Communication method and communication device

Abstract

An embodiment of the present application provides a communication method and a communication device. The method includes: acquiring p groups to be transmitted on a first resource, where the p groups belong to N downlink channel state information (CSI) reports, where p is an integer greater than 1 and N is an integer greater than 1; and transmitting some or all of the p groups on the first resource based on priorities of the p groups, where the priorities of the p groups are based on priorities of the N CSI reports, the priorities of the CSI reports are based on types of the CSI reports, and the type of at least one of the N CSI reports is an artificial intelligence (AI)-related type. The solution in the embodiment of the present application can realize efficient transmission of CSI reports, thereby enabling a network device to obtain desired related information in the CSI reports.

Description

[Technical Field]

[0001] This application claims priority to Chinese Patent Application No. 202310176651.5, entitled "Communication Method and Communication Apparatus," filed with the State Intellectual Property Office of China on February 16, 2023, the entire contents of which are incorporated herein by reference.

[0002] Technical Field TECHNICAL FIELD Embodiments of the present application relate to the field of communications, and more particularly to a communication method and a communication device. [Background technology]

[0003] In a communication system, a network device needs to determine downlink channel-related configuration information, such as resources for scheduling a downlink data channel of a terminal device, a modulation and coding scheme (MCS), and precoding, based on CSI parameters. The terminal device calculates downlink CSI parameters by measuring a downlink reference signal and feeds back the downlink CSI parameters to the network device through a CSI report. CSI feedback modes include codebook-based CSI feedback, AI model-based CSI feedback, etc. In some scenarios, the size of a CSI report to be transmitted may exceed the channel's carrying capacity. For example, when multiple CSI reports need to be transmitted on the same resource, the size of the CSI report to be transmitted may exceed the channel's carrying capacity.

[0004] Therefore, how to implement CSI feedback to allow network devices to obtain CSI reports becomes an urgent problem to be solved. Summary of the Invention [Problem to be solved by the invention]

[0005] Embodiments of the present application provide a communication method and apparatus for implementing efficient transmission of CSI reports and allowing network devices to obtain desired relevant information in the CSI reports. [Means for solving the problem]

[0006] According to a first aspect, there is provided a communication method, which may be performed by a terminal device or by a chip or circuit located within the terminal device, although this is not a limitation of the present application.

[0007] The method includes the steps of: obtaining p groups to be transmitted on a first resource, where the p groups belong to N channel state information (CSI) reports, where p is an integer greater than 1 and N is an integer greater than 1; and transmitting some or all of the p groups on the first resource based on priorities of the p groups, where the priorities of the p groups are based on priorities of the N CSI reports, the priorities of the CSI reports are based on types of the CSI reports, and the type of at least one of the N CSI reports is an AI-related type.

[0008] According to the solution in this embodiment of the present application, CSI reports to be transmitted on the same resource include AI-related CSI reports, and the priority of the CSI report is based on the type of the CSI report. Priorities of p groups in N CSI reports may be determined based on the priorities of the N CSI reports, whereby some or all of the p groups are transmitted on the same resource based on the priorities of the p groups. This helps network devices appropriately use resources to obtain necessary information. For example, if the same resource cannot carry all of the p groups, some groups may be selected for transmission based on the priorities of the p groups, whereby higher-priority groups are transmitted and lower-priority groups are discarded. This allows network devices to obtain channel information within higher-priority groups, for example, within important groups.

[0009] Referring to the first aspect, in some implementations of the first aspect, the priority of a CSI report is based on one or more parameters indicating at least one of the following: a periodicity of the CSI report, a channel on which the CSI report is carried, whether the CSI report includes a layer 1-reference signal received power (L1-RSRP) or a layer 1 signal-to-interference-plus-noise ratio (L1-SINR), a serving cell to which the CSI report belongs, or an identity (ID) of the CSI report. At least one of the one or more parameters further indicates the type of the CSI report, where the type of the CSI report includes the AI-related type or the AI-non-related type.

[0010] For example, for a certain parameter, different values ​​of the parameter may indicate different information.

[0011] With reference to the first aspect, in some implementations of the first aspect, the priority of a CSI report is based on one or more parameters, where at least one of the one or more parameters indicates a type of the CSI report, where the type of the CSI report includes an AI-related type or an AI-non-related type, and another parameter of the one or more parameters indicates at least one of: a periodicity of the CSI report, a channel on which the CSI report is carried, whether the CSI report includes L1-RSRP or L1-SINR, a serving cell to which the CSI report belongs, or an ID of the CSI report.

[0012] For example, the at least one parameter indicating the type of CSI report may not indicate other information.

[0013] With reference to the first aspect, in some implementations of the first aspect, the AI-related type includes one or more of a first CSI report type, a second CSI report type, or a third CSI report type, and the report content of the first CSI report type includes channel information obtained through inference by an AI model, the report content of the second CSI report type includes performance information of the AI ​​model, or the report content of the third CSI report type includes channel information for implementing an operation related to the AI ​​model.

[0014] Referring to the first aspect, in some implementations of the first aspect, the priority of the CSI report may alternatively be based on an AI model associated with the CSI report.

[0015] With reference to the first aspect, in some implementations of the first aspect, the N CSI reports include a first CSI report, a type of the first CSI report is a first CSI report type, the first CSI report includes p1 groups among the p groups, and priorities of the p1 groups are based on a priority of the first CSI report and priorities of the p1 groups in the first CSI report, where p1 is an integer greater than 1.

[0016] For example, the N CSI reports may further include a fourth CSI report, the type of the fourth CSI report being the first CSI report type, and the fourth CSI report may include p1' groups among the p groups, and the priorities of the p1' groups may be based on the priority of the fourth CSI report and the priorities of the p1' groups in the fourth CSI report. The priority of the fourth CSI report is lower than the priority of the first CSI report, and the priority of each group in the fourth CSI report is lower than the priority of the group in the first CSI report corresponding to that group in the fourth CSI report. p1' = p1. The p1 groups in the first CSI report include group p1-1 and group p1-2, and correspondingly, the p1' groups in the fourth CSI report include group p1'-1 and group p1'-2. The priority of group p1-1 is higher than the priority of group p1-2, the priority of group p1'-1 is higher than the priority of group p1'-2, the priority of group p1-1 is higher than the priority of group p1'-1, the priority of group p1'-1 is higher than the priority of group p1-2, and the priority of group p1-2 is higher than the priority of group p1'-2.

[0017] There is a correspondence between a group in the fourth CSI report and a group in the first CSI report. If the priority of a group in the fourth CSI report in the fourth CSI report is the same as the priority of a group in the first CSI report in the first CSI report, the two groups may be considered to be groups having a correspondence.

