, LTE Frequency Division Duplex (FDD) systems, LTE Time Division Duplex (TDD) systems, sidelink communication systems, Universal Mobile Telecommunication System (UMTS), Worldwide Interoperability for Microwave Access (WiMAX) communication systems, non-terrestrial network (NTN) communication systems, 5th generation (5G) mobile communication systems, or new radio access technology (NR). Among these, 5G mobile communication systems can include non-standalone (NSA) and / or standalone (SA) networking. The technical solutions provided in this application can also be applied to future communication systems. This application does not limit the scope of these applications.

[0123] Figure 1 is a schematic diagram of a communication system 100 according to an embodiment of this application. The communication system 100 may include network devices, such as network device 110 shown in Figure 1. The communication system 100 may also include terminal devices, such as terminal device 120 shown in Figure 1. Network device 110 and terminal device 120 can communicate via a wireless link.

[0124] Figure 1 illustrates an exemplary network device 110 and a terminal device 120. Optionally, the communication system 100 may also include multiple network devices and / or multiple terminal devices.

[0125] The network equipment in this application can be network-side equipment such as access network equipment and core network equipment. Access network equipment is sometimes also called access node. Access network equipment has wireless transceiver capabilities and is used to communicate with terminals. Access network equipment includes, but is not limited to, base stations, evolved NodeBs (eNodeBs), transmission reception points (TRPs) in the above-mentioned communication systems, next-generation NodeBs (gNBs) in 5G mobile communication systems, access network equipment or modules of access network equipment in open RAN (ORAN) systems, satellites in NTN communication systems, base stations in future mobile communication systems, or access nodes in WiFi systems. Access network equipment can also be modules or units that can implement some of the functions of a base station. Access network equipment can be a macro base station (as shown in Figure 1, 110a), a micro base station or indoor station (as shown in Figure 1, 110b), a relay node or donor node, or a wireless controller in a cloud radio access network (CRAN) scenario. Optionally, access network equipment can also be a server, wearable device, or vehicle-mounted equipment, etc. For example, the access network equipment in vehicle-to-everything (V2X) technology can be a roadside unit (RSU). Multiple access network devices in a communication system can be base stations of the same type or different types. Base stations can communicate with terminals directly or via relay stations. Terminals can communicate with multiple base stations using different access technologies. The embodiments of this application do not limit the specific technology or device form used in the access network equipment. In this application, the access network equipment is referred to as a network device.

[0126] In this application, the means for implementing the functions of a network device can be a network device itself, or a means capable of supporting the network device in implementing those functions, such as a processor, circuit, chip, or chip system. This means can be installed in or connected to the network device. In the technical solutions provided in this application, the example of a network device being used to implement the functions of a network device is used to describe the technical solutions provided in this application.

[0127] The terminal device in this application can be a wireless terminal device capable of receiving network device scheduling and instruction information. The wireless terminal device can be a device providing voice and / or data connectivity to a user, a handheld device with wireless connectivity, or other processing devices connected to a wireless modem. For example, the terminal device can communicate with one or more core networks or the Internet via a radio access network (RAN). The terminal device can also be referred to as a terminal, user equipment (UE), mobile station, mobile terminal, etc. Terminal devices can be widely used in various scenarios, such as device-to-device (D2D), vehicle-to-everything (V2X) communication, machine-type communication (MTC), Internet of Things (IoT), ultra-reliable low-latency communication (URLLC), virtual reality, augmented reality, industrial control, autonomous driving, telemedicine, smart grids, smart furniture, smart offices, smart wearables, smart transportation, smart cities, or satellite communication, etc. The terminal can be a mobile phone, tablet computer, computer with wireless transceiver capabilities, wearable device, vehicle, aircraft (such as drone, helicopter, airplane), hot air balloon, ship, robot, robotic arm, or smart home device, etc. The embodiments of this application do not limit the form of the terminal device.

[0128] In this application, the apparatus for implementing the functions of the terminal device can be the terminal device itself, or any apparatus capable of supporting the terminal device in implementing those functions, such as a processor, circuit, chip, or chip system. This apparatus can be installed in or connected to the terminal device. In the technical solutions provided in this application, the example of a terminal device being used to implement the functions of the terminal device is used to describe the technical solutions provided in this application.

[0129] Access network devices and / or terminals can be fixed or mobile. They can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; on water; or in the air on aircraft, balloons, and satellites. This application does not limit the application scenarios of the access network devices and terminals. Access network devices and terminal devices can be deployed in the same or different scenarios; for example, both can be deployed on land; or the access network device can be deployed on land and the terminal device on water, etc., and so on.

[0130] In practical applications, multiple network devices can collaborate to assist terminals in achieving wireless access, with different network devices each implementing some of the functions of a base station. For example, network devices can be central units (CUs), distributed units (DUs), CU-control plane (CPs), CU-user plane (UPs), or radio units (RUs), etc. CUs and DUs can be set up separately or included in the same network element, such as a baseband unit (BBU). RUs can be included in radio frequency equipment or radio frequency units, such as remote radio units (RRUs), active antenna units (AAUs), or remote radio heads (RRHs).

[0131] In different systems, CU (or CU-CP and CU-UP), DU, or RU may have different names, but those skilled in the art will understand their meaning. For example, in an ORAN system, CU can also be called O-CU (Open CU), DU can also be called O-DU, CU-CP can also be called O-CU-CP, CU-UP can also be called O-CU-UP, and RU can also be called O-RU. Any of the units among CU (or CU-CP, CU-UP), DU, and RU in this application can be implemented through software modules, hardware modules, or a combination of software and hardware modules. CU (or CU-CP and CU-UP), DU, and RU can implement different protocol layer functions.

[0132] To facilitate understanding of the embodiments of this application, the terminology and related technologies involved in this application will be briefly explained first. Optionally, the explanation of some terms may also refer to the explanation in the 3rd Generation Partnership Project (3GPP) standard protocol.

[0133] 1. Channel State Information (CSI)

[0134] In communication systems, network devices can determine and schedule downlink data channel resources, modulation and coding schemes (MCS), and precoding configurations for terminal devices based on Channel Information System (CSI). CSI is a type of channel information that reflects channel characteristics and quality.

[0135] CSI can include at least one of the following: rank indicator (RI), channel quality indicator (CQI), precoding matrix indicator (PMI), CSI-RS resource indicator (CRI), layer indicator (LI), reference signal receiving power (RSRP), or signal to interference plus noise ratio (SINR). The signal to interference plus noise ratio can also be called the signal-to-noise ratio.

[0136] 2. CSI Measurement

[0137] CSI can be generated by the terminal device. In traditional technologies, a CSI measurement process can be executed between the terminal device and the network device to provide CSI feedback. CSI can then be contained in a CSI report.

[0138] The CSI measurement process may include:

[0139] 1) The network device sends channel measurement configuration information to the terminal device.

[0140] The channel measurement configuration information is used to instruct channel measurements to be performed, as well as the configuration parameters for performing channel measurements. The channel measurement configuration information may include, for example, resources for channel measurement (CMR), or resources for interference (IMR) corresponding to the CMR.

[0141] 2) The network device sends a reference signal, such as a channel state information reference signal (CSI-RS), to the terminal device. Correspondingly, the terminal device receives the CSI-RS from the network device.

[0142] In NR, network devices transmit CSI-RS signals for terminal devices to probe the downlink channel. Terminal devices receive CSI-RS signals on a pre-configured CMR to perform channel measurements. Additionally, terminal devices can also receive interference measurement signals on a pre-configured IMR for interference measurement.

[0143] 3) The terminal device measures the CSI according to the CSI-RS and reports the CSI to the network device based on the CSI. Correspondingly, the network device receives the CSI report from the terminal device.

[0144] The terminal device calculates the CSI based on the measurement results from the CMR and IMR. A CSI report can then be generated based on the CSI. The CMR (referred to as CSI-RS resource) used to transmit the CSI-RS and the IMR used to transmit the interference measurement signal can be periodic, semi-persistent, or aperiodic, depending on the configuration provided to the terminal device via higher-layer signaling. Correspondingly, the generated CSI report can also be periodic, semi-persistent, or aperiodic.

[0145] 4) Network devices send data based on CSI reports.

[0146] 3. CSI Prediction

[0147] With the development of AI technology, it can be used to predict Communication Service Index (CSI) in the communications field. For example, AI models can be used to predict future CSI based on CSI measurements taken at historical times. The principle of CSI prediction is explained below.

[0148] For example, Figure 2 is a schematic diagram of a CSI prediction principle provided in an embodiment of this application. As shown in Figure 2, in CSI prediction based on an AI model, different CSIs can be defined using an observation window and a prediction window. Within the observation window, the network device can send CSI-RS, and the terminal device measures the CSI-RS within the observation window to obtain one or more sets of CSI measurement results. The terminal device inputs the CSI measurement results into the AI ​​model after preprocessing to predict one or more sets of CSI prediction results within the prediction window following the observation window (i.e., within a future time period).

[0149] As shown in Figure 2, the observation window includes K (K is a positive integer) measurement opportunities, with a time interval of m between adjacent measurement opportunities. Each measurement opportunity corresponds to a CSI-RS, and each measurement opportunity corresponds to a set of CSI measurement results. The measurement opportunity, also known as the measurement moment, can be understood as the time-domain location indicated by the CSI-RS. In other words, each set of CSI measurement results characterizes the channel state corresponding to a measurement opportunity. The K measurement opportunities within the observation window correspond to K sets of CSI measurement results; or, the observation window includes K sets of CSI measurement results.

[0150] The terminal device inputs K sets of CSI measurement results into the AI ​​model, determines the CSI prediction result based on the AI ​​model's output, and predicts a specific time point between the observation window and the prediction window based on the CSI prediction result (shown as diagonal filling in Figure 2), outputting a report containing the CSI prediction result, hereinafter referred to as the predicted CSI report. In one embodiment, the output of the AI ​​model can be channel information (e.g., a channel matrix), and the CSI prediction result can include PMI. After the AI ​​model outputs the channel information, the PMI can be further determined based on the channel information.

[0151] Specifically, the prediction window includes N4 (N4 is a positive integer) prediction opportunities, with a time interval of d between adjacent prediction opportunities. Each prediction opportunity corresponds to a set of CSI prediction results. The prediction CSI report includes N4 sets of CSI prediction results. Here, a prediction opportunity is also called a prediction moment, a time to predict channel information, or a prediction instance, etc. That is, each set of CSI prediction results represents the channel state corresponding to a prediction opportunity. The N4 prediction opportunities within the prediction window correspond to N4 sets of CSI prediction results, or in other words, the prediction window includes N4 sets of CSI prediction results.

[0152] The units for m and d mentioned above are time-domain units. These time-domain units can be time slots, half-time slots, sub-time slots, or symbols. The values ​​of m and d can be equal or unequal.

[0153] The parameters K, m, N4, and d mentioned above can be configured by the network device. Specifically, the network device can send configuration information for the predicted CSI report to the terminal device, which may include configurations for parameters such as K, m, N4, and d. Based on the configuration information for the predicted CSI report, the terminal device uses an AI model to predict CSI and generates a predicted CSI report.

[0154] 4. Forecasting PMI

[0155] As mentioned above, CSI can include PMI; therefore, the predicted CSI report can include the predicted PMI, hereinafter referred to as the predicted PMI. When the predicted CSI report includes the predicted PMI, the configuration information of the predicted CSI report sent by the network device to the terminal device can also include the configuration of relevant parameters for the predicted PMI. The terminal device generates the predicted PMI based on the configuration of the relevant parameters for the predicted PMI and the predicted channel information.

[0156] In 3GPP Release 18, the predicted PMI can be generated based on the enhanced Type II codebook for predicted PMI and the corresponding codebook parameters. The enhanced Type II codebook for predicting PMI is also called the Type II Doppler codebook, etc.

[0157] In the enhanced Type II codebook used to predict PMI, the precoding matrix indicated by PMI can be derived from L+M. v +Q vectors are determined. L vectors are spatial domain-related vectors, referred to in this embodiment as the first basis vectors (also called spatial domain basis vectors), M v The L vectors are frequency-domain related vectors, referred to as the second basis vectors (also called frequency-domain basis vectors) in this embodiment, and the Q vectors are time-domain (corresponding to the Doppler domain) related basis vectors, referred to as the third basis vectors in this embodiment. The terminal device can be based on the L first basis vectors and M... v The predicted PMI is determined by using Q second basis vectors, Q third basis vectors, and weighting coefficients corresponding to the first, second, and third basis vectors.

[0158] In other words, the relevant parameters for predicting PMI include: L parameters related to the first basis vector, M... v The parameters related to the second basis vector, the parameters related to the Q third basis vectors, and the parameters related to the corresponding weighting coefficients, etc.

[0159] Specifically, regarding the L first basis vectors: the dimension of each first basis vector is determined by the number of antenna ports of the reference signal. The value of L can be configured by the network device using configuration information from the predicted CSI report.

[0160] Regarding M v The second basis vector: M varies depending on the number of layers in the precoding matrix. v The value of M may vary. In one possible implementation, M... v The value of can be obtained, for example, through the formula Confirmed. Here, v represents the number of layers in the precoding matrix, N3 represents the number of precoding subbands (i.e., quantity), and R represents the number of precoding subbands contained in a frequency domain subband; that is, a frequency domain subband contains R precoding subbands. R represents the number of frequency domain subbands, and can typically be 1 or 2. The dimension of each second basis vector can be determined by the total number of frequency domain subbands. Here, the parameter p... υ N3 and R can be configured by network devices using configuration information from predicted CSI reports.