[0018] In other words, for two CSI reports of the first CSI report type, during priority determination, the priority of the group is considered first, and then the priority of the CSI report to which the corresponding group belongs is considered.

[0019] In this embodiment of the present application, the first CSI report may be divided into multiple groups, thereby improving the flexibility of information transmission and increasing the probability that a network device will acquire a group in the first CSI report. This helps the network device acquire partial information in the CSI report. For example, the first CSI report is divided into multiple groups, and the groups correspond to different priorities. When the first resource cannot carry p groups, even if the entire first CSI report cannot be transmitted, the probability that the network device will acquire some groups in the first CSI report can be increased. This helps the network device acquire partial information in the CSI report.

[0020] With reference to the first aspect, in some implementations of the first aspect, the channel information obtained through inference by the AI ​​model includes channel information of multiple layers, and the p1 groups each include channel information of at least one layer among the multiple layers, and the priorities of the p1 groups in the first CSI report match the priorities of the channel information of the at least one layer in the p1 groups.

[0021] For example, the priority of the channel information of a layer is related to the eigenvalues ​​or singular values ​​corresponding to the eigenvectors of the layer. The priority of the channel information of a layer is sorted in ascending order of the singular values ​​corresponding to the eigenvalues ​​or eigenvectors of the layer, or the priority of the channel information of a layer is sorted in descending order of the singular values ​​corresponding to the eigenvalues ​​or eigenvectors of the layer. The beamforming gain of different layers may be different. In general, larger eigenvalues ​​or singular values ​​corresponding to a layer indicate larger beamforming gain of that layer. The sparsity of the eigenvectors of different layers may be different. In general, larger eigenvalues ​​or singular values ​​corresponding to a layer indicate stronger sparsity of the eigenvectors of that layer, i.e., higher feedback accuracy for the same feedback overhead.

[0022] In this embodiment of the present application, the priority of the channel information acquired through AI inference may be classified based on layers, and the priority of the channel information of a layer may match the priority of the group to which the layer belongs in the first CSI report. Different layers may have different beamforming gains. Therefore, the beamforming gains of different layers may be distinguished through layer-based priority classification. This is useful for implementing priority classification based on different beamforming gains. For example, a layer with a larger beamforming gain may be used as a higher-priority layer, and the higher-priority layer belongs to a higher-priority group. In this way, it may be ensured that this portion of CSI can be fed back preferentially. In other words, the probability that a network device acquires channel information for a layer with a larger beamforming gain is increased. In addition, the sparsity of the eigenvectors of different layers may be different, i.e., the feedback accuracy may be different. Therefore, the feedback accuracy of different layers may be distinguished through layer-based priority classification. This is useful for implementing priority classification based on different feedback accuracy. For example, a layer with a higher feedback accuracy may be used as a higher-priority layer, and the higher-priority layer belongs to a higher-priority group. In this way, it can be ensured that this portion of CSI can be fed back preferentially, i.e., the probability that the network device acquires channel information for a layer with higher feedback accuracy increases.

[0023] With reference to the first aspect, in some implementations of the first aspect, the channel information obtained through inference by the AI ​​model includes channel information of a plurality of segments, and p1 groups each include channel information of at least one of the plurality of segments, and the priority of the p1 groups in the first CSI report matches the priority of the channel information of the at least one segment in the p1 groups.

[0024] For example, the priority of the channel information of a segment is related to whether CSI can be independently recovered based on the segment, and the channel information of the segment based on which CSI can be independently recovered has the highest priority.

[0025] In this embodiment of the present application, priority classification is performed based on segments, whereby segments based on which CSI can be independently recovered have higher priorities, and higher priority segments belong to higher priority groups. In this way, it can be preferentially guaranteed that this portion of CSI can be fed back. In other words, the probability that a network device will obtain a segment based on which CSI can be independently recovered increases, which helps the network device recover CSI. In this way, even if the entire first CSI report cannot be transmitted, the probability that the network device will obtain a higher priority group can be increased. Even if a segment with a lower priority is discarded, the network device can still recover CSI based on the remaining segments with a higher priority.

[0026] With reference to the first aspect, in some implementations of the first aspect, the priority of the at least one segment is related to a layer in which the at least one segment is located and a position of the at least one segment in the layer in which the at least one segment is located.

[0027] The channel information acquired through inference by the AI ​​model may include channel information of multiple layers. Some or all of the channel information of the multiple layers is segmented to acquire channel information of multiple segments. The priority of a segment is related to the layer in which the segment is located and the position of the segment in the layer in which the segment is located. The position of a segment in the layer in which the segment is located may be understood as the segment in that layer. For example, the position of a segment in the layer in which the segment is located may be indicated by the index of the segment in the layer in which the segment is located. The layer in which the segment is located may be indicated by the index of the layer in which the segment is located. For example, the priority of the ith segment in the jth layer is related to the values ​​of i and j, where i indicates the index of the segment in the layer in which the segment is located, and j indicates the index of the layer in which the segment is located.

[0028] With reference to the first aspect, in some implementations of the first aspect, the N CSI reports include a second CSI report, a type of the second CSI report is a second CSI report type, the second CSI report includes p2 groups among the p groups, and priorities of the p2 groups are based on a priority of the second CSI report and a priority of the p2 groups in the second CSI report, where p2 is an integer greater than 1.

[0029] For example, the N CSI reports may further include a fifth CSI report, the type of the fifth CSI report being the second CSI report type, and the fifth CSI report may include p2' groups out of the p groups, and the priorities of the p2' groups may be based on the priority of the fifth CSI report and the priorities of the p2' groups in the fifth CSI report. The priority of the fifth CSI report may be lower than the priority of the second CSI report, and the priority of each group in the fifth CSI report may be lower than the priority of the group in the second CSI report corresponding to that group in the fifth CSI report. p2' = p2. The p2 groups in the second CSI report include group p2-1 and group p2-2, and correspondingly, the p2' groups in the fifth CSI report include group p2'-1 and group p2'-2. The priority of group p2-1 is higher than the priority of group p2-2, the priority of group p2'-1 is higher than the priority of group p2'-2, the priority of group p2-1 is higher than the priority of group p2'-1, the priority of group p2'-1 is higher than the priority of group p2-2, and the priority of group p2-2 is higher than the priority of group p2'-2.

[0030] There is a correspondence between a group in the fifth CSI report and a group in the second CSI report. If the priority of a group in the fifth CSI report in the fifth CSI report is the same as the priority of a group in the second CSI report in the second CSI report, the two groups may be considered to be groups having a correspondence.

[0031] In other words, for two CSI reports of the second CSI report type, during priority determination, the priority of the group is considered first, and then the priority of the CSI report to which the corresponding group belongs is considered.