[0161] Regarding weighting coefficients: Weighting coefficients can include amplitude coefficients and phase coefficients. To reduce the feedback overhead of the terminal, the total number of weighting coefficients (i.e., the total number of coefficients) is usually required to not exceed 2K0. Here, β can be configured by the network device using configuration information from the predicted CSI report.

[0162] The codebook parameters involved in determining the PMI forecast include, but are not limited to, L and ρ mentioned above. v The quantization accuracy of L, ρ, β, amplitude coefficients, and phase coefficients, etc. v The codebook parameters, such as β, can be configured through parameter combination indexes based on the mapping relationship shown in Table 1 below.

[0163] Table 1

[0164] In Table 1, parameter combination indices can correspond to higher-level parameters and are configured by the network device. Each parameter combination index value corresponds to a set of L, ρ... v The value of β is also important. Generally speaking, the larger the parameter combination index, the higher the accuracy of the predicted PMI determined based on the predicted channel information. Correspondingly, the overhead required for the terminal device to indicate the predicted PMI in the predicted CSI report is also greater. Furthermore, as the parameter combination index increases, the computational complexity required for the terminal device to obtain the predicted PMI also increases. Therefore, for parameter combination indices 8 and 9, the terminal device generally needs additional capability signaling to notify the network device whether it supports parameter combination indices 8 and 9.

[0165] After generating the predicted PMI based on the aforementioned parameters related to the predicted PMI, the terminal device includes the predicted PMI in the predicted CSI report and reports it to the network device. Based on the predicted PMI, the network device can determine the precoding matrix corresponding to each prediction timing, layer, and predicted coding sub-band. For example, the precoding matrix for the n4th prediction timing, the lth layer, and the n3rd precoding sub-band is represented as follows: Where n3∈{1,…,N3}, l∈{1,…,v}, n4∈{1,…,N4}. This is the precoding matrix. It is obtained based on the prediction results of the AI ​​model, and is referred to as the prediction precoding matrix in this application.

[0166] 5. First parameter and second parameter

[0167] To allow terminal devices sufficient time to generate and submit CSI reports, NR Release 15 defines two delay parameters, Z and Z′, for CSI calculation latency. In this embodiment, delay parameter Z′ is also referred to as the first parameter, or simply the first parameter Z′, and delay parameter Z is also referred to as the second parameter, or simply the second parameter Z.

[0168] The first parameter Z′ represents the minimum number of symbols between the last symbol of the CSI-RS resource currently used to determine the CSI report and the first symbol of the uplink data channel used to carry the CSI report. In other words, the time interval between the last symbol of the CSI-RS resource used to generate the CSI report and the first symbol of the uplink data channel used to carry the CSI report is greater than the first parameter Z′.

[0169] CSI-RS resources include CMR and / or IMR. The time Z′ preceding the first symbol of the uplink data channel used to carry the CSI report can also be called the CSI reference time. The last symbol of the CSI-RS resources used to generate the CSI report is no later than the CSI reference time. In other words, the first parameter Z′ determines the last CSI-RS resource used for the CSI report. After the CSI reference time, the terminal device no longer receives or measures the CSI-RS resources used for the current CSI report because the terminal does not have enough time to process the measured resources after the CSI reference time, nor enough time to generate the CSI report. The value of the first parameter Z′ may differ for different types of CSI reports.

[0170] The second parameter Z represents the minimum number of symbols between the last symbol of the physical downlink control channel (PDCCH) that triggers the CSI report and the last symbol of the uplink data channel used to carry the CSI report. The PDCCH carries physical layer downlink control information (DCI). Alternatively, the second parameter Z can be described as the minimum number of symbols between the end time of the DCI that triggers the CSI report and the end time of the CSI report. The value of the second parameter Z may differ for different types of CSI reports. The second parameter Z corresponding to a specific CSI report can be used to determine the timing of downlink control channel detection and decoding, channel estimation and CSI calculation, and CSI report transmission by the terminal equipment.

[0171] The values ​​of the first parameter Z′ and the second parameter Z can be predefined and are related to the CSI report type and subcarrier spacing.

[0172] It should be noted that the value of the second parameter Z can be different for different configuration parameters μ, and the value of the first parameter Z′ can also be different for different configuration parameters μ. Here, μ is the configuration parameter corresponding to the subcarrier spacing. Different configuration parameters correspond to different subcarrier spacings.

[0173] The current correspondence between the CSI calculation delay (predefined values ​​of Z and Z′) for the Type II codebook and the configuration parameter μ in the protocol is shown in Table 2 below.

[0174] Table 2

[0175] The values ​​of the first parameter Z′ and the second parameter Z corresponding to a certain CSI report can be used to determine the time required for the terminal to process the CSI report, and thus determine the length of time the terminal device occupies resources to process the CSI report (called the resource occupation time). The resource occupation time is, for example, the time occupied by the processing unit (CPU) used to process channel state information (CSI processing unit).

[0176] Specifically, for periodic CSI reports or semi-persistent CSI reports, the last symbol of the CSI-RS resource used by the CSI report can be determined based on the value of the first parameter Z′, thereby determining the duration (called the first duration) between the last symbol of the CSI-RS resource used by the CSI report and the first symbol of the uplink data channel used to carry the CSI report. The CPU occupancy time is greater than or equal to the first duration.

[0177] For non-periodic CSI reports, the CPU usage duration is greater than or equal to the value of the second parameter corresponding to the report.

[0178] The following explanation is based on the accompanying drawings.

[0179] For example, Figure 3 is a schematic diagram illustrating the principle of determining CPU usage time according to an embodiment of this application. As shown in Figure 3(a), for a certain periodic CSI report or semi-persistent CSI report, the first symbol of the uplink data channel used to carry the CSI report is time t3. Based on time t3 and the value of the first parameter Z′, the CSI reference time t2 can be determined. Specifically, the CSI reference time t2 is the time before time t3, and the time difference between time t3 and the CSI reference time t2 is the value of the first parameter Z′.

[0180] Based on the CSI reference time t2, the last CSI-RS resource used to generate the CSI report can be determined. The last symbol of this last CSI-RS resource is time t1, which is no later than the CSI reference time t2. The first symbol of this last CSI-RS resource is time t0. Furthermore, the last symbol of the uplink data channel used to carry the CSI report is time t4. Therefore, the CPU usage time is greater than or equal to t4-t0.

[0181] It is understandable that the CPU usage time for each cycle is the same for both periodic CSI reports and semi-continuous CSI reports, as illustrated in Figure 3(a) which shows two cycles.

[0182] As shown in Figure 3(b), for a certain non-periodic CSI report, the last character of the PDCCH that triggers the CSI report, i.e., the end time of the DCI, is t0. The last symbol of the uplink data channel carrying the CSI report is t4. Therefore, t4-t0 is greater than or equal to the value of the second parameter Z, and the CPU usage time is greater than or equal to t4-t0. Thus, the CPU usage time is greater than or equal to the value of the second parameter Z. In this figure, the definitions of t2 and t3 are the same as in Figure 3(a), and will not be repeated.

[0183] It should be understood that the technical terms used in this application are for illustrative purposes only and not as limiting. For example, as technology evolves, technical terms may also change, and other technical terms that have the same technical meaning should also apply to this application.

[0184] The technical problems faced by this application are explained below.

[0185] Predicting CSI using AI models can improve CSI feedback efficiency. However, due to the generalization limitations of AI models, their inference performance (or prediction performance) is unstable, potentially leading to inaccurate CSI predictions. Specifically, AI models perform well when the training data used for model training and the data used for model inference have similar distributions. Conversely, their inference performance is poor when the distributions of the training data and the data used for model inference differ significantly. In actual deployments, factors such as the time-varying nature of wireless channels, environmental interference, and user mobility mean that AI models may only maintain their inference performance in specific environments. Therefore, it is necessary to monitor the performance of AI models and provide feedback to network devices to assist them in deciding whether to continue using AI models for CSI prediction or revert to traditional CSI measurement methods.

[0186] In view of this, this application provides a communication method for determining first information based on a first CSI and a second CSI, wherein the first CSI is based on measurement and the second CSI is based on prediction, and the first information characterizes the difference (or similarity) between the first CSI and the second CSI. Through the first information, the magnitude of the difference between the first CSI and the second CSI can be quantified, thus enabling the monitoring of the inference performance of the AI ​​model.

[0187] The solution provided in this application will be described in detail below with reference to the corresponding flowcharts. It is understood that the illustrative flowcharts provided in this application primarily use different devices (e.g., terminal devices, network devices) as examples of the execution subjects of this interactive illustration to illustrate the method, but this application does not limit the execution subjects of the interactive illustrations. For example, the devices (e.g., terminal devices, network devices) in the illustrative flowcharts can also be chips, chip systems, or processors that support the implementation of this method on the device, or logic modules or software that can implement all or part of the functions of the device.

[0188] As a general statement, the message or signaling interactions involved in the interaction process of this application embodiment can be standard messages or signaling or newly introduced messages or signaling. This application embodiment does not make specific limitations on this.

[0189] Figure 4 is a schematic diagram of a communication method 400 according to an embodiment of this application. It can be understood that the terminal device in Figure 4 can be any of the terminal devices in Figure 1, or it can refer to a device within the terminal device (e.g., a processor, chip, or chip system). The network device can be any access network device in Figure 1, or it can refer to a device within the access network device (e.g., a processor, chip, or chip system). As shown in Figure 4, the method 400 includes the following steps:

[0190] S410, the terminal device acquires the first CSI, which is based on measurement.

[0191] Optionally, the first CSI may include a set of CSI measurement results corresponding to one measurement time point, or it may include multiple sets of CSI measurement results corresponding to multiple measurement time points. In this embodiment, the measurement time point corresponding to the CSI measurement results included in the first CSI is simply referred to as the measurement time point included in the first CSI.

[0192] Optionally, the first CSI can be measured by the terminal device based on CSI-RS, or it can be measured by the terminal device based on a measurement timing indicated by the network device, or it can be measured by the terminal device based on a first reference signal (RS). The first RS is an RS sent by the network device for AI model performance monitoring.

[0193] Specifically, in one embodiment, the terminal device can acquire one or more sets of CSI measurement results based on CSI-RS measurements, as the first CSI. In another embodiment, the network device can also indicate a measurement timing to the terminal device, and the terminal device measures one or more sets of CSI measurement results based on the measurement timing indicated by the network device, as the first CSI. In yet another embodiment, the network device can send a first RS for AI model performance monitoring to the terminal device, the first RS including signals for one or more measurement timings. The terminal device measures one or more sets of CSI measurement results based on the first RS, as the first CSI.

[0194] Optionally, the first CSI may include one or more of the following: RI, CQI, PMI, precoding matrix, CRI, LI, reference signal received power RSRP, or SINR.

[0195] S420, the terminal device determines first information based on the first CSI and the second CSI. The second CSI is obtained based on prediction, and the first information characterizes the difference between the first CSI and the second CSI.

[0196] Optionally, the second CSI may include a set of CSI prediction results corresponding to one prediction time, or it may include multiple sets of CSI prediction results corresponding to multiple measurement time points. In this embodiment, the measurement time points corresponding to the CSI measurement results included in the second CSI are simply referred to as the measurement time points included in the second CSI.

[0197] The second CSI may include one or more of the following: RI, CQI, PMI, precoding matrix, CRI, LI, reference signal received power RSRP, or SINR, based on prediction. It should be understood that the second CSI may contain the same type of information as the first CSI; the difference is that the first CSI is based on measurement, while the second CSI is based on prediction.

[0198] In one embodiment, the second CSI is predicted based on an AI model. The AI ​​model can be a neural network, a linear regression model, a decision tree model, a support vector machine (SVM), a Bayesian network, a Q-learning model, or other machine learning (ML) models.

[0199] Optionally, the second CSI can be obtained from the prediction CSI report generated by the terminal device before the prediction window. Specifically, as shown in Figure 2, the terminal device generates and transmits the prediction CSI report before the prediction window. Therefore, the terminal device can obtain the CSI prediction results for one or more prediction windows from the prediction CSI report as the second CSI. In this way, the prediction CSI report can be reused, simplifying the process of obtaining the second CSI and simplifying the model performance monitoring process.

[0200] The first information represents the difference between the first CSI and the second CSI; in other words, the first information represents the similarity between the first CSI and the second CSI. Optionally, the first information can be information that can characterize the difference or similarity between the first CSI and the second CSI, such as the difference, ratio, or similarity value. In one embodiment, the first information can be the normalized mean square error (NMSE) or the square of generalized cosine similarity (SGCS), etc.

[0201] It is understandable that the first CSI is based on measurement and has high accuracy. Therefore, the first CSI can be used as a label to compare with the second CSI, determining the accuracy of the predicted second CSI, and thus the accuracy or performance of the AI ​​model used to predict the second CSI. If, based on the first information, the difference between the second CSI and the first CSI is large (i.e., their similarity is low), it indicates that the predicted second CSI has low accuracy, and therefore the performance of the AI ​​model used to predict the second CSI is poor. Conversely, if, based on the first information, the difference between the second CSI and the first CSI is small (i.e., their similarity is high), it indicates that the predicted second CSI has high accuracy, and therefore the performance of the AI ​​model used to predict the second CSI is good.

[0202] In this embodiment, by acquiring the first CSI and determining the first information based on the first CSI and the second CSI, the difference between the measured CSI and the predicted CSI is determined, thereby determining the accuracy of the predicted CSI, and further determining the performance of the AI ​​model used to predict the CSI, realizing performance monitoring of the AI ​​model, and facilitating network devices to decide whether to continue using the AI ​​model to predict CSI or to fall back to the traditional mode for CSI measurement, i.e., switching the CSI acquisition mode.