[0032] In this embodiment of the present application, the second CSI report may be divided into multiple groups, thereby improving the flexibility of information transmission and increasing the probability that a network device will acquire a group in the second CSI report. This helps the network device acquire partial information in the CSI report. For example, the second CSI report is divided into multiple groups, and the groups correspond to different priorities. When the first resource cannot carry p groups, even if the entire second CSI report cannot be transmitted, the probability that the network device will acquire some groups in the second CSI report can be increased. This helps the network device acquire partial information in the CSI report.

[0033] With reference to the first aspect, in some implementations of the first aspect, the performance information of the AI ​​model includes overall performance information, performance information of multiple layers, or performance information of multiple subbands, each of the p2 groups includes at least one performance information in the performance information of the AI ​​model, and the priority of the p2 groups in the second CSI report matches the priority of the at least one performance information in the p2 groups.

[0034] For example, the overall performance information of the AI ​​model may be the overall accuracy of the reconstructed CSI information. The overall accuracy of the reconstructed CSI information may be determined based on the accuracy of each layer and / or each subband within the reconstructed CSI information. For example, the overall accuracy of the reconstructed CSI information may be expressed as an average value of the SGCS between the reconstructed information of each layer and / or each subband and the raw information of each layer and / or each subband.

[0035] With reference to the first aspect, in some implementations of the first aspect, the priority of the overall performance information is higher than the priority of the performance information of multiple layers, or the priority of the overall performance information is higher than the priority of the performance information of multiple subbands.

[0036] In this embodiment of the present application, the overall performance information may be prioritized over other performance information, which helps increase the probability that the network device will obtain the overall performance information, thereby determining the overall performance of the AI ​​model and implementing subsequent configuration.

[0037] With reference to the first aspect, in some implementations of the first aspect, a third CSI report among the N CSI reports includes one of the p groups.

[0038] The third CSI report may be any one of the N CSI reports.

[0039] In other words, for p groups, some CSI reports contain only one group.

[0040] For example, the CSI reports may not be divided into multiple groups, ie, the CSI reports may be considered as one group as a whole.

[0041] According to the solution in this embodiment of the present application, for some information with high integrity requirements, if part of the information is discarded, the subsequent configuration may be affected. The information with high integrity requirements may be used as one of the p groups. In this way, all of the information with high integrity requirements may be discarded or uploaded.

[0042] With respect to the first aspect, in some implementations of the first aspect, p=N.

[0043] That is, each CSI report contains one of the p groups.

[0044] For example, none of the N CSI reports may be divided into multiple groups, i.e., each of the N CSI reports may be considered as a group as a whole. The priority of the p groups is based on the priorities of the N CSI reports.

[0045] According to the solution in this embodiment of the present application, for any information with high integrity requirements, if part of the information is discarded, subsequent configuration may be affected. If information with high integrity requirements exists in each of N CSI reports, the information with high integrity requirements in each CSI report may be used separately as one of p groups. In this way, all of the information with high integrity requirements may be discarded or uploaded.

[0046] Referring to the first aspect, in some implementations of the first aspect, at least two of the N CSI reports have the same priority.

[0047] For example, the prioritization scheme may be applied to a scenario in which multiple CSI reports need to be used together. For multiple CSI reports with the same priority, when the multiple CSI reports need to be discarded, the multiple CSI reports may be discarded simultaneously, or a part of a group with the same priority may be discarded based on the priority in the multiple CSI reports.

[0048] With reference to the first aspect, in some implementations of the first aspect, the method further includes receiving first indication information from the network device, the first indication information indicating priorities of some or all of the N CSI reports.

[0049] With reference to the first aspect, in some implementations of the first aspect, the first indication information is carried in one or more of radio resource control (RRC) signaling, a medium access control element (MAC CE), or downlink control information (DCI).

[0050] With reference to the first aspect, in some implementations of the first aspect, the step of transmitting some or all of the p groups on the first resource based on the priorities of the p groups may include the steps of selecting some groups from the p groups based on the priorities of the p groups, and transmitting the some groups on the first resource.

[0051] With reference to the first aspect, in some implementations of the first aspect, the step of transmitting some or all of the p groups on the first resource based on the priorities of the p groups may include determining the positions of some or all of the p groups on the first resource based on the priorities of the p groups, and transmitting some or all of the p groups on the first resource.

[0052] According to a second aspect, there is provided a communication method, which may be performed by a terminal device or by a chip or circuit located within the terminal device, although this is not a limitation of the present application.

[0053] The method includes obtaining p groups of CSI reports to be transmitted on a first resource, where p is an integer greater than 1; and transmitting some or all of the p groups on the first resource based on priorities of the p groups in the CSI reports, where the type of the CSI report is an AI-related type.

[0054] According to the solution in this embodiment of the present application, an AI-related CSI report may be internally divided into multiple groups, and some or all of the p groups may be transmitted based on the priorities of the p groups to help appropriately use resources, thereby allowing the network device to obtain necessary information. For example, if the same resource cannot carry all of the p groups, some groups may be selected for transmission based on the priorities of the p groups, with higher-priority groups being transmitted and lower-priority groups being discarded, thereby allowing the network device to obtain channel information within the higher-priority groups, for example, within important groups. In this way, the flexibility of information transmission can be improved, and even if the entire AI-related CSI report cannot be transmitted, the probability that the network device will obtain some groups can be increased, thereby allowing the network device to obtain partial information within the CSI report.

[0055] With reference to the second aspect, in some implementations of the second aspect, the AI-related type includes one or more of a first CSI report type, a second CSI report type, or a third CSI report type, and the report content of the first CSI report type includes channel information obtained through inference by an AI model, the report content of the second CSI report type includes performance information of the AI ​​model, or the report content of the third CSI report type includes channel information for implementing an operation related to the AI ​​model.

[0056] With reference to the second aspect, in some implementations of the second aspect, the type of the CSI report is a first CSI report type, the channel information obtained through inference by the AI ​​model includes channel information of multiple layers, each of the p groups includes channel information of at least one layer among the multiple layers, and priorities of the p groups in the CSI report match priorities of the channel information of the at least one layer in the p groups.

[0057] With reference to the second aspect, in some implementations of the second aspect, the channel information obtained through inference by the AI ​​model includes channel information for a plurality of segments, each of the p groups includes channel information for at least one segment among the plurality of segments, and the priority of the p groups in the CSI report matches the priority of the channel information for the at least one segment in the p groups.

[0058] Referring to the second aspect, in some implementations of the second aspect, the priority of the at least one segment is related to a layer in which the at least one segment is located and a position of the at least one segment in the layer in which the at least one segment is located.

[0059] With reference to the second aspect, in some implementations of the second aspect, the type of the CSI report is a second CSI report type, the performance information of the AI ​​model includes overall performance information, performance information of multiple layers, or performance information of multiple subbands, each of the p groups includes at least one performance information among the performance information of the AI ​​model, and the priority of the p groups in the CSI report matches the priority of the at least one performance information among the p groups.

[0060] With reference to the second aspect, in some implementations of the second aspect, the priority of the overall performance information is higher than the priority of the performance information of multiple layers, or the priority of the overall performance information is higher than the priority of the performance information of multiple subbands.

[0061] According to a third aspect, there is provided a communication method, which may be performed by a network device or by a chip or circuit located within the network device, although this is not a limitation of the present application.

[0062] The method includes receiving q groups on a first resource, where the q groups are some or all of p groups, the p groups belong to N CSI reports, q is a positive integer, p is an integer greater than 1, and N is an integer greater than 1, priorities of the p groups are based on priorities of the N CSI reports, priorities of the CSI reports are based on types of the CSI reports, and a type of at least one report among the N CSI reports is an AI-related type; and communicating with a terminal device based on the q groups.

[0063] With reference to the third aspect, in some implementations of the third aspect, the priority of a CSI report is based on one or more parameters indicating at least one of the following: a periodicity of the CSI report, a channel on which the CSI report is carried, whether the CSI report includes a Layer 1 reference signal received power (L1-RSRP) or a Layer 1 signal-to-interference-and-noise ratio (L1-SINR), a serving cell to which the CSI report belongs, or an identity (ID) of the CSI report. At least one of the one or more parameters further indicates the type of the CSI report, wherein the type of the CSI report includes the AI-related type or the AI-non-related type.

[0064] With reference to the third aspect, in some implementations of the third aspect, the priority of a CSI report is based on one or more parameters, where at least one of the one or more parameters indicates a type of the CSI report, where the type of the CSI report includes an AI-related type or an AI-non-related type, and another parameter of the one or more parameters includes at least one of a periodicity of the CSI report, a channel on which the CSI report is carried, whether the CSI report includes L1-RSRP or L1-SINR, a serving cell to which the CSI report belongs, or an ID of the CSI report.

[0065] With reference to the third aspect, in some implementations of the third aspect, the AI-related type includes one or more of a first CSI report type, a second CSI report type, or a third CSI report type, and the report content of the first CSI report type includes channel information obtained through inference by an AI model, the report content of the second CSI report type includes performance information of the AI ​​model, or the report content of the third CSI report type includes channel information for implementing an operation related to the AI ​​model.

[0066] Referring to the third aspect, in some implementations of the third aspect, the N CSI reports include a first CSI report, a type of the first CSI report is a first CSI report type, the first CSI report includes p1 groups among the p groups, and priorities of the p1 groups are based on a priority of the first CSI report and priorities of the p1 groups in the first CSI report, where p1 is an integer greater than 1.

[0067] With reference to the third aspect, in some implementations of the third aspect, the channel information obtained through inference by the AI ​​model includes channel information of multiple layers, each of the p1 groups includes channel information of at least one layer among the multiple layers, and priorities of the p1 groups in the first CSI report match priorities of the channel information of the at least one layer in the p1 groups.

[0068] With reference to the third aspect, in some implementations of the third aspect, the channel information obtained through inference by the AI ​​model includes channel information of a plurality of segments, each of the p1 groups includes channel information of at least one segment among the plurality of segments, and the priority of the p1 groups in the first CSI report matches the priority of the channel information of the at least one segment in the p1 groups.

[0069] With reference to the third aspect, in some implementations of the third aspect, the priority of the at least one segment is related to a layer in which the at least one segment is located and a position of the at least one segment in the layer in which the at least one segment is located.

[0070] Referring to the third aspect, in some implementations of the third aspect, the N CSI reports include a second CSI report, the type of the second CSI report is a second CSI report type, the second CSI report includes p2 groups among the p groups, and priorities of the p2 groups are based on a priority of the second CSI report and a priority of the p2 groups among the second CSI report, where p2 is an integer greater than 1.

[0071] With reference to the third aspect, in some implementations of the third aspect, the performance information of the AI ​​model includes overall performance information, performance information of multiple layers, or performance information of multiple subbands, each of the p2 groups includes at least one performance information in the performance information of the AI ​​model, and the priority of the p2 groups in the second CSI report matches the priority of the at least one performance information in the p2 groups.

[0072] With reference to the third aspect, in some implementations of the third aspect, the priority of the overall performance information is higher than the priority of the performance information of multiple layers, or the priority of the overall performance information is higher than the priority of the performance information of multiple subbands.

[0073] With respect to the third aspect, in some implementations of the third aspect, p=N.

[0074] Referring to the third aspect, in some implementations of the third aspect, at least two of the N CSI reports have the same priority.

[0075] In relation to the third aspect, in some implementations of the third aspect, the method further includes sending first indication information indicating priorities of some or all of the N CSI reports.

[0076] According to a fourth aspect, there is provided a communication method, which may be performed by a network device or by a chip or circuit located within the network device, although this is not a limitation of the present application.

[0077] The method includes: transmitting first indication information indicating priorities of some or all of the N CSI reports, where N is an integer greater than 1, and a report type of at least one of the N CSI reports is an AI-related type; and receiving, on a first resource, some or all of the p groups, where some or all of the p groups are based on priorities of the p groups, where the p groups belong to the N CSI reports, and the priorities of the p groups are based on priorities of the N CSI reports, where p is an integer greater than 1.

[0078] With reference to the fourth aspect, in some implementations of the fourth aspect, the first indication information is carried in one or more of RRC signaling, MAC CE, or DCI.

[0079] Referring to the fourth aspect, in some implementations of the fourth aspect, at least two of the N CSI reports have the same priority.

[0080] Referring to the fourth aspect, in some implementations of the fourth aspect, the priority of the CSI report is based on the type of the CSI report, and the type of the CSI report includes an AI-related type or an AI-non-related type.

[0081] With reference to the fourth aspect, in some implementations of the fourth aspect, the AI-related type includes one or more of a first CSI report type, a second CSI report type, or a third CSI report type, and the report content of the first CSI report type includes channel information obtained through inference by an AI model, the report content of the second CSI report type includes performance information of the AI ​​model, or the report content of the third CSI report type includes channel information for implementing an operation related to the AI ​​model.

[0082] It will be understood that extensions, limitations, explanations and descriptions of relevant content in the first aspect are also applicable to the same content in the second, third and fourth aspects.