[0203] In one embodiment, after step S420 above, the method may further include:

[0204] The terminal device sends the first information to the network device. Correspondingly, the network device receives the first information.

[0205] In one embodiment, the terminal device may also generate a report based on the first information to provide feedback on the monitoring results of AI prediction performance, and submit the report to the network device. In this embodiment, the report providing feedback on the monitoring results of AI prediction performance is referred to as the first report. Specifically, after step S420 above, the terminal device sending the first information to the network device may include:

[0206] The terminal device sends a first report to the network device, which includes first information. Correspondingly, the network device receives the first report.

[0207] Optionally, the first report can be a periodic CSI report, a semi-continuous CSI report, or an aperiodic CSI report.

[0208] In one embodiment, the first CSI is determined based on the target value of the target parameter, where the target value is a set of candidate values ​​for the target parameter, and the candidate values ​​are one or more sets of predefined values ​​for the target parameter.

[0209] In other words, the target parameters are the parameters used to determine the first CSI, or, to put it another way, the target parameters are the parameters related to the first CSI. It should be understood that the target parameters may include one parameter or multiple parameters.

[0210] The target parameter can take different values. Network devices, terminal devices, or protocols can predefine one or more sets of values ​​for the target parameter, which are referred to as the predefined values ​​of the target parameter in this application.

[0211] In this embodiment of the application, the value of the target parameter used to determine the first CSI is called the target value. The target value is a set of candidate values ​​for the target parameter, and the candidate values ​​are one or more sets of predefined values ​​for the target parameter. In other words, the candidate values ​​of the target parameter are values ​​obtained after applying certain value restrictions to the predefined values, and the target value is a set of the restricted values.

[0212] Optionally, restrictions on predefined values ​​(i.e., the selection of candidate values) can be set as needed. In a specific embodiment, candidate values ​​are those among the predefined values ​​that result in relatively high accuracy of the first CSI. When the target parameter is a candidate value, the determined first CSI value has higher accuracy compared to when it is not a candidate value. Higher accuracy can be understood as the first CSI being closer to the true or ideal result, or a greater probability that the first CSI is close to the true or ideal result. In other words, candidate values ​​can be values ​​among the predefined values ​​that satisfy a first condition, which is a condition related to the accuracy of the first CSI.

[0213] In this implementation, the target parameter value used for the first CSI is determined to be a set of candidate values. These candidate values ​​are obtained by applying certain constraints to predefined values. Therefore, the accuracy of the first CSI determined based on the target value is relatively high. The first CSI serves as a label, used to compare with the second CSI to determine the first information. Thus, improving the accuracy of the first CSI enhances the accuracy of the first information, enabling it to more accurately reflect the difference or similarity between the predicted CSI and the actual or ideal result, thereby improving the accuracy of AI model performance monitoring.

[0214] In one embodiment, the first CSI includes first precoding information. The first precoding information is measurement-based, precoding-related information. The first precoding information may include at least one of a measured first precoding matrix and a measured first PMI.

[0215] In one embodiment, the second CSI includes second precoding information. The second precoding information is prediction-based, precoding-related information. The second precoding information may include at least one of a predicted second precoding matrix and a predicted second PMI.

[0216] In one embodiment, the target parameters may include one or more of the codebook parameters corresponding to the first CSI. As one possible implementation, the codebook parameters corresponding to the first CSI may include at least one of the following parameters: parameters related to the first basis vector, parameters related to the second basis vector, parameters related to the third basis vector, and parameters related to the weighting coefficients corresponding to at least one of the first, second, and third basis vectors.

[0217] Parameters related to the first base vector include, but are not limited to: the number of first base vectors L, the number of antenna ports of the reference signal, etc.

[0218] Parameters related to the second basis vector include, but are not limited to, the number M of the second basis vectors. v The number of precoded subbands N3 and the number of precoded subbands R contained in a frequency domain subband, etc.

[0219] Parameters related to the third basis vector include, but are not limited to, the number of third basis vectors, Q.

[0220] The parameters related to the weighting coefficients corresponding to at least one of the first, second, and third basis vectors include, but are not limited to: β, L, M1, K0, and the quantization precision of the weighting coefficients. or Weighting coefficients can include amplitude coefficients and phase coefficients, and the quantization accuracy of weighting coefficients includes the quantization accuracy of amplitude coefficients and the quantization accuracy of phase coefficients, etc.

[0221] All of the above parameters are related to precoding information. The target parameters include one or more of the above parameters, and the target value used to determine the first precoding information is a set of candidate values. That is, the value of one or more of the parameters related to the precoding information is a value among the candidate values ​​obtained by restricting the value of the predefined value. Therefore, this method can improve the accuracy of the determined first precoding information and improve the accuracy of performance monitoring of AI models that predict precoding information.

[0222] In this embodiment of the application, the first precoding information can be generated based on different types of codebooks. Correspondingly, the target parameters can be different, and the candidate values ​​and predefined values ​​corresponding to the target parameters can be different. Two examples are given below:

[0223] Example 1: In one possible implementation, the first precoding information is determined based on a first codebook type, which indicates the precoding information determined according to a first basis vector and a second basis vector. The target parameters include one or more codebook parameters corresponding to the first codebook type, and the predefined values ​​of the target parameters include one or more codebook parameter configurations corresponding to the first codebook type.

[0224] The first codebook type can be, for example, Enhanced Type II. That is, the first precoding information can be generated based on the Enhanced Type II codebook. Under the Enhanced Type II codebook, the precoding matrix indicated by the PMI can be L+M. v The vectors are determined. Among them, the L vectors are also the L first basis vectors, and M... v The vector, i.e., M v There are L second basis vectors. The first and second basis vectors have corresponding weighting coefficients. Regarding the L first basis vectors, M... v The parameters, such as the second basis vector and weighting coefficients, are the same as or similar to the corresponding parameters in the enhanced Type II codebook used to predict PMI, as described in the above embodiments. The difference lies in:

[0225] The enhanced Type II codebook does not involve the third basis vector and the weighting coefficients corresponding to the third basis vector;

[0226] In the enhanced Type II codebook, the total number of weighting coefficients does not exceed 2K0, where,

[0227] Additionally, the L and ρ values ​​corresponding to Enhanced Type II v The mapping relationship between the β codebook parameter configuration and the β codebook parameter configuration can be shown in Table 3, which is different from Table 1 above.

[0228] Table 3

[0229] In Table 3, the parameter combination indexes can correspond to higher-level parameters and are configured by the network device. Each parameter combination index value corresponds to a set of L, ρ... v The value of β is also important. Generally speaking, the larger the parameter combination index value, the higher the accuracy of the PMI. Correspondingly, the overhead required for the terminal device to indicate the PMI in the CSI report also increases. In addition, as the parameter combination index increases, the computational complexity required for the terminal device to obtain the PMI also increases. Therefore, for parameter combination indices 7 and 8, the terminal device generally needs additional capability signaling to notify the network device whether it supports parameter combination indices 7 and 8.

[0230] When the first precoding information is determined based on Enhanced Type II, the target parameters may include one or more of the codebook parameters corresponding to Enhanced Type II. The predefined values ​​of the target parameters may include one or more of the codebook parameter configurations corresponding to Enhanced Type II. For example, the target parameters may include one or more parameters from Table 3 above, i.e., the target parameters may include the codebook parameter combination index, the number L of the first basis vectors, and the parameter ρ related to the number of the second basis vectors. v And one or more of the parameters β related to the total number of weighting coefficients. Predefined values ​​for the target parameters include the codebook parameter combination index, L, ρ, as shown in Table 3 above. v One or more in the β configuration.

[0231] The following sections explain the candidate and predefined values ​​of the target parameter in several cases:

[0232] 1) The target parameters include the parameter combination index.

[0233] In this case, the predefined values ​​of the parameter combination index can include one or more of the codebook parameter combination index configurations in Table 3 above. For example, the predefined values ​​of the parameter combination index include 1, 2, 3, 4, 5, 6, 7, and 8.

[0234] In this case, the candidate values ​​for the parameter combination index can be determined in one of the following ways:

[0235] a. The candidate values ​​for the parameter combination index are those that are greater than the preset index threshold among the predefined values ​​of the parameter combination index.

[0236] For example, if the preset index threshold can be 4, then the candidate values ​​for the parameter combination index can be values ​​greater than 4 in the parameter combination index configuration in Table 3 above. That is, the candidate values ​​for the parameter combination index include: 5, 6, 7, and 8.

[0237] Optionally, a preset index threshold can be set based on the capabilities supported by the terminal device to determine candidate values ​​for the parameter combination index. For example, if the terminal device does not support parameter combination indices 7 and 8, setting the preset index threshold to 4 will result in candidate values ​​for parameter combination indices including 5 and 6. If the terminal device supports parameter combination indices 7 and 8, setting the preset index threshold to 6 will result in candidate values ​​for parameter combination indices including 7 and 8.

[0238] b. The candidate values ​​of the parameter combination index are values ​​that belong to the preset index range among the predefined values ​​of the parameter combination index.

[0239] For example, if the preset index range is [5,8], then the candidate values ​​for the parameter combination index include: 5, 6, 7, 8.

[0240] Optionally, a preset index range can be set based on the capabilities supported by the terminal device to determine the candidate values ​​for the parameter combination index. For example, if the terminal device does not support parameter combination indices 7 and 8, setting the preset index range to [5, 6] will result in candidate values ​​for the parameter combination index including 5 and 6. If the terminal device supports parameter combination indices 7 and 8, setting the preset index range to [7, 8] will result in candidate values ​​for the parameter combination index including 7 and 8.

[0241] c. The candidate value of the parameter combination index is the maximum value supported by the terminal device among the predefined values ​​of the parameter combination index.

[0242] For example, if the terminal device does not support parameter combination indices 7 and 8, the candidate value for the parameter combination index is 6. If the terminal device supports parameter combination indices 7 and 8, the candidate value for the parameter combination index is 8.

[0243] In this implementation, the target parameter includes a parameter combination index. The larger the value of the parameter combination index, the higher the accuracy of PMI. Therefore, by limiting the value of the parameter combination index through candidate values, a larger value of the parameter combination index used to determine the first precoding information is made, thereby improving the accuracy of the determined first precoding information, and consequently improving the accuracy of the obtained first information, thus enhancing the accuracy of performance monitoring of the AI ​​model. Moreover, the parameter combination index corresponds to a combination of multiple codebook parameter values. By using the parameter combination index, multiple codebook parameter values ​​can be limited simply, quickly, and directly, improving device processing efficiency.

[0244] 2) The target parameters include the number L of the first basis vectors.

[0245] In this case, the predefined value of L can include one or more of the configurations of L in Table 3 above, such as 2, 4, and 6.

[0246] In this case, the candidate values ​​for the number L of the first basis vectors can be determined in one of the following ways:

[0247] a. Candidate values ​​of L are those among the predefined values ​​of L that are greater than the preset threshold L. thr The value of .

[0248] For example, a preset threshold L thr If the value can be 2, then the candidate values ​​for L can be values ​​greater than 4 in the configurations of L in Table 3 above, that is, the candidate values ​​for L include: 4 and 6.

[0249] Optionally, a preset threshold L can be set based on the capabilities supported by the terminal device. thr This allows us to determine candidate values ​​for L. For example, if the terminal device does not support parameter combination indices 7 and 8, a preset threshold L is set. thr If the value is 2, then the candidate values ​​for the parameter combination index include 4. If the terminal device supports parameter combination indices 7 and 8, a preset threshold L is set. thr If the value is 4, then the candidate values ​​for the parameter combination index include 8.

[0250] b. The candidate values ​​of L are those that fall within the preset range of L's predefined values. range The value of .

[0251] For example, the preset range L range If the range is [4,6], then the candidate values ​​for L include 4 and 6.

[0252] Optionally, a preset range L can be set based on the capabilities supported by the terminal device. range This allows us to determine candidate values ​​for the parameter combination index. For example, if the terminal device does not support parameter combination indices 7 and 8, a preset range L is set. range If the parameter combination index is [4, 4], then the candidate values ​​for the parameter combination index include 4. If the terminal device supports parameter combination indices 7 and 8, and the preset index range is set to [4, 5], then the candidate values ​​for the parameter combination index include 4 and 6.

[0253] The candidate values ​​for c and L are the maximum values ​​supported by the terminal device among the predefined values ​​of L.

[0254] For example, if the terminal device does not support parameter combination indices 7 and 8, the candidate value for the parameter combination index is 4. If the terminal device supports parameter combination indices 7 and 8, the candidate value for the parameter combination index is 6.

[0255] In this implementation, the target parameters include the number L of the first basis vectors. The larger the value of L, the larger the initial range for filtering the first precoding information from the codebook, the higher the probability of successful filtering, and the more accurate the obtained first precoding information. Therefore, by limiting the value of L by candidate values, a larger value of L is used to determine the first precoding information, thereby improving the accuracy of the determined first precoding information, which in turn improves the accuracy of the obtained first information and enhances the accuracy of performance monitoring of the AI ​​model.

[0256] 3) The target parameters include the parameter p, which is related to the number of second basis vectors. υ .

[0257] In this case, p υ The predefined values ​​can include p in Table 3 above. υ One or more of the configurations, such as 1 / 4, 1 / 2, 1 / 8.