[0083] According to a fifth aspect, there is provided a communications apparatus. The communications apparatus may be a terminal device, or may be a device, module, circuit, chip, etc. located within the terminal device, or may be an apparatus that can be used in conjunction with the terminal device. In one design, the communications apparatus may include modules in a one-to-one correspondence with the methods / operations / steps / actions described in the first or second aspect. The modules may be implemented by hardware circuits, software, or a combination of hardware circuits and software. In one design, the communications apparatus may include a processing module and a communications module.

[0084] The sending module is configured to perform the sending action in the method described in the first aspect, and the processing module is configured to perform the processing-related action in the method described in the first aspect. Alternatively, the sending module is configured to perform the sending action in the method described in the second aspect, and the processing module is configured to perform the processing-related action in the method described in the second aspect.

[0085] According to a sixth aspect, a communications apparatus is provided. The communications apparatus may be a network device, or a device, module, circuit, chip, etc. located within the network device, or may be a device usable with the network device. In one design, the communications apparatus may include modules in a one-to-one correspondence with the methods / operations / steps / actions described in the third or fourth aspect. The modules may be implemented by hardware circuits, software, or a combination of hardware circuits and software. In one design, the communications apparatus may include a processing module and a communications module.

[0086] The receiving module is configured to perform the receiving operations of the methods described in the third or fourth aspect, the transmitting module is configured to perform the transmitting operations of the methods described in the third or fourth aspect, and the processing module is configured to perform the processing-related operations of the methods described in the third or fourth aspect.

[0087] According to a seventh aspect, there is provided a communications device. The communications device includes a processor and a storage medium. The storage medium stores instructions. When executed by the processor, the instructions implement a method according to the first aspect or any one of possible implementations of the first aspect, a method according to the second aspect or any one of possible implementations of the second aspect, a method according to the third aspect or any one of possible implementations of the third aspect, or a method according to the fourth aspect or any one of possible implementations of the fourth aspect.

[0088] According to an eighth aspect, a communication device is provided. The communication device includes a processor. The processor is configured to process data and / or information, thereby implementing a method according to the first aspect or any one of its possible implementations, a method according to the second aspect or any one of its possible implementations, a method according to the third aspect or any one of its possible implementations, or a method according to the fourth aspect or any one of its possible implementations. Optionally, the communication device may further include a communication interface. The communication interface is configured to receive the data and / or information and transmit the received data and / or information to the processor. Optionally, the communication interface is further configured to output the data and / or information processed by the processor.

[0089] According to a ninth aspect, a chip is provided. The chip includes a processor. The processor is configured to execute a program or instructions to implement a method according to the first aspect or any one of its possible implementations, a method according to the second aspect or any one of its possible implementations, a method according to the third aspect or any one of its possible implementations, or a method according to the fourth aspect or any one of its possible implementations. Optionally, the chip may further include a memory, the memory configured to store the program or instructions. Optionally, the chip may further include an input / output circuit or an interface.

[0090] According to a tenth aspect, there is provided a computer-readable storage medium comprising instructions that, when executed by a processor, implement a method according to the first aspect or any one of its possible implementations, a method according to the second aspect or any one of its possible implementations, a method according to the third aspect or any one of its possible implementations, or a method according to the fourth aspect or any one of its possible implementations.

[0091] According to an eleventh aspect, there is provided a computer program product comprising computer program code or instructions that, when executed, implement a method according to the first aspect or any one of possible implementations of the first aspect, a method according to the second aspect or any one of possible implementations of the second aspect, a method according to the third aspect or any one of possible implementations of the third aspect, or a method according to the fourth aspect or any one of possible implementations of the fourth aspect.

[0092] According to a twelfth aspect, there is provided a communication system including one or a combination of the following devices: a communication device performing the first aspect or any one of possible implementations of the first aspect, a communication device performing the second aspect or any one of possible implementations of the second aspect, a communication device performing the third aspect or any one of possible implementations of the third aspect, or a communication device performing the fourth aspect or any one of possible implementations of the fourth aspect. [Brief explanation of the drawings]

[0093] [Figure 1] 1 is a diagram of a communication system to which certain embodiments of the present application are applicable;

[0094] [Figure 2] FIG. 1 is a diagram of another communication system to which certain embodiments of the present application can be applied.

[0095] [Figure 3] FIG. 1 is a block diagram of an autoencoder.

[0096] [Figure 4] Diagram of AI application framework.

[0097] [Figure 5] 1 is a schematic flowchart of a communication method according to an embodiment of the present application;

[0098] [Figure 6] FIG. 2 is a diagram of a method for determining segment priorities according to an embodiment of the present application.

[0099] [Figure 7] FIG. 10 is a diagram of another method for determining segment priorities according to an embodiment of the present application.

[0100] [Figure 8] FIG. 1 is a diagram of a report format for a CSI report according to an embodiment of the present application.

[0101] [Figure 9] FIG. 1 is a diagram of a report format for multiple CSI reports, according to an embodiment of the present application.

[0102] [Figure 10] FIG. 10 is another diagram of a report format for multiple CSI reports according to an embodiment of the present application.

[0103] [Figure 11] FIG. 10 is yet another diagram of a report format for multiple CSI reports according to an embodiment of the present application.

[0104] [Figure 12] FIG. 10 is yet another diagram of a report format for multiple CSI reports according to an embodiment of the present application.

[0105] [Figure 13] 1 is a block diagram of a communication device according to an embodiment of the present application;

[0106] [Figure 14] FIG. 2 is a block diagram of another communication device according to an embodiment of the present application. DETAILED DESCRIPTION OF THE INVENTION

[0107] The following describes the technical solution of the present application with reference to the accompanying drawings.

[0108] The technical solutions provided herein may be applied to various communication systems, for example, future communication systems such as a 5th generation (5G) or new radio (NR) system, a long term evolution (LTE) system, an LTE frequency division duplex (FDD) system, an LTE time division duplex (TDD) system, a wireless local area network (WLAN) system, a satellite communication system, a 6th generation (6G) mobile communication system, or an integrated system of multiple systems. The technical solutions provided herein may further be applied to device-to-device (D2D) communication, vehicle-to-everything (V2X) communication, machine-to-machine (M2M) communication, machine type communication (MTC), internet of things (IoT) communication system, or another communication system.

[0109] A network element in a communication system can transmit a signal to or receive a signal from another network element. The signal may include information, signaling, data, etc. A network element may alternatively be referred to as an entity, a network entity, a device, a communication device, a communication module, a node, a communication node, etc. In the present disclosure, a network element is used as an illustrative example. For example, a communication system may include at least one terminal device and at least one network device. The network device may transmit a downlink signal to the terminal device, and / or the terminal device may transmit an uplink signal to the network device. It may be understood that the terminal device in the present disclosure may be replaced by a first network element, and the network device may be replaced by a second network element, and the terminal device and the network device perform the corresponding communication method in the present disclosure.

[0110] In embodiments herein, a terminal device may also be referred to as user equipment (UE), access terminal, subscriber unit, subscriber station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent, or user equipment.

[0111] The terminal device may be a device that provides voice / data to a user, for example, a handheld device or an in-vehicle device with wireless connectivity. Currently, some examples of terminals are mobile phones, tablet computers, notebook computers, palmtop computers, mobile internet devices (MIDs), wearable devices, virtual reality (VR) devices, augmented reality (AR) devices, wireless terminals in industrial control, wireless terminals in self-driving, wireless terminals in remote medical surgery, wireless terminals in smart grids, wireless terminals in transportation safety, wireless terminals in smart cities, wireless terminals in smart homes, cellular phones, cordless phones, session initiation protocol (SIP) phones, wireless local loop (WLL) stations, personal digital assistants (PDAs), handheld devices or computing devices with wireless communication capabilities, separate processing devices or wearable devices connected to a wireless modem, terminal devices in 5G networks, future evolved public land mobile networks, and the like. The present invention is not limited to this embodiment.

[0112] By way of example and not limitation, in embodiments of the present application, the terminal device may alternatively be a wearable device. A wearable device may also be referred to as a wearable intelligent device, which is a general term for wearable devices, such as glasses, gloves, watches, clothing, and shoes, that are intelligently designed and developed for everyday wear by using wearable technology. A wearable device is a portable device that can be worn directly on the body or integrated into a user's clothing or accessories. A wearable device is not only a hardware device, but also implements powerful functions through software support, data exchange, and cloud interaction. In a broad sense, wearable intelligent devices include full-featured large devices, such as smart watches or smart glasses, that can implement full or partial functions without relying on a smartphone, and devices that are dedicated to only one type of application function and need to cooperate with other devices, such as smartphones, such as various smart bands or smart jewelry for monitoring physical symptoms.

[0113] In an embodiment of the present application, an apparatus configured to implement the functions of a terminal device may be a terminal device itself, or may be an apparatus capable of supporting a terminal device in implementing the functions, such as a chip system. The apparatus may be installed in a terminal device or used together with a terminal device. In this embodiment of the present application, the chip system may include a chip, or may include a chip and another discrete component. In the embodiment of the present application, the fact that the apparatus for implementing the functions of a terminal device is a terminal device is merely used as an example for explanation and does not constitute any limitation on the solution in the embodiment of the present application.

[0114] A network device in an embodiment of the present application may be a device configured to communicate with a terminal device. A network device may also be referred to as an access network device or a radio access network device. For example, a network device may be a base station. A network device in an embodiment of the present application may be a radio access network (RAN) node (or device) that connects a terminal device to a wireless network. A base station may broadly cover or be replaced by the following various names: For example, a NodeB (NodeB), evolved NodeB (eNB), next generation NodeB (gNB), relay station, access point, transmitting and receiving point (TRP), transmitting point (TP), primary station, secondary station, multi-standard radio (motor slide retainer (MSR)) node, home base station, network controller, access node, radio node, access point (AP), transmitting node, transceiver node, baseband unit (BBU), remote radio unit (RRU), active antenna unit (AAU), remote radio head (RRH), central unit (CU), distributed unit (DU), and positioning node. The base station may be a macro base station, a micro base station, a relay node, a donor node, etc., or a combination thereof. The base station may alternatively be a communication module, a modem, or a chip disposed in the aforementioned device or apparatus. The base station may alternatively be a mobile switching center, a device performing base station functions in D2D, V2X and M2M communications, a network side device in a 6G network, a device performing base station functions in future communication systems, etc.The base stations may support networks of the same access technology or different access technologies. The specific technology used for the network devices and the specific device configuration are not limited to the embodiments of this application.

[0115] A base station may be fixed or mobile. For example, a helicopter or unmanned aerial vehicle may be configured as a mobile base station, and one or more cells may move based on the location of the mobile base station. In another example, a helicopter or unmanned aerial vehicle may be configured as a device for communicating with another base station.

[0116] In some deployments, a network device referred to in embodiments of the present application may be a device including a CU or DU, a device including a CU and a DU, or a device including a control plane CU node (central unit-control plane (CU-CP)), a user plane CU node (central unit-user plane (CU-UP)), and a DU node. For example, the network device may include a gNB-CU-CP, a gNB-CU-UP, and a gNB-DU.

[0117] In the embodiments of the present application, the apparatus for implementing the functions of the network device may be a network device, or may be an apparatus capable of supporting the network device in implementing the functions, such as a chip system, a hardware circuit, a software module, or a combination of a hardware circuit and a software module. The apparatus may be mounted on the network device or used together with the network device. In the embodiments of the present application, the fact that the apparatus for implementing the functions of the network device is a network device is merely used as an example for explanation and does not constitute any limitation on the solution in the embodiments of the present application.

[0118] The network devices and / or terminal devices may be located on land, water, or airborne, including indoor, outdoor, handheld, or vehicle-mounted devices. The scenarios in which the network devices and terminal devices exist are not limited in the embodiments of this application. Additionally, each of the terminal devices and network devices may be a hardware device, a software function running on dedicated hardware, a software function running on general-purpose hardware, a virtualization function instantiated on a platform (e.g., a cloud platform), or an entity that includes dedicated or general-purpose hardware devices and software functions. The specific forms of the terminal devices and network devices are not limited in this application.

[0119] In wireless communication networks, e.g., mobile communication networks, the services supported by the network are becoming increasingly diverse, and therefore the requirements that must be met are becoming increasingly diverse. For example, networks need to be able to support ultra-high speeds, ultra-low latency, and / or a large number of connections. This characteristic increases the complexity of network planning, network configuration, and / or resource scheduling. In addition, as networks have increasingly powerful capabilities, e.g., supporting increasingly higher spectrum and new technologies such as higher-order multiple input multiple output (MIMO) technology, beamforming, and / or beam management, network energy conservation has become a hot research topic. These new requirements, scenarios, and characteristics pose unprecedented challenges to network planning, operation and maintenance, and efficient operation. To address these challenges, artificial intelligence techniques can be introduced into wireless communication networks to implement network intelligence.

[0120] To support AI technology in wireless networks, AI nodes may be further introduced into the network.

[0121] Optionally, the AI ​​node may be located in one or more of the following locations within a communications system: an access network device, a terminal device, a core network device, etc. Alternatively, the AI ​​node may be independently located, e.g., located in a location other than any one of the aforementioned devices, e.g., in a host or cloud server of an over-the-top (OTT) system. The AI ​​node may communicate with another device within the communications system. The other device may be, e.g., one or more of a network device, a terminal device, a network element of a core network, etc.