[0258] In this case, p υ The method for determining the candidate values ​​of p is similar to that for determining the candidate values ​​of L, and will not be elaborated further. It can be understood that determining p... υ When selecting candidate values, different thresholds, ranges, and candidate values ​​can be set based on the set to which υ belongs.

[0259] In this implementation, the target parameters include a parameter p that is related to the number of second basis vectors. υ .according to p υ The larger the value of the second basis vector M, the more important it is to maintain the higher the value of M. v The larger the number of candidate values, the larger the initial range for filtering the first precoded information from the codebook, the higher the probability of successful filtering, and the more accurate the obtained first precoded information. Therefore, limiting p by candidate values... υ The value of p determines the p used for the first precoding information. υ A larger value improves the accuracy of the determined first pre-encoded information, which in turn improves the accuracy of the obtained first information, thereby improving the accuracy of performance monitoring of the AI ​​model.

[0260] 4) Target parameters include L and p υ and β.

[0261] In this case, the predefined values ​​of the target parameter include multiple sets, each set containing an L value and a p value. υ The target parameter has a value of and a value of β. Candidate values ​​for the target parameter can be one or more sets of predefined values; for example, candidate values ​​for the target parameter could include two sets: {L = 2, p...} υ =1 / 2, β=1 / 2}, {L=6, p υ =1 / 4, β=3 / 4}.

[0262] In this implementation, the target parameters L and p υ And β. Limiting L and p by candidate values. υ The combination of values ​​for β and β can improve the accuracy of the determined first pre-encoded information, thereby improving the accuracy of the obtained first information and thus improving the accuracy of performance monitoring of the AI ​​model.

[0263] In Example 1, based on the first codebook type, the first precoding information is determined according to the first basis vector and the second basis vector. That is, the first precoding information is determined from two dimensions: spatial domain and frequency domain. This simplifies the determination process of the first precoding information, thereby simplifying the calculation process of the first information, simplifying the performance monitoring process of the AI ​​model, and reducing the power consumption of the terminal device.

[0264] Example 2: In another possible implementation, the first precoding information is determined based on a second codebook type, which indicates the determination of precoding information according to a first base vector, a second base vector, and a third base vector. The target parameters include one or more codebook parameters corresponding to the second codebook type, and the predefined values ​​of the target parameters include one or more codebook parameter configurations corresponding to the second codebook type.

[0265] The second codebook type could be, for example, the enhanced Type II codebook used for predicting PMI. That is, the first precoding information can be generated based on the enhanced Type II codebook. Under the enhanced Type II codebook, the precoding matrix indicating the PMI can be generated by L+M. v +Q vectors are determined. Among them, L vectors are also the L first basis vectors, M... v The vectors are frequency domain dependent vectors, that is, M v There are L first basis vectors and Q third basis vectors. The first, second, and third basis vectors have corresponding weighting coefficients. Regarding the L first basis vectors and M... v The parameters, such as the Q second basis vectors, Q third basis vectors, and weighting coefficients, can be found in the description of the enhanced Type II codebook used to predict PMI in the above embodiments, and will not be repeated here.

[0266] When the first precoding information is based on Enhanced Type II for predicting PMI, the target parameters may include one or more of the codebook parameters corresponding to Enhanced Type II for predicting PMI. Predefined values ​​for the target parameters may include one or more of the codebook parameter configurations corresponding to Enhanced Type II for predicting PMI. For example, the target parameters may include one or more parameters from Table 1 above, i.e., the target parameters may include the codebook parameter combination index, the number L of the first basis vectors, and the parameter ρ related to the number of the second basis vectors.v And one or more of the parameters β associated with the total number of weighting coefficients. Predefined values ​​for the target parameters include the codebook parameter combination index, L, ρ shown in Table 1 above. v One or more in the β configuration.

[0267] It should be understood that the target parameter in Example 2 can be the same as that in Example 1, but the predefined values ​​of the target parameter in Example 2 and Example 1 can be different.

[0268] In Example 2, the method for determining the candidate values ​​of the target parameter can be similar to that in Example 1, and the corresponding beneficial effects are also similar. However, it should be understood that since Example 2 and Example 1 use different codebook types, have different codebook parameter configurations, and different predefined values ​​of the target parameter, the preset threshold, preset range, etc., used in Example 2 to determine the candidate values ​​can be the same as or different from those in Example 1, and the determined candidate values ​​may be different.

[0269] Here, taking the target parameters including the parameter combination index as an example, the process of determining candidate values ​​is explained when the first precoding information is determined based on the second codebook type. Regarding the target parameters including L and p... υ Or L, p υ For the case of combination with β, please refer to Example 1 above, which will not be repeated here.

[0270] When the target parameter includes a parameter combination index, the predefined values ​​of the parameter combination index can include one or more of the codebook parameter combination index configurations in Table 1 above. For example, the predefined values ​​of the parameter combination index include 1, 2, 3, 4, 5, 6, 7, 8, and 9.

[0271] In this case, the candidate values ​​for the parameter combination index can be determined in one of the following ways:

[0272] a. The candidate values ​​for the parameter combination index are those that are greater than the preset index threshold among the predefined values ​​of the parameter combination index.

[0273] For example, if the preset index threshold can be 5, then the candidate values ​​for the parameter combination index can be values ​​greater than 5 in the parameter combination index configuration in Table 3 above. That is, the candidate values ​​for the parameter combination index include: 6, 7, 8, and 9.

[0274] Optionally, a preset index threshold can be set based on the capabilities supported by the terminal device to determine candidate values ​​for the parameter combination index. For example, if the terminal device does not support parameter combination indices 8 and 9, setting the preset index threshold to 5 will result in candidate values ​​for parameter combination indices including 7 and 8. If the terminal device supports parameter combination indices 8 and 9, setting the preset index threshold to 7 will result in candidate values ​​for parameter combination indices including 8 and 9.

[0275] b. The candidate values ​​of the parameter combination index are values ​​that belong to the preset index range among the predefined values ​​of the parameter combination index.

[0276] For example, if the preset index range is [6,9], then the candidate values ​​for the parameter combination index include: 6, 7, 8, and 9.

[0277] Optionally, a preset index range can be set based on the capabilities supported by the terminal device to determine the candidate values ​​for the parameter combination index. For example, if the terminal device does not support parameter combination indices 8 and 9, setting the preset index range to [6, 7] will result in candidate values ​​for the parameter combination index including 6 and 7. If the terminal device supports parameter combination indices 8 and 9, setting the preset index range to [8, 9] will result in candidate values ​​for the parameter combination index including 8 and 9.

[0278] c. The candidate value of the parameter combination index is the maximum value supported by the terminal device among the predefined values ​​of the parameter combination index.

[0279] For example, if the terminal device does not support parameter combination indices 8 and 9, the candidate value for the parameter combination index is 7. If the terminal device supports parameter combination indices 8 and 9, the candidate value for the parameter combination index is 9.

[0280] In this implementation, the target parameter includes a parameter combination index. The larger the value of the parameter combination index, the higher the accuracy of the PMI. Therefore, by limiting the value of the parameter combination index through candidate values, a larger value of the parameter combination index used to determine the first precoding information is made, thereby improving the accuracy of the determined first precoding information and thus improving the accuracy of the obtained first information, thereby improving the accuracy of performance monitoring of the AI ​​model. Moreover, the parameter combination index corresponds to a combination of multiple codebook parameter values. By using the parameter combination index, multiple codebook parameter values ​​can be limited simply, quickly, and directly, improving the processing efficiency of the terminal device.

[0281] In Example 2, based on the second codebook type, the first precoding information is determined according to the first, second, and third basis vectors. This means the first precoding information is determined from three dimensions: spatial, frequency, and time domains, further improving the accuracy of the obtained first precoding information, thereby enhancing the accuracy of the first information and ultimately improving the accuracy of AI model performance monitoring. Furthermore, in Example 2, the codebook type used by the terminal device to obtain the first CSI is the same as the codebook type used to predict the second CSI. Therefore, this method only requires the terminal device to have the ability to determine CSI based on this single codebook type to achieve model performance monitoring, making the method highly applicable.

[0282] The determination methods for the target and candidate values ​​of the target parameters are further explained below. In the embodiments of this application, the target and candidate values ​​of the target parameters can be determined in at least three ways:

[0283] Method 1: In one embodiment, the target value and / or candidate value of the target parameter are determined based on the first configuration information sent by the network device.

[0284] Specifically, the network device can send configuration information for monitoring the performance of the AI ​​model to the terminal device, or in other words, configuration information for sending a first report. In this embodiment, this configuration information is referred to as first configuration information. The first configuration information may include target values ​​of the target parameters, or candidate values ​​of the target parameters, or both target and candidate values ​​of the target parameters.

[0285] Taking the first configuration information as an example, which includes the target value of the target parameter and the target parameter includes the parameter combination index, the network device can determine the codebook type and the supported parameter combination index supported by the terminal device based on the capability signaling sent by the terminal device, and then determine the candidate value of the target parameter, and determine the target value from the candidate value.

[0286] 1) When the terminal device determines the first precoding information based on the first codebook type:

[0287] If the terminal device's capabilities do not support parameter combination indices 7 and 8 as shown in Table 3 above, the network device can determine the candidate values ​​for the parameter combination index as 5 and 6. These candidate values ​​5 and 6 can be pre-defined by protocols or specifications. Based on the candidate values ​​of the parameter combination index, combined with the current resource situation and the terminal device's status, the network device determines the target value for the parameter combination index as 6. The network device then sends the target value 6 to the terminal device through the first configuration information.

[0288] If the terminal device's capabilities support parameter combination indices 7 and 8 as shown in Table 3 above, the network device can determine the candidate values ​​for the parameter combination index as 7 and 8. These candidate values ​​7 and 8 can be pre-defined by protocols or specifications. Based on the candidate values ​​of the parameter combination index, combined with the current resource situation and the terminal device's status, the network device determines the target value for the parameter combination index as 8. The network device then sends the target value 8 to the terminal device through the first configuration information.

[0289] 2) When the terminal device determines the first precoding information based on the second codebook type:

[0290] If the terminal device's capabilities do not support parameter combination indices 8 and 9 as shown in Table 1 above, the network device can determine candidate values ​​for the parameter combination index as 6 and 7. Candidate values ​​6 and 7 can be pre-defined by protocols or specifications. Based on the candidate values ​​of the parameter combination index, combined with current resource conditions, the terminal device's status, etc., the network device determines the target value of the parameter combination index as 7, and then sends the target value 7 to the terminal device through the first configuration information.

[0291] If the terminal device's capabilities support parameter combination indices 8 and 9 as shown in Table 1 above, the network device can determine the candidate values ​​for the parameter combination index as 8 and 9. These candidate values ​​8 and 9 can be pre-defined by protocols or specifications. Based on the candidate values ​​of the parameter combination index, combined with the current resource situation and the terminal device's status, the network device determines the target value for the parameter combination index as 9. The network device then sends the target value 9 to the terminal device through the first configuration information.

[0292] Of course, network devices can also configure candidate values ​​of target parameters to terminal devices through the first configuration information. The terminal devices then determine the target value from the candidate values ​​based on the first configuration information. The specifics will not be elaborated further.

[0293] Method 2: In another embodiment, the target value of the target parameter and / or the candidate value of the target parameter is a preset value.

[0294] Specifically, the terminal device may pre-set the target value of the target parameter to a preset value based on its capabilities or according to protocols, regulations, etc., and / or pre-set the candidate value of the target parameter to a preset value.

[0295] Continuing with the example of target parameters including parameter combination indexes:

[0296] 1) When the first codebook type of the terminal device determines the first precoding information:

[0297] If the terminal device does not support the parameter combination indices 7 and 8 shown in Table 3 above, the terminal device can preset the candidate values ​​of the parameter combination index to 5 and 6, or directly preset the target value of the parameter combination index to 6.

[0298] If the terminal device supports parameter combination indices 7 and 8 as shown in Table 3 above, then the candidate values ​​for the preset parameter combination index are 7 and 8, or the target value for the preset parameter combination index is 8.

[0299] 2) When the terminal device determines the first precoding information based on the second codebook type:

[0300] If the terminal device does not support the parameter combination indices 8 and 9 shown in Table 1 above, the terminal device can preset the candidate values ​​of the parameter combination index to 6 and 7, or directly preset the target value of the parameter combination index to 7.

[0301] If the terminal device supports parameter combination indices 8 and 9 as shown in Table 1 above, then the candidate values ​​for the preset parameter combination index are 8 and 9, or the target value for the preset parameter combination index is 9.

[0302] Method 3: In yet another embodiment, the target value and / or candidate value of the target parameter are determined based on a first value, which is the value of the relevant parameter used to determine the second CSI.

[0303] Optionally, the first value may be the value of the parameter of the same type as the target parameter among the parameters used to determine the second CSI. For example, if the target parameter includes a parameter combination index, the first value may be the value of the parameter combination index used to determine the second CSI. That is, the target value and / or candidate value of the parameter combination index used by the first CSI are determined based on the value of the parameter combination index used to determine the second CSI.

[0304] In one specific embodiment, a mapping relationship can be pre-established between multiple available values ​​of the relevant parameters of the second CSI and the target value and / or candidate value of the target parameter of the first CSI. Then, the mapping relationship is queried based on the first value to determine the target value and / or candidate value of the target parameter.

[0305] The following explanation uses the example of a first CSI including first precoding information, a second CSI including second precoding information, a target parameter including a parameter combination index, and determining the target value of the target parameter.