[0122] It may be understood that the number of AI nodes is not limited in this application. For example, if there are multiple AI nodes, the multiple AI nodes may be divided based on functions. For example, different AI nodes are responsible for different functions.

[0123] It may be further understood that an AI node may be an independent device, or may be integrated into the same device to implement different functions, or may be a network element within a hardware device, or may be a software function running on dedicated hardware, or may be a virtualized function instantiated on a platform (e.g., a cloud platform). The specific form of an AI node is not limited herein.

[0124] An AI node may be an AI network element or an AI module.

[0125] 1 is a diagram of a communication system to which a communication method according to an embodiment of the present application can be applied. As shown in FIG. 1, the communication system 100 may include at least one network device, for example, the network device 110 shown in FIG. 1. The communication system 100 may further include at least one terminal device, for example, the terminal device 120 and the terminal device 130 shown in FIG. 1. The network device 110 may communicate with the terminal devices (e.g., the terminal device 120 and the terminal device 130) through a wireless link. The communication devices in the communication system, for example, the network device 110 and the terminal device 120, may communicate with each other by using multiple antenna technology.

[0126] 2 is a diagram of another communication system to which a communication method according to an embodiment of the present application can be applied. Compared with the communication system 100 shown in FIG. 1, the communication system 200 shown in FIG. 2 further includes an AI network element 140. The AI ​​network element 140 is configured to perform AI-related operations, such as building a training dataset or training an AI model.

[0127] In one possible implementation, the network device 110 may transmit data related to the training of the AI ​​model to the AI ​​network element 140, which constructs a training dataset and trains the AI ​​model. For example, the data related to the training of the AI ​​model may include data reported by the terminal device. The AI ​​network element 140 may transmit results of operations related to the AI ​​model to the network device 110, which forwards the results to the terminal device. For example, the results of operations related to the artificial intelligence model may include at least one of the trained AI model, evaluation results or test results for the model, etc. For example, a portion of the trained AI model may be located on the network device 110, and another portion may be located on the terminal device. Alternatively, the trained AI model may be located on the network device 110. Alternatively, the trained AI model may be deployed on the terminal device.

[0128] It should be understood that in FIG. 2 , the AI ​​network element 140 is directly connected to the network device 110 is used merely as an example for purposes of illustration. In another scenario, the AI ​​network element 140 may alternatively be connected to a terminal device. Alternatively, the AI ​​network element 140 may be connected to both the network device 110 and the terminal device. Alternatively, the AI ​​network element 140 may be connected to the network device 110 via a third-party network element. The connection relationship between the AI ​​network element and another network element is not limited in the embodiments of the present application.

[0129] The AI ​​network element 140 may alternatively be located as a module in a network device and / or terminal device, for example, in the network device 110 or terminal device shown in FIG. 1 .

[0130] 1 and 2 are merely simplified diagrams of examples for ease of understanding. For example, the communication system may further include other devices, such as a wireless relay device and / or a wireless backhaul device not shown in FIGS. 1 and 2. In actual applications, the communication system may include multiple network devices and may also include multiple terminal devices. The number of network devices and terminal devices included in the communication system is not limited in the embodiments of the present application.

[0131] To facilitate understanding of the solutions in the embodiments of the present application, the following describes terms that may be used in the embodiments of the present application.

[0132] (1) AI model:

[0133] An AI model is an algorithm or computer program that can implement an AI function. An AI model describes the mapping relationship between the model's inputs and outputs. The type of AI model can be a neural network, linear regression model, decision tree model, support vector machine (SVM), Bayesian network, Q-learning model, or another machine learning (ML) model.

[0134] (2) Two-sided model:

[0135] A two-sided model is sometimes called a bilateral model, collaborative model, dual model, or two-sided model. A two-sided model is a model that contains multiple submodels. The multiple submodels contained in the model must match each other. The multiple submodels can be deployed on different nodes.

[0136] An embodiment of the present application relates to an encoder for compressing CSI and a decoder for restoring the compressed CSI. The encoder and decoder are used together. The encoder and decoder can be understood as AI models that match each other. One encoder can include one or more AI models, and the decoder that matches the encoder also includes one or more AI models. The encoder and decoder used together include the same number of AI models, and the AI ​​models included in the encoder have a one-to-one correspondence with the AI ​​models included in the decoder.

[0137] In one possible design, the encoder and decoder used together can be two parts of the same auto-encoder (AE), for example, as shown in Figure 3. An AE model in which the encoder and decoder are separately located in different nodes is a typical bilateral model. The encoder and decoder of an AE model are generally trained jointly and used together. The encoder processes an input V to obtain a processed result z, and the decoder can decode the encoder output z into the expected output V'.

[0138] An autoencoder is an unsupervised learning neural network that uses input data as label data. Therefore, an autoencoder may be understood as a self-supervised learning neural network. An autoencoder may be configured to compress and decompress data. For example, an encoder within an autoencoder may compress (encode) data A to obtain data B, and a decoder within an autoencoder may decompress (decode) data B to restore data A. Alternatively, a decoder may be understood as the inverse operation of an encoder.

[0139] For example, the AI ​​model in the embodiment of the present application may include an encoder and a decoder. The encoder and the decoder are used together. It can be understood that the encoder and the decoder are AI models that match each other. The encoder and the decoder can be separately located on a terminal device and a network device.

[0140] Alternatively, the AI ​​model in the embodiments of the present application may be a one-sided model, and the AI ​​model may be located in a terminal device or a network device.

[0141] (3) Neural network (NN):

[0142] A neural network is a specific implementation of AI or machine learning. According to the universal approximation theorem, a neural network can theoretically approximate any continuous function, thereby giving it the ability to learn any mapping.

[0143] A neural network contains neurons, each of which has a set of x s and intercept 1 as inputs. A neural network is a network formed by connecting many single neurons together. Specifically, the output of a neuron can be the input of another neuron. The input of each neuron can be connected to the local receptive field of the previous layer to extract features of the local receptive field. A local receptive field can be an area containing several neurons.

[0144] For example, the type of the AI ​​model is a neural network. The AI ​​model in the present disclosure may be a deep neural network (DNN). Based on the network construction mode, the DNN may include a feedforward neural network (FNN), a convolutional neural network (CNN), a recurrent neural network (RNN), etc.

[0145] (4) Training dataset and inference data:

[0146] A training dataset is used to train an AI model. The training dataset may include inputs for the AI ​​model, or may include inputs and target outputs for the AI ​​model. The training dataset includes one or more training data. The training data may include training samples that are input to the AI ​​model, or may include target outputs for the AI ​​model. The target outputs may also be referred to as labels or label samples.