[0306] 1) When the terminal device determines the first precoding information based on the first codebook type, the terminal device can pre-establish the mapping relationship of the target values ​​of the parameter combination index, as shown in Table 4 below:

[0307] Table 4

[0308] In other words, if the value of the parameter combination index used by the second precoding information is determined to be one of 1 to 7 in Table 1 above, then the target value of the parameter combination index used by the first precoding information is determined to be 6 in Table 3 above; if the value of the parameter combination index used by the second precoding information is determined to be 8 or 9 in Table 1 above, then the target value of the parameter combination index used by the first precoding information is determined to be 8 in Table 3.

[0309] 2) When the terminal device determines the first precoding information based on the first codebook type, the terminal device can pre-establish the mapping relationship of the target values ​​of the target parameters, as shown in Table 5 below:

[0310] Table 5

[0311] In other words, if the value of the parameter combination index used by the second precoding information is determined to be one of 1 to 7 in Table 1 above, then the target value of the parameter combination index used by the first precoding information is determined to be 7 in Table 1 above; if the value of the parameter combination index used by the second precoding information is determined to be 8 or 9 in Table 1 above, then the target value of the parameter combination index used by the first precoding information is determined to be 9 in Table 1.

[0312] The above example illustrates the process of determining the target value and / or candidate values ​​based on the second value, using a target parameter that includes a parameter combination index and a mapping method as an example. It should be understood that the target parameter may include other parameters, such as L and p from the codebook parameters. υ In cases where one or more of the following are considered, such as β, quantization precision of the weighting coefficients, the first value can be a value related to the codebook parameters used to determine the second CSI. Additionally, besides the mapping method, in some other embodiments, the target value and / or candidate values ​​can also be determined in other ways.

[0313] First, let's explain the process of determining the target value. Optionally, the target value of the target parameter can be determined according to a preset second condition, so that the determined target value and the first value satisfy the second condition, thereby making the accuracy of the first pre-encoded information higher, and thus improving the accuracy of AI model performance monitoring.

[0314] Optionally, the second condition may be, for example, that the target value of the target parameter is greater than or equal to the first value used to determine the second precoding information. It should be understood that different target parameters correspond to different first values, and thus different second conditions. The second condition will be explained below in two cases, depending on the codebook type used to determine the first and second precoding information.

[0315] Scenario A: When the terminal device determines the first precoding information based on the first codebook type and the terminal device determines the second precoding information based on the second codebook type, the second condition may include at least one of the following:

[0316] a. The target value of the number L of the first base vectors is greater than or equal to the value of the number L of the first base vectors used by the terminal device to determine the second precoding.

[0317] Specifically, when the target parameter includes the number L of the first basis vectors, the first value includes the value of the number L of the first basis vectors used to determine the second precoding information. Simply put, this condition means that the number L of the first basis vectors used by the terminal device to determine the first precoding information is greater than or equal to the number L of the first basis vectors used by the terminal device to determine the second precoding.

[0318] b. The number M of the second basis vectors v The target value is greater than or equal to the number M of second basis vectors used by the terminal device to determine the second precoding information. v The value of .

[0319] Specifically, the target parameters include the number M of the second basis vectors. v In this case, the first value includes the number M of the second basis vectors used to determine the second precoding information. v The value of M. Simply put, this condition is: the number M of the second basis vectors used by the terminal device to determine the first precoded information. v The number M of second basis vectors used by the terminal device to determine the second precoding information is greater than or equal to the number of second basis vectors. v .

[0320] c. The target value of the total number of weighting coefficients is greater than or equal to the value of the total number of weighting coefficients used by the terminal device to determine the second pre-compiled information.

[0321] Specifically, when the target parameter includes the total number of weighting coefficients, the first value includes the total number of weighting coefficients used to determine the second precoding information. Simply put, this condition means that the total number of weighting coefficients used by the terminal device to determine the first precoding information is greater than or equal to the total number of weighting coefficients used by the terminal device to determine the second precoding information.

[0322] d. The target value of the phase coefficient quantization precision is greater than or equal to the value of the phase coefficient quantization precision used by the terminal device to determine the second precoded information.

[0323] Specifically, when the target parameter includes the phase coefficient quantization precision, the first value includes the value used to determine the phase coefficient quantization precision of the second precoding information. Simply put, this condition means that the phase coefficient quantization precision used by the terminal device to determine the first encoding information is greater than or equal to the phase coefficient quantization precision used by the terminal device to determine the second precoding information.

[0324] e. The target value of the amplitude coefficient quantization precision is greater than or equal to the value of the amplitude coefficient quantization precision used by the terminal device to determine the second precoded information.

[0325] Specifically, when the target parameter includes the amplitude coefficient quantization precision, the first value includes the value of the amplitude coefficient quantization precision used to determine the second precoding information. Simply put, this condition means that the value of the amplitude coefficient quantization precision used by the terminal device to determine the first precoding information is greater than or equal to the value of the amplitude coefficient quantization precision used by the terminal device to determine the second precoding information.

[0326] Case B: When the terminal device determines the first precoding information based on the second codebook type, and the terminal device determines the second precoding information based on the second codebook type, the second condition may include at least one of the following:

[0327] a. The target value of the parameter combination index is greater than or equal to the value of the parameter combination index used by the terminal device to determine the second precoding.

[0328] Specifically, when the target parameter includes a parameter combination index, the first value includes the value of the parameter combination index used to determine the second precoding information. Simply put, this condition means that the parameter combination index used by the terminal device to determine the first precoding information is greater than or equal to the parameter combination index used by the terminal device to determine the second precoding information.

[0329] It should be understood that in case B, the terminal device determines that both the first precoding information and the second precoding information are determined based on the second codebook type. Therefore, the corresponding codebook parameters and their configurations are the same, and the parameter combination index configurations are the same. Thus, the second condition may include the relevant limitations of the parameter combination index.

[0330] b. The target value of the number L of the first basis vectors is greater than or equal to the value of the number L of the first basis vectors used by the terminal device to determine the second precoding.

[0331] c. The number M of the second basis vectors v The target value is greater than or equal to the number M of second basis vectors used by the terminal device to determine the second precoding information. v The value of .

[0332] d. The target value of the number of second base vectors Q is greater than or equal to the value of the number of third base vectors Q used by the terminal device to determine the second precoding information.

[0333] e. The target value of the total number of weighting coefficients is greater than or equal to the value of the total number of weighting coefficients used by the terminal device to determine the second pre-compiled information.

[0334] f. The target value of the phase coefficient quantization precision is greater than or equal to the value of the phase coefficient quantization precision used by the terminal device to determine the second precoded information.

[0335] g. The target value of the amplitude coefficient quantization precision is greater than or equal to the value of the amplitude coefficient quantization precision used by the terminal device to determine the second precoded information.

[0336] The above b, c, e, f, and g are the same as a to e in case A, and will not be repeated.

[0337] Optionally, the aforementioned first value can be indicated by the network device, namely, the value of the parameter combination index used to determine the second precoding information, the value of the number L of the first basis vectors, and the number M of the second basis vectors. v One or more of the following values, such as the value of the total number of weighting coefficients, the value of the quantization precision of the phase coefficients, and the value of the quantization precision of the amplitude coefficients, can be indicated by the network device.

[0338] Next, the determination of candidate values ​​will be explained. As mentioned above, optionally, candidate values ​​for the target parameter can be determined based on a preset first condition. In this embodiment, the first condition can be a condition related to the first value. Candidate values ​​satisfying the first condition can improve the accuracy of the first pre-encoded information, thereby enhancing the accuracy of AI model performance monitoring.

[0339] Optionally, the first condition may be, for example, that the candidate value of the target parameter is greater than or equal to the first value of the corresponding parameter used to determine the second precoding information. It should be understood that different target parameters correspond to different first values, and thus different first conditions. The specific content of the first condition can be similar to the second condition described above, except that the second condition characterizes the relationship between the target value and the first value, while the first condition characterizes the relationship between the candidate value and the first value. Here, we will use case a in the above case A as an example to illustrate the first condition; the rest will refer to the second condition and will not be repeated. The first condition may include: the candidate value of the number L of the first basis vectors is greater than or equal to the value of the number L of the first basis vectors used by the terminal device to determine the second precoding.

[0340] The process of determining the first information based on the first CSI and the second CSI will be further explained below.

[0341] In one embodiment, the first CSI includes X measurement opportunities, meaning the first CSI includes CSI measurement results corresponding to X measurement opportunities. X is an integer greater than or equal to 1. The second CSI includes N4 prediction opportunities, meaning the second CSI includes CSI prediction results corresponding to N4 prediction opportunities. N4 is an integer greater than or equal to X. Optionally, the second CSI can be obtained from the predicted CSI report.

[0342] In one embodiment, step S420 above, determining the first information based on the first CSI and the second CSI, may include:

[0343] The third CSI is determined based on the second CSI; the third CSI includes the CSI prediction results corresponding to X prediction opportunities out of N4 prediction opportunities;

[0344] The first information is determined based on the first CSI and the third CSI.

[0345] In other words, the third CSI includes either the CSI prediction results corresponding to some or all of the prediction opportunities in the second CSI, and the number of prediction opportunities included in the third CSI is the same as the number of measurement opportunities included in the first CSI. Optionally, the prediction opportunities included in the third CSI are matched one-to-one with the measurement opportunities included in the first CSI. Matching prediction opportunities with measurement opportunities can be understood as: the time difference between the measurement opportunity and the prediction opportunity does not exceed a preset time threshold.

[0346] For example, Figure 5 is a schematic diagram of a timing example provided in an embodiment of this application. As shown in Figure 5, the second CSI includes N4 prediction timings, which are the N4 prediction timings in the prediction window. The first CSI includes X measurement timings from 1 to X. In Figure 5, to distinguish them from the measurement timings in the observation window, the measurement timings included in the first CSI are shown as filled with horizontal lines. The X measurement timings included in the first CSI can be matched one-to-one with any X prediction timings among the N4 prediction timings in the prediction window, and these X prediction timings are also the prediction timings included in the third CSI.

[0347] Based on the CSI measurement results corresponding to the X measurement opportunities in the first CSI and the CSI prediction results corresponding to the X prediction opportunities in the third CSI, the first information can be determined and a first report can be generated. The first report can be transmitted to the network device at some point after the X measurement opportunities, as shown in black in Figure 5.

[0348] The value of X can be set according to actual needs, and the following will explain them respectively.

[0349] 1) X = 1

[0350] When X=1, the third CSI includes the CSI prediction result corresponding to one of the N4 prediction opportunities of the second CSI. In this case, the terminal device only needs to determine the first information based on the CSI measurement result of one measurement opportunity and the CSI prediction result of one prediction opportunity, which reduces the amount of computation and the power consumption of the terminal device.

[0351] It is understandable that when X=1, since the first precoding information contains only one measurement opportunity, the terminal device can determine the first precoding information based on the first codebook type according to Example 1 above, that is, without using the third base vector; or, the terminal device can determine the first precoding information based on the second codebook type according to Example 2 above, and set the number Q of the third base vectors to 1.

[0352] The timing of this prediction can be determined in one of the following ways:

[0353] a. This 1 prediction opportunity is the first prediction opportunity among N4 prediction opportunities.

[0354] Generally, AI models are more accurate at earlier prediction points than at later prediction points. This method determines the first piece of information (CSI) based on the CSI prediction result at the first (i.e., the earliest) prediction point, thus assessing the AI ​​model's performance. If the CSI prediction result at the first prediction point is inaccurate, it indicates poor AI model performance. This method provides an accurate and simple way to monitor AI model performance.

[0355] b. This 1 prediction opportunity is the N4th prediction opportunity out of N4 prediction opportunities.

[0356] As mentioned above, the AI ​​model's CSI prediction accuracy is higher at earlier prediction times out of N4 prediction opportunities compared to later ones. In this method, the first piece of information is determined by the CSI prediction result at the N4th (and therefore the latest) prediction opportunity to assess the AI ​​model's performance. If the CSI prediction accuracy at the last prediction opportunity is good, then the AI ​​model's performance is considered good. This method provides an accurate and simple way to monitor the performance of AI models.

[0357] c. This 1 prediction opportunity is the nth prediction opportunity among N4 prediction opportunities, where n is an integer greater than 1 and less than N4.

[0358] In cases where the first or N4th prediction timing among the N4 prediction timings cannot match the measurement timings included in the first CSI, other prediction timings besides the first and N4th prediction timings can be selected to determine the first information, thereby improving the reliability of AI model performance monitoring.

[0359] d. The prediction timing is the prediction timing closest to the reception timing of a first RS among the N4 prediction timings.

[0360] As described above, the X sets of CSI measurement results in the first CSI can be obtained based on the first RS measurement, and therefore the X measurement times can be the reception times of the first RS. In this method, the first information is calculated based on the CSI prediction result corresponding to the prediction time closest to the reception time of the first RS, so that the matching degree between the prediction time and the measurement time is the highest, thereby making the calculated first information more accurate, and thus making the performance monitoring results of the AI ​​model more accurate.

[0361] 2) 1 < X ≤ N4

[0362] When 1 < X ≤ N4, the first CSI includes CSI measurement results corresponding to X measurement times, and the third CSI includes CSI prediction results corresponding to X prediction times. In other words, in this case, the terminal device determines the first information based on the CSI measurement results from multiple measurement times and the CSI prediction results from multiple prediction times. Thus, the first information can characterize the differences between the CSI measurement results and CSI prediction results at multiple times, reflecting the inference performance of the AI ​​model at different times, resulting in a more comprehensive and accurate monitoring of the AI ​​model's performance.

[0363] It is understandable that the specific methods for determining the first information can differ depending on the type of the first information and the specific information contained in the first CSI and the third CSI. In this embodiment, taking the example that the first CSI includes first precoding information, the third CSI includes second precoding information, and the first information includes one or more SGCS averages, determining the first information based on the first CSI and the third CSI can include:

[0364] Multiple SGCS values ​​are determined based on the first precoding information and the second precoding information;

[0365] One or more SGCS means are determined based on multiple SGCS values.

[0366] The first precoding information may include a first PMI and / or a first precoding matrix. The second precoding information may include a second PMI and / or a second precoding matrix.

[0367] In one specific embodiment, multiple SGCS values ​​can be determined according to the first precoding matrix and the second precoding matrix, according to the following formula (1):

[0368] Where l represents a layer, n3 represents a precoding subband, n4 represents the prediction timing included in the third CSI, l∈{1,…,v}, n3∈{1,…,N3}, n4∈{1,…,X}. The representation layer l, precoding subband n3, and prediction timing n4 correspond to the SGCS value. The representation layer l, precoding subband n3, prediction timing n4, and the second precoding matrix (the predicted precoding matrix) are shown. The representation layer l, precoding subband n3, and prediction timing n4 are represented by the first precoding matrix (the measured precoding matrix).

[0369] In other words, v×N3×X SGCS values ​​can be calculated according to the above formula (1). In this embodiment of the application, one or more SGCS averages are calculated based on the v×N3×X SGCS values, and a first report is generated based on the one or more SGCS averages and then fed back to the network device, which can reduce the overhead of feeding back the first report.

[0370] Optionally, the SGCS mean can be calculated in different ways depending on the value of X:

[0371] a. When X = 1, v × N3 SGCS values ​​can be calculated according to formula (1).

[0372] In this case, the average of the N3 SGCS values ​​corresponding to each layer can be calculated from the v×N3 SGCS values ​​to obtain the v SGCS mean. The v SGCS mean values ​​correspond to the similarity between the second and first precoding matrices of the v layers, respectively. This method allows network devices to easily understand the accuracy of the CSI predictions for different layers by the AI ​​model.

[0373] b. When 1 < X ≤ N4, v × N3 × X SGCS values ​​can be calculated according to formula (1).

[0374] In one implementation, the average of the v×N3×X SGCS values ​​corresponding to each prediction timing can be calculated to obtain X SGCS mean values. These X SGCS mean values ​​correspond to the similarity between the second precoding matrix and the first precoding matrix at the X prediction timings. This method facilitates network devices in understanding the accuracy of the CSI predictions made by the AI ​​model at different prediction timings.

[0375] In another implementation, the average of the N3 SGCS values ​​corresponding to each layer at each prediction time can be calculated from the v×N3×X SGCS values ​​to obtain the v×X SGCS mean. The v×X SGCS mean values ​​correspond to the similarity between the second and first precoding matrices of the v layers at X prediction times. This method allows network devices to understand the accuracy of the AI ​​model's predictions of CSI for different layers at different prediction times.

[0376] In one embodiment, at least one of the parameters related to the first CSI and the parameters related to the second CSI has the same value.

[0377] The parameters related to the first CSI can be understood as those that determine the parameters used in the first CSI. The parameters related to the second CSI can be understood as those that determine the parameters used in the second CSI. There are one or more common parameters between the parameters related to the first and second CSI. Common parameters can be understood as parameters that are defined identically in both the first and second CSI, or parameters that are defined differently but have the same type or attribute.

[0378] Optionally, if the first CSI includes the first precoding information, the parameters related to the first CSI may include, but are not limited to: the parameters related to the first RS, the number of physical resources contained in a frequency domain subband, the number of precoding subbands R contained in a frequency domain subband, the number of precoding subbands N3, the number of layers of the precoding matrix, the number of measurement opportunities contained in the first CSI, etc.

[0379] Optionally, if the second CSI includes the first precoding information, the parameters related to the second CSI may include, but are not limited to: the parameters related to CSI-RS in the observation window, the number of physical resources contained in a frequency domain subband, the number of precoding subbands R contained in a frequency domain subband, the number of precoding subbands N3, the number of layers of the precoding matrix, the number of prediction opportunities contained in the second CSI, etc.

[0380] Among the above parameters, the number of physical resources contained in a frequency domain subband, the number of precoding subbands R contained in a frequency domain subband, the number of precoding subbands N3, and the number of layers of the precoding matrix are defined with the same parameters, which are common parameters of the first CSI and the second CSI.

[0381] The parameters related to the first RS and the parameters related to CSI-RS are parameters with the same attributes or types, and are also common parameters of the first CSI and the second CSI. The parameters related to the first RS and the parameters related to CSI-RS can be collectively referred to as RS-related parameters. RS-related parameters may include, but are not limited to: bandwidth, frequency domain transmission density, number of antenna ports, etc.

[0382] The number of measurement opportunities included in the first CSI and the number of prediction opportunities included in the second CSI are parameters with the same attributes or types, and are also common parameters of the first and second CSIs.

[0383] In this embodiment, one or more of the aforementioned common parameters take values ​​consistent with those taken in determining the first CSI and the second CSI. This ensures that the dimensions of the first and second CSIs are consistent, thereby making the calculated first information more accurate and consequently, making the monitoring results of AI model performance more accurate.

[0384] The following describes the processing resources and processing time required for the terminal device to process the first report.

[0385] The processing of CSI reports by terminal devices requires a certain amount of processing resources and a certain processing time. Different CSI reports typically require different amounts of processing resources and processing time. The following explanation uses the processing time for processing CSI reports as an example, focusing on the time spent consuming processing resources. The terminal device pre-determines the required processing resources and the required resource consumption time based on the type of CSI report to be processed, in order to allocate processing resources appropriately.

[0386] In this embodiment, the process of the terminal device generating the first report includes both the process of acquiring the first CSI and the process of generating first information based on the first CSI and the second CSI. This process is relatively complex, and therefore the processing resources and time required to generate the first report may be lengthy. Considering this issue, the method provided in this embodiment sets appropriate processing resources for the first report and sets appropriate resource usage time for the first report by setting the first parameter and the second parameter corresponding to the first report. These will be explained separately below.

[0387] 1. Resource processing

[0388] Optionally, processing resources can be quantified through logical processing units. For example, the resource usage of a CSI report can be quantified through the processing unit (CPU) used to process the CSI. The following explanation uses CPU as an example of processing resources.

[0389] Different CSI reports require different numbers of CPUs. The terminal device can calculate and set the number of CPUs before processing a CSI report.

[0390] In this embodiment of the application, different CSI CPU occupancy numbers can be set according to the different number of measurement opportunities (i.e., the value of X) included in the first CSI.

[0391] 1) When X=1, the number of CPUs corresponding to the first report can be set to the predefined number of CPUs M. M is a positive integer.

[0392] Optionally, M is a predefined number of CPUs required to process CSI-related reports, such as those generated by a protocol or specification. Alternatively, M can be a predefined number of CPUs required to process reports of the same or similar type as the first report. For example, M could be a predefined number of CPUs required to process CSI report 1, which is a CSI report generated according to a second codebook type. As another example, M could be a predefined number of CPUs required to process CSI report 2, which is a report of CSIs predicted using a non-AI model and generated according to a second codebook type.

[0393] Optionally, M can be configured by the network device, preset by the terminal device, or set as a default value. For example, M can be 4 or 8.

[0394] In other words, when X=1, the original CPU number setting does not need to be changed. This is because when X=1, only one CSI prediction result corresponding to the measurement time and one CSI measurement result corresponding to the prediction time need to be processed, which does not require excessive processing resources. Therefore, not changing the original CPU number setting can prevent resource waste.

[0395] 2) When 1 < X ≤ N4, the number of CPUs corresponding to the first report can be determined based on the value of X and the predefined number of CPUs M.

[0396] Optionally, the number of CPUs corresponding to the first report can be set to the larger of Y3·X and M. For example, when X = N4, the number of CPUs = max(Y3·N4,M). Where Y3 ∈ {1,2,3}, and the value of Y3 depends on the capabilities of the terminal device. M is, for example, 4.

[0397] In other words, when 1 < X ≤ N4, if the number of CPUs Y3·X determined based on X is greater than the predefined number of CPUs M, then the number of CPUs corresponding to the first report is set to Y3·X; if the number of CPUs Y3·X determined based on X is less than the predefined number of CPUs M, then the number of CPUs corresponding to the first report is set to M. This ensures that the terminal device has sufficient processing resources to process the first report, conforms to the protocol or regulations, prevents resource waste, matches the processing resources occupied by the first report with the capabilities of the terminal device, and improves the effectiveness of the first report submission.

[0398] 2. Resource usage time

[0399] In this embodiment of the application, the value of the first parameter Z′ corresponding to the first report is greater than the first predefined value (hereinafter referred to as T1), and / or the value of the second parameter Z corresponding to the first report is greater than the second predefined value (hereinafter referred to as T2).

[0400] Here, the first predefined value T1 is a value of the first parameter Z′ corresponding to the predicted CSI-related report, predefined by relevant protocols or regulations. Alternatively, the first predefined value T1 is a value of the first parameter Z′ corresponding to a report of the same or similar type as the first report, predefined by protocols or regulations. For example, the first predefined value T1 is the value of the first parameter Z′ corresponding to CSI report 1. CSI report 1 is a CSI report generated according to the second codebook type. Another example is the value of the first parameter Z′ corresponding to CSI report 2. CSI report 2 is a CSI report predicted using a non-AI model and generated according to the second codebook type. The value of the first predefined value T1 can be found in Table 2 above.

[0401] The second predefined value T2 is a value of the second parameter Z corresponding to the predicted CSI-related report, predefined by relevant protocols or regulations. Alternatively, the second predefined value T2 is a value of the second parameter Z corresponding to a report of the same or similar type as the first report, predefined by protocols or regulations. For example, the second predefined value T2 is the value of the second parameter Z corresponding to CSI report 1, which is a CSI report generated based on the second codebook type. As another example, the second predefined value T2 is the value of the second parameter Z corresponding to CSI report 2, which is a CSI report predicted using a non-AI model and generated based on the second codebook type. The value of the second predefined value T2 can be found in Table 2 above.

[0402] As described above, the values ​​of the first parameter Z′ and the second parameter Z can be used to determine the resource occupancy duration, such as CPU occupancy duration. In this embodiment, the value of the first parameter Z′ corresponding to the first report is greater than the first predefined value T1, and / or the value of the second parameter Z corresponding to the first report is greater than the second predefined value T2, so that the resource occupancy duration determined by the terminal device is longer. This allows the terminal device sufficient time to determine the first information, generate the first report, and transmit the first report, thereby achieving a match between the processing time of the first report and the capabilities of the terminal device, and improving the effectiveness of the first report reporting.

[0403] The specific methods for determining the values ​​of the first parameter Z′ and the second parameter Z corresponding to the first report are explained below:

[0404] 1) The value of the first parameter Z′

[0405] In one embodiment, the value of the first parameter Z′ corresponding to the first report is related to at least one of the following:

[0406] The first predefined value T1, the first offset value ΔT1, and the first multiple s1; wherein the first offset value ΔT1 is a value greater than or equal to 0, and the first multiple s1 is a value greater than or equal to 1.

[0407] The specific values ​​of the first offset value ΔT1 and the first multiple s1 can be configured by the network device, preset by the terminal device, or set by the terminal device and then reported to the network device for determination.

[0408] In one specific embodiment, the value of the first parameter Z′ corresponding to the first report can be determined in one of the following ways:

[0409] a. The value of the first parameter Z′ corresponding to the first report is: the first predefined value T1, that is: the value of the first parameter Z′ = T1.

[0410] b. The value of the first parameter Z′ corresponding to the first report is the product of the first predefined value T1 and the first multiple s1, that is: the value of the first parameter Z′ = s1·T1.

[0411] c. The value of the first parameter Z′ corresponding to the first report is the sum of the first predefined value T1 and the first offset value ΔT1, that is: the value of the first parameter Z′ = T1 + ΔT1.

[0412] d. The value of the first parameter Z′ corresponding to the first report is the sum of the product of the first predefined value T1 and the first multiple s1 and the first offset value ΔT1, that is: the value of the first parameter Z′ = s1·T1+ΔT1.

[0413] Optionally, the values ​​of s1 and / or ΔT1 can differ depending on the number X of measurement opportunities included in the first CSI. For example, s1 can be 1.5 when X = 1; and s1 can be 2 when 1 < X ≤ N4.

[0414] In this embodiment, the value of the first parameter Z′ is determined based on at least one of the first predefined value T1, the first offset value ΔT1, and the first multiple s1. Furthermore, different methods can be used to determine the value of the first parameter Z′ in combination with the value of X. This not only allows the terminal device to have sufficient time to process the first report, but also references the predefined values ​​of protocols or regulations, thus meeting the requirements of the protocols or regulations. It also prevents the waste of time and resources, achieves the matching of the resource occupation time of the first report with the actual needs, and improves the effectiveness of the first report.

[0415] 2) The value of the second parameter Z

[0416] In one embodiment, the value of the second parameter Z corresponding to the first report is related to at least one of the following:

[0417] The second predefined value T2, the number of measurement opportunities X included in the first CSI, the interval d of the prediction opportunities included in the second CSI, the second offset value ΔT2, and the second multiple s2; wherein the second offset value ΔT2 is a value greater than or equal to 0, and the second multiple s2 is a value greater than or equal to 1.

[0418] The second offset value ΔT2 can be the same as or different from the first offset value ΔT1. The second multiple s2 can be the same as or different from the first multiple s1.

[0419] In one specific embodiment, the value of the first parameter Z′ corresponding to the first report can be determined in one of the following ways:

[0420] a. The value of the second parameter Z corresponding to the first report is: the second predefined value T2, that is: the value of the second parameter Z = T2.

[0421] b. The value of the second parameter Z corresponding to the first report is the product of the second predefined value T2 and the second multiple s2, that is: the value of the second parameter Z′ = s2·T2.

[0422] In some embodiments, this scheme can be selected when X=1.

[0423] c. The value of the second parameter Z corresponding to the first report is equal to the sum of at least one of the first predefined value T1, the second offset value ΔT2, and the second value, and the second predefined value T2. The second value is determined based on the number of measurement opportunities X included in the first CSI and the interval d between the prediction opportunities included in the second CSI.

[0424] Optionally, under this scheme, the value of the second parameter Z can be determined in different ways depending on the number X of measurement opportunities included in the first CSI.

[0425] For example, when X=1, the value of the second parameter Z corresponding to the first report can be the sum of the second predefined value T2 and the second offset value ΔT2, that is: the value of the first parameter Z′ = T2+ΔT2.

[0426] For example, when 1 < X ≤ N4, the value of the second parameter Z corresponding to the first report can be determined in one of the following ways:

[0427] ① The value of the second parameter Z corresponding to the first report is the sum of the second predefined value T2 and the second value, that is, the value of the second parameter Z = T2 + p;

[0428] ②The value of the second parameter Z corresponding to the first report is: the sum of the second predefined value T2, the second value and the first predefined value T1, that is, the value of the second parameter Z = T2 + p + T1; where p represents the second value.

[0429] ③ The value of the second parameter Z corresponding to the first report is: the second predefined value T2, the second value, the first predefined value T1 and the second offset ΔT2, that is, the value of the second parameter Z = T2 + p + T1 + ΔT2.

[0430] Optionally, the second value p can be determined in different ways depending on the type of the first RS.

[0431] For example, when the first RS is an aperiodic reference signal, the second value p can be: q(X-1)d, where q is a value greater than 1, for example, q can be 14.

[0432] For example, when the first RS is a periodic or semi-continuous reference signal, the second value p can be: q·X·d or q(X-1)d. q is a value greater than 1, for example, q can be 14.

[0433] In this embodiment, the value of the second parameter Z is determined based on at least one of the second predefined value T2, the number X of measurement opportunities included in the first CSI, the interval d of prediction opportunities included in the second CSI, the second offset value ΔT2, and the second multiple s2. Furthermore, different methods can be used to determine the value of the second parameter Z in combination with the value of X. This not only allows the terminal device to have sufficient time to process the first report, but also references predefined values ​​such as protocols or regulations, and prevents the waste of time and resources. It also achieves the matching of the resource occupation time of the first report with the actual needs, thereby improving the effectiveness of the first report.

[0434] It should be understood that the flowcharts or scenario diagrams shown in Figures 1 to 5 are for ease of understanding only and are not intended to limit the embodiments of this application to the examples shown. In fact, those skilled in the art can make equivalent transformations based on the examples in Figures 1 to 5 to obtain more implementation methods.

[0435] The communication methods provided by the embodiments of this application have been described in detail above with reference to Figures 1 to 5. The apparatus embodiments of this application will now be described in detail with reference to Figures 6 and 7. It should be understood that the communication apparatus of the embodiments of this application can execute the various communication methods described in the foregoing embodiments of this application; that is, the specific working processes of the various products described below can be referred to the corresponding processes in the foregoing method embodiments.

[0436] In the embodiments described above, the terminal device may execute some or all of the steps in each embodiment; the network device may execute some or all of the steps in each embodiment. These steps or operations are merely examples, and the embodiments of this application may also perform other operations or variations thereof. Furthermore, the steps may be executed in different orders as presented in the embodiments, and it is not necessary to execute all the operations in the embodiments of this application. Moreover, the sequence number of each step does not imply the order of execution; the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0437] Figure 6 is a schematic block diagram of a communication device provided in an embodiment of this application. As shown in Figure 6, the communication device 600 can have a processing module 610, which can implement corresponding processing functions.

[0438] Optionally, the communication device 600 further includes a storage module, which can be used to store instructions and / or data; the processing module 610 can read the instructions and / or data in the storage module so that the communication device 600 can implement the aforementioned method embodiments.

[0439] Optionally, the communication device 600 further includes a communication module 620. The communication module 620 can implement corresponding communication functions, which can be internal communication functions of the communication device 600 or communication functions between the communication device 600 and other devices. Optionally, the communication module 620 can also be referred to as a communication interface or transceiver module.

[0440] In one possible design, the communication device 600 may correspond to the terminal device in the above method embodiments, or to a component (such as a circuit, chip, or chip system) configured in the terminal device. The communication device 600 can be used to perform the steps or processes performed by the terminal device in any of the above method embodiments.

[0441] For example, the processing module 610 is used to acquire first channel state information (CSI); the first CSI is based on measurement and is determined according to the target value of the target parameter, the target value is a set of candidate values ​​of the target parameter, and the candidate values ​​are a set or more sets of predefined values ​​of the target parameter; first information is determined based on the first CSI and the second CSI; the second CSI is based on prediction, and the first information characterizes the difference between the first CSI and the second CSI.

[0442] In one embodiment, the first CSI includes first precoding information, and the target parameters include at least one of the following: parameters related to a first basis vector; the first basis vector is a spatially related basis vector; parameters related to a second basis vector; the second basis vector is a frequency-related basis vector; parameters related to a third basis vector; the third basis vector is a time-related basis vector; and parameters related to the weighting coefficients corresponding to at least one of the first, second, and third basis vectors.

[0443] In one embodiment, the first precoding information is determined based on a first codebook type, which indicates the precoding information determined according to a first base vector and a second base vector; the target parameter includes one or more codebook parameters corresponding to the first codebook type, and the predefined value of the target parameter includes one or more codebook parameter configurations corresponding to the first codebook type; or,

[0444] The first precoding information is determined based on the second codebook type, which is used to indicate the precoding information determined according to the first base vector, the second base vector, and the third base vector; the target parameter includes one or more of the codebook parameters corresponding to the second codebook type, and the predefined value of the target parameter includes one or more of the codebook parameter configurations corresponding to the second codebook type.

[0445] In one embodiment, the target parameters include the parameter combination index, the number L of the first basis vectors, and the parameter p related to the number of the second basis vectors. υ And at least one of the parameters β associated with the total number of weighting coefficients, wherein each value of the parameter combination index corresponds to a value of L, a value of p υ The value of and a value of β.

[0446] In one embodiment, the target parameter includes a parameter combination index; the candidate values ​​for the parameter combination index are:

[0447] The value that is greater than the preset index threshold among the predefined values ​​of the parameter combination index; or, the value that falls within the preset index range among the predefined values ​​of the parameter combination index; or, the maximum value supported by the terminal device among the predefined values ​​of the parameter combination index.

[0448] In one embodiment, the target value and / or candidate value of the target parameter are determined based on the first configuration information sent by the network device; or, the target value and / or candidate value of the target parameter are preset values; or, the target value and / or candidate value of the target parameter are determined based on a first value, wherein the first value is the value of the relevant parameter used to determine the second CSI.

[0449] In one embodiment, the first information is used to generate a first report; the value of the first parameter corresponding to the first report is greater than or equal to a first predefined value; and / or, the value of the second parameter corresponding to the first report is greater than or equal to a second predefined value; the value of the first parameter corresponding to the first report or the value of the second parameter corresponding to the first report is used to determine the time required for the terminal device to process the first report.

[0450] In one embodiment, the value of the first parameter corresponding to the first report is related to at least one of the following: a first predefined value, a first offset value, and a first multiple; wherein the first offset value is a value greater than or equal to 0, and the first multiple is a value greater than or equal to 1; the value of the second parameter corresponding to the first report is related to at least one of the following: a second predefined value, the number of measurement opportunities included in the first CSI, the interval of prediction opportunities included in the second CSI, a second offset value, and a second multiple; the measurement opportunities included in the first CSI refer to the measurement opportunities corresponding to the CSI measurement results included in the first CSI, the prediction opportunities included in the second CSI refer to the prediction opportunities corresponding to the CSI prediction results included in the second CSI, the second offset value is a value greater than or equal to 0, and the second multiple is a value greater than or equal to 1.

[0451] In one embodiment, the value of the first parameter corresponding to the first report is: the product of a first predefined value and a first multiple; or, the sum of the first predefined value and a first offset value; or, the sum of the product of the first predefined value and the first multiple and the first offset value.

[0452] In one embodiment, the value of the second parameter corresponding to the first report is:

[0453] The sum of at least one of a first predefined value, a second offset value, and a second value, and the second predefined value; wherein the second value is determined based on the number of measurement opportunities included in the first CSI and the interval between the prediction opportunities included in the second CSI; or, the product of the second predefined value and a second multiple.

[0454] In one embodiment, the first information is used to generate a first report; the number of processing resources required to process the first report is: a predefined number of processing resources, or determined according to the number of measurement opportunities included in the first CSI; the measurement opportunities included in the first CSI refer to the prediction opportunities corresponding to the CSI measurement results included in the first CSI.

[0455] In one embodiment, at least one of the parameters related to the first CSI and the parameters related to the second CSI has the same value.

[0456] In one embodiment, at least one parameter includes at least one of the following: bandwidth, frequency domain transmission density, number of antenna ports, number of physical resources contained in a frequency domain subband, number of precoding subbands contained in a frequency domain subband, number of precoding subbands, number of layers of the precoding matrix, and number of measurement opportunities contained in the first CSI, wherein the measurement opportunities contained in the first CSI refer to the measurement opportunities corresponding to the CSI measurement results contained in the first CSI.

[0457] In one embodiment, the first CSI includes CSI measurement results corresponding to X measurement opportunities, and the second CSI includes CSI prediction results corresponding to N4 prediction opportunities, where X is an integer greater than or equal to 1 and N4 is an integer greater than or equal to X; the processing module 610 is specifically used to determine the third CSI based on the second CSI; the third CSI includes CSI prediction results corresponding to X prediction opportunities out of the N4 prediction opportunities; and the first information is determined based on the first CSI and the third CSI.

[0458] In one embodiment, when X is 1: the X prediction opportunities are the first prediction opportunity among N4 prediction opportunities; or, the X prediction opportunities are the N4th prediction opportunity among N4 prediction opportunities; or, the X prediction opportunities are the nth prediction opportunity among N4 prediction opportunities, where n is an integer greater than 1 and less than N4; or, the X prediction opportunities are the prediction opportunities among N4 prediction opportunities that are closest to the reception opportunity of a first reference signal, which is used to obtain the first CSI.

[0459] In one embodiment, the first CSI includes first pre-encoded information, and the third CSI includes second pre-encoded information. The first information includes one or more squared generalized cosine similarity (SGCS) mean values. The processing module 610 is specifically used to determine multiple SGCS values ​​based on the first pre-encoded information and the second pre-encoded information, and to determine one or more SGCS mean values ​​based on the multiple SGCS values.

[0460] In one embodiment, the communication module 620 is used to send first information to the network device.

[0461] In one embodiment, the communication module 620 is specifically used to send a first report to a network device, the first report including first information.

[0462] The above are merely examples; for detailed steps or procedures, please refer to the descriptions in the foregoing embodiments.

[0463] Figure 7 is a schematic block diagram of another communication device provided in an embodiment of this application. As shown in Figure 7, the communication device 700 can have a processing module 710, which can implement corresponding processing functions.

[0464] Optionally, the communication device 700 further includes a storage module, which can be used to store instructions and / or data; the processing module 710 can read the instructions and / or data in the storage module so that the communication device 600 can implement the aforementioned method embodiments.

[0465] Optionally, the communication device 700 further includes a communication module 720. The communication module 720 can implement corresponding communication functions, which can be internal communication functions of the communication device 700 or communication functions between the communication device 700 and other devices. Optionally, the communication module 720 can also be referred to as a communication interface or transceiver module.

[0466] In one possible design, the communication device 700 may correspond to the network device in the above method embodiments, or to a component (such as a circuit, chip, or chip system) configured in the network device. The communication device 700 can be used to perform the steps or processes performed by the network device in any of the above method embodiments.

[0467] For example, the communication module 720 is used to receive first information sent by the terminal device; the first information characterizes the difference between the first CSI and the second CSI, the first CSI is based on measurement, the second CSI is based on prediction, the first CSI is determined according to the target value of the target parameter, the target value is a set of candidate values ​​of the target parameter, and the candidate values ​​are one or more sets determined from the predefined values ​​of the target parameter.

[0468] In one embodiment, the communication module 720 is further configured to send first configuration information to the terminal device, the first configuration information including the target value of the target parameter and / or the candidate value of the target parameter.

[0469] In one embodiment, the communication module 720 is further configured to send a first reference signal to the terminal device, the first reference signal being used to measure a first CSI.

[0470] In one embodiment, the processing module 710 is configured to switch the CSI acquisition mode based on a first report. The acquisition mode may include an AI model-based prediction mode and a measurement mode.

[0471] The above are merely examples; for detailed steps or procedures, please refer to the descriptions in the foregoing embodiments.

[0472] Figure 8 is another schematic block diagram of a communication device 800 provided in an embodiment of this application. The communication device 800 may be a chip, chip system, or processor, etc., used by a terminal device or network device to implement the above-described methods. The communication device 800 can be used to implement the methods described in the above-described method embodiments; for details, please refer to the descriptions in the above-described method embodiments.

[0473] As shown in Figure 8, the communication device 800 may include one or more processors 810, which may also be referred to as processing units or processing modules, and can implement certain control functions. The processor 810 may be a general-purpose processor or a dedicated processor, such as 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 800 (e.g., base station, baseband chip, user, user chip), execute software programs, and process data from the software programs.

[0474] In an alternative design, the processor 810 may also store instructions and / or data, which can be executed by the processor 810 to cause the communication device 800 to perform the methods described in the above method embodiments.

[0475] In another alternative design, the communication device 800 may include a communication interface 820 for implementing receiving and transmitting functions. For example, the communication interface 820 may be a transceiver circuit, interface, interface circuit, or transceiver. The transceiver circuit, interface, interface circuit, or transceiver for implementing receiving and transmitting functions may be separate or integrated. The aforementioned transceiver circuit, interface, interface circuit, or transceiver may be used for reading and writing code / data, or it may be used for transmitting or relaying signals.

[0476] Optionally, the communication device 800 may include one or more memories 830, which may store instructions that can be executed on the processor 810, causing the communication device 800 to perform the methods described in the above method embodiments. Optionally, the memories 830 may also store data. Optionally, the processor 810 may also store instructions and / or data. The processor 810 and the memories 830 may be provided separately or integrated together.

[0477] It should be understood that, in one possible design, the steps in the method embodiments provided in this application can be implemented by integrated logic circuits in the processor's hardware or by instructions in software form. The steps of the methods disclosed in the embodiments of this application can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules in the processor. The software modules can reside in random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, registers, or other mature storage media in the art. This storage medium is located in memory, and the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method. To avoid repetition, detailed descriptions are not provided here.

[0478] In one implementation, the communication device 800 may correspond to the terminal device in the above method embodiments and may be used to execute the various steps and / or processes executed by the terminal device in the above method embodiments. The processor 810 may be used to execute instructions stored in the memory 830, and when the processor 810 executes the instructions stored in the memory, the processor 810 is used to execute the various steps and / or processes of the above method embodiments corresponding to the terminal device.

[0479] In another implementation, the communication device 800 may correspond to the network device in the above method embodiments and may be used to execute the various steps and / or processes executed by the network device in the above method embodiments. The processor 810 may be used to execute instructions stored in the memory 830, and when the processor 810 executes the instructions stored in the memory, the processor 810 is used to execute the various steps and / or processes of the above method embodiments corresponding to the network device.

[0480] It should be understood that the aforementioned processing device can be one or more chips. For example, the processing device can be a field-programmable gate array (FPGA), an application-specific integrated circuit (ASIC), a system-on-chip (SoC), a central processor unit (CPU), a network processor (NP), a digital signal processor (DSP), a microcontroller unit (MCU), a programmable logic device (PLD), or other integrated chips.

[0481] It is understood that the memory in the embodiments of this application can be volatile memory or non-volatile memory, or may include both volatile and non-volatile memory. The non-volatile memory can be read-only memory (ROM), programmable read-only memory (PROM), erasable programmable read-only memory (EPROM), electrically erasable programmable read-only memory (EEPROM), or flash memory. The volatile memory can be random access memory (RAM), which is used as an external cache. By way of example, but not limitation, many forms of RAM are available, such as static random access memory (SRAM), dynamic random access memory (DRAM), synchronous dynamic random access memory (SDRAM), double data rate synchronous dynamic random access memory (DDR SDRAM), enhanced synchronous dynamic random access memory (ESDRAM), synchronous linked dynamic random access memory (SLDRAM), and direct rambus RAM (DR RAM). It should be noted that the memory used in the systems and methods described herein is intended to include, but is not limited to, these and any other suitable types of memory.

[0482] According to the method provided in the embodiments of this application, this application also provides a chip system, which includes one or more processors for calling and executing instructions stored in memory, thereby causing the method described in the embodiments of this application to be executed. The chip system may be composed of chips or may include chips and other discrete devices.

[0483] The chip system may include input circuits or interfaces for transmitting information or data, and output circuits or interfaces for receiving information or data.

[0484] According to the method provided in the embodiments of this application, this application also provides a communication system, which includes the aforementioned network device and terminal device.

[0485] According to the method provided in the embodiments of this application, this application also provides a computer program product, which includes: computer program code, which, when run on a computer, causes the computer to execute the various steps or processes executed by the network device or terminal device in any of the foregoing method embodiments.

[0486] According to the method provided in the embodiments of this application, this application also provides a computer-readable storage medium storing program code, which, when run on a computer, causes the computer to execute the various steps or processes executed by the network device or terminal device in any of the foregoing method embodiments.

[0487] The computer-readable storage medium may be the aforementioned volatile memory or non-volatile memory, or it may include both volatile memory and non-volatile memory.

[0488] In the embodiments of this application, the terms and English abbreviations are exemplary examples given for ease of description and should not be construed as limiting the application in any way. This application does not preclude the possibility of defining other terms that can achieve the same or similar functions in existing or future agreements.

[0489] In the above embodiments, implementation can be achieved, in whole or in part, through software, hardware, firmware, or any combination thereof. When implemented in software, it can be implemented, in whole or in part, as a computer program product. A computer program product includes one or more computer instructions. When the computer instructions are loaded and executed on a computer, all or part of the flow or function according to the embodiments of this application is generated.

[0490] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of 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 system, 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 apparatuses or units may be electrical, mechanical, or other forms.

[0491] It should be understood that in the various embodiments of this application, the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0492] In summary, the above are merely preferred embodiments of the technical solutions of this application and are not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A communication method, executed by a terminal device, characterized in that, include: Acquire first channel state information (CSI); the first CSI is based on measurement and is determined according to the target value of the target parameter, wherein the target value is a set of candidate values ​​of the target parameter, and the candidate values ​​are one or more sets of predefined values ​​of the target parameter; First information is determined based on the first CSI and the second CSI; The second CSI is obtained based on prediction, and the first information characterizes the difference between the first CSI and the second CSI.

2. The method according to claim 1, characterized in that, The first CSI includes first precoding information, and the target parameter includes at least one of the following: Parameters related to the first basis vector; the first basis vector is a spatial domain-related basis vector; The parameters are related to the second basis vector; the second basis vector is a frequency-domain related basis vector. Parameters related to the third basis vector; the third basis vector is a time-domain related basis vector; The parameters related to the weighting coefficients of at least one of the first basis vector, the second basis vector, and the third basis vector.

3. The method according to claim 2, characterized in that, The first precoding information is determined based on a first codebook type, which is used to indicate the determination of precoding information according to the first base vector and the second base vector; the target parameter includes one or more codebook parameters corresponding to the first codebook type, and the predefined value of the target parameter includes one or more codebook parameter configurations corresponding to the first codebook type. or, The first precoding information is determined based on a second codebook type, which is used to indicate the determination of precoding information according to the first base vector, the second base vector, and the third base vector; the target parameter includes one or more codebook parameters corresponding to the second codebook type, and the predefined value of the target parameter includes one or more codebook parameter configurations corresponding to the second codebook type.

4. The method according to claim 2 or 3, characterized in that, The target parameters include the parameter combination index, the number L of the first basis vectors, and the parameter p related to the number of the second basis vectors. υ and at least one of the parameters β associated with the total number of the weighting coefficients, wherein each value of the parameter combination index corresponds to a value of L, a value of p υ The value of and the value of β.

5. The method according to claim 4, characterized in that, The target parameter includes the parameter combination index; the candidate values ​​of the parameter combination index are: The predefined values ​​of the parameter combination index are greater than the preset index threshold; or, The values ​​that belong to the preset index range among the predefined values ​​of the parameter combination index; or, The maximum value supported by the terminal device among the predefined values ​​of the parameter combination index.

6. The method according to any one of claims 1 to 5, characterized in that, The target value and / or candidate value of the target parameter are determined based on the first configuration information sent by the network device; or, The target value of the target parameter and / or the candidate value of the target parameter are preset values; or, The target value of the target parameter and / or the candidate value of the target parameter are determined according to a first value, which is the value of the relevant parameter used to determine the second CSI.

7. The method according to any one of claims 1 to 6, characterized in that, The first information is used to generate the first report; The value of the first parameter corresponding to the first report is greater than or equal to a first predefined value; and / or, The value of the second parameter corresponding to the first report is greater than or equal to the second predefined value; The value of the first parameter corresponding to the first report or the value of the second parameter corresponding to the first report is used to determine the time required for the terminal device to process the first report.

8. The method according to claim 7, characterized in that, The value of the first parameter corresponding to the first report is related to at least one of the following: The first predefined value, the first offset value, and the first multiple; wherein, the first offset value is a value greater than or equal to 0, and the first multiple is a value greater than or equal to 1; The value of the second parameter corresponding to the first report is related to at least one of the following: The second predefined value, the number of measurement opportunities included in the first CSI, the interval of prediction opportunities included in the second CSI, the second offset value, and the second multiple; the measurement opportunities included in the first CSI refer to the measurement opportunities corresponding to the CSI measurement results included in the first CSI, the prediction opportunities included in the second CSI refer to the prediction opportunities corresponding to the CSI prediction results included in the second CSI, the second offset value is a value greater than or equal to 0, and the second multiple is a value greater than or equal to 1.

9. The method according to claim 8, characterized in that, The value of the first parameter corresponding to the first report is: The product of the first predefined value and the first multiple; or, The sum of the first predefined value and the first offset value; or, The product of the first predefined value and the first multiple, and the sum of the first offset value.

10. The method according to claim 8 or 9, characterized in that, The value of the second parameter corresponding to the first report is: The sum of at least one of a first predefined value, a second offset value, and a second value, and the second predefined value; wherein the second value is determined based on the number of measurement opportunities included in the first CSI and the interval of prediction opportunities included in the second CSI; or, The product of the second predefined value and the second multiple.

11. The method according to any one of claims 1 to 10, characterized in that, The first information is used to generate the first report; The number of processing resources required to process the first report is: a predefined number of processing resources, or determined based on the number of measurement opportunities included in the first CSI; the measurement opportunities included in the first CSI refer to the prediction opportunities corresponding to the CSI measurement results included in the first CSI.

12. The method according to any one of claims 1 to 11, characterized in that, The value of at least one of the parameters related to the first CSI and the parameters related to the second CSI is the same.

13. The method according to claim 12, characterized in that, The at least one parameter includes at least one of the following: The parameters include bandwidth, frequency domain transmission density, number of antenna ports, number of physical resources contained in a frequency domain subband, number of precoding subbands contained in a frequency domain subband, number of precoding subbands, number of layers in the precoding matrix, and number of measurement opportunities contained in the first CSI. The measurement opportunities contained in the first CSI refer to the measurement opportunities corresponding to the CSI measurement results contained in the first CSI.

14. The method according to any one of claims 1 to 13, characterized in that, The first CSI includes CSI measurement results corresponding to X measurement opportunities, and the second CSI includes CSI prediction results corresponding to N4 prediction opportunities, where X is an integer greater than or equal to 1 and N4 is an integer greater than or equal to X. The step of determining the first information based on the first CSI and the second CSI includes: A third CSI is determined based on the second CSI; the third CSI includes the CSI prediction results corresponding to X prediction opportunities out of the N4 prediction opportunities; The first information is determined based on the first CSI and the third CSI.

15. The method according to claim 14, characterized in that, When X is 1: The X prediction opportunities are the first prediction opportunities among the N4 prediction opportunities; Alternatively, the X prediction opportunities are the N4th prediction opportunity among the N4 prediction opportunities; Alternatively, the X prediction opportunities are the nth prediction opportunity among the N4 prediction opportunities, where n is an integer greater than 1 and less than N4; Alternatively, the X prediction opportunities are the prediction opportunities among the N4 prediction opportunities that are closest to the reception opportunity of a first reference signal, and the first reference signal is used to acquire the first CSI.

16. The method according to claim 14 or 15, characterized in that, The first CSI includes first precoding information, and the third CSI includes second precoding information. The first information includes one or more squared generalized cosine similarity (SGCS) mean values. Determining the first information based on the first CSI and the third CSI includes: Based on the first precoding information and the second precoding information, multiple SGCS values ​​are determined; Based on the plurality of SGCS values, determine the mean of one or more SGCS values.

17. A communication method, executed by a network device, characterized in that, include: Receive the first information sent by the terminal device; The first information characterizes the difference between the first CSI and the second CSI. The first CSI is based on measurement, and the second CSI is based on prediction. The first CSI is determined according to the target value of the target parameter. The target value is a set of candidate values ​​of the target parameter, and the candidate values ​​are one or more sets determined from the predefined values ​​of the target parameter.

18. The method according to claim 17, characterized in that, Before receiving the first information sent by the receiving terminal device, the method further includes: Send first configuration information to the terminal device, the first configuration information including the target value of the target parameter and / or the candidate value of the target parameter.

19. The method according to claim 18, characterized in that, The method further includes: A first reference signal is sent to the terminal device, the first reference signal being used to measure the first CSI.

20. A communication device, characterized in that, The device includes at least one processor coupled to a memory storing a program or instructions, the processor executing the program or instructions to cause the device to perform the method as described in any one of claims 1 to 19.

21. A computer-readable storage medium having a computer program or instructions stored thereon, characterized in that, When the computer program or instructions are executed, they cause the computer to perform the method as described in any one of claims 1 to 19.

22. A communication system, characterized in that, Includes the communication device as described in claim 20.

23. A chip system comprising one or more processors, the one or more processors being configured to retrieve and execute instructions stored in memory, such that the method of any one of claims 1 to 19 is performed.