[0147] In the communications field, training datasets may include simulated data collected using a simulation platform, experimental data collected in experimental scenarios, or actual measurement data collected in a real communications network. Because the geographical environments and channel conditions in which the data are generated vary, e.g., indoors, outdoors, moving speeds, frequency bands, antenna configurations, etc., the collected data may be classified when the data is acquired. For example, data with the same channel propagation environment and the same antenna configuration may be classified into one type.

[0148] Model training essentially involves learning some features of the training data from the training data. In the process of training an AI model (e.g., a neural network model), the AI ​​model's output is expected to be as close as possible to the actual desired predicted value. Therefore, the current network's predicted value may be compared with the actual desired target value, and the weight vector of each layer of the AI ​​model is updated based on the difference between the predicted value and the target value. (Of course, before the first update, an initialization process is generally performed, specifically, parameters for all layers of the AI ​​model are preconfigured.) For example, if the network's predicted value is large, the weight vector is adjusted to reduce the predicted value, and adjustments are continuously performed until the AI ​​model can predict the actual desired target value or a value very close to the actual desired target value. Therefore, it is necessary to predefine how to obtain the difference between the predicted value and the target value through comparison. This is the loss function or objective function. Loss functions and objective functions are important formulas that measure the difference between the predicted value and the target value. The loss function is used as an example. A larger output value (loss) of the loss function indicates a larger difference. In this case, training for an AI model is the process of minimizing the loss so that the value of the loss function is less than a threshold or so that the value of the loss function meets a target requirement. For example, if the AI ​​model is a neural network, adjusting the model parameters of the neural network includes adjusting at least one of the number and width of the neural network layers, the weights of the neurons, or parameters in the activation functions of the neurons.

[0149] The inference data may be used as input to a trained AI model and used for inference by the AI ​​model. During model inference, the inference data is input to the AI ​​model to obtain a corresponding output, i.e., an inference result.

[0150] (5) AI model design:

[0151] AI model design mainly includes a data collection phase (e.g., collecting training data and / or inference data), a model training phase, and a model inference phase, and may further include an inference result application phase.

[0152] Figure 4 shows the AI ​​application framework.

[0153] In the data collection phase, a data source is used to provide training data sets and inference data. In the model training phase, the training data provided by the data source is analyzed or trained to obtain an AI model. The AI ​​model represents a mapping relationship between the model's input and output. Obtaining an AI model through learning using a model training node is equivalent to obtaining a mapping relationship between the model's input and output through learning using training data. In the model inference phase, the AI ​​model obtained through training in the model training phase is used to perform inference based on the inference data provided by the data source to obtain an inference result. This phase can also be understood as follows: inference data is input to the AI ​​model, and output through the AI ​​model is obtained, and the output is the inference result. The inference result may indicate the configuration parameters used (executed) by the actor object and / or the action performed by the actor object. The inference result is published in the inference result application phase. For example, the inference result may be planned in a unified manner by actor entities. For example, an actor entity may send inference results to one or more actor objects (e.g., network devices or terminal devices) for execution. In another example, an actor entity may further feed back model performance to a data source to facilitate subsequent model update training.

[0154] It may be understood that a communication system may include network elements having artificial intelligence capabilities. The above-described phases of AI model design may be performed by one or more network elements having artificial intelligence capabilities. In one possible design, AI functionality (e.g., an AI module or AI entity) may be configured within an existing network element in the communication system to perform AI-related operations, such as AI model training and / or inference. For example, the existing network element may be a network device or a terminal device. Alternatively, in another possible design, a separate network element may be introduced into the communication system to perform AI-related operations, such as AI model training. The separate network element may be referred to as an AI network element, an AI node, or the like. The names are not limited in the embodiments of this application. For example, the AI ​​network element may be directly connected to a network device in the communication system or indirectly connected to the network device via a third-party network element. The third-party network element may be a core network element, such as an authentication management function (AMF) network element or a user plane function (UPF) network element, an operations, administration, and maintenance (OAM) network element, a cloud server, or another network element. This is not limited to this. For example, the independent network element may be located on one or more of the network device side, the terminal device side, or the core network side. Optionally, the independent network element may be located on a cloud-side server. For example, the AI ​​network element 140 is introduced into the communication system shown in FIG. 2.

[0155] Training processes for different models may be deployed on different devices or nodes, or on the same device or node. Inference processes for different models may be deployed on different devices or nodes, or on the same device or node. For example, a model training phase is performed by a terminal device. After training a matching encoder and decoder, the terminal device may send model parameters of the decoder to a network device. For example, the model training phase is performed by a network device. After training a matching encoder and decoder, the network device may indicate the model parameters of the encoder to the terminal device. For example, the model training phase is performed by an independent AI network element. After training a matching encoder and decoder, the AI ​​network element may send the model parameters of the encoder to the terminal device and the model parameters of the decoder to the network device. Then, a model inference phase corresponding to the encoder is performed in the terminal device, and a model inference phase corresponding to the decoder is performed in the network device.

[0156] The model parameters may include one or more of the following: structural parameters of the model (e.g., the number of layers and / or model weights), input parameters of the model (e.g., input dimensions and the number of input ports), or output parameters of the model (e.g., output dimensions and the number of output ports). It may be understood that the input dimensions may be the size of a piece of input data. For example, if the input data is a sequence, the input dimensions corresponding to the sequence may indicate the length of the sequence. The number of input ports may be the number of input data. Similarly, the output dimensions may be the size of a piece of output data. For example, if the output data is a sequence, the output dimensions corresponding to the sequence may indicate the length of the sequence. The number of output ports may be the number of output data.

[0157] (6) Channel State Information (CSI):

[0158] In a communication system (e.g., an LTE communication system or an NR communication system), a network device needs to determine configurations such as resources, MCS, and precoding to be used to schedule a downlink data channel for a terminal device based on CSI. It may be understood that CSI is channel information and may reflect channel characteristics and channel quality. For example, the CSI may be represented by using a channel matrix. For example, the CSI may include a channel matrix, or the CSI may include a channel eigenvector.

[0159] The CSI measurement means that the receiving end solves the channel information based on the reference signal transmitted by the transmitting end, that is, estimates the channel information by using a channel estimation method. The propagation form of the wireless signal in the channel is expressed as Y=HX+N noise where H is the CSI, X is the reference signal, and N noisewhere X is noise, and Y is a received signal. The reference signal X is known information specified by the terminal device and the network device. After the received signal Y is obtained, channel estimation may be performed using a channel estimation algorithm, such as the least squares method or the minimum mean square error method. For example, the reference signal X may include one or more of a channel state information reference signal (CSI-RS), a synchronization signal / physical broadcast channel block (SSB), a sounding reference signal (SRS), a demodulation reference signal (DMRS), etc. The CSI-RS, SSB, DMRS, etc. may be used to measure downlink CSI. The SRS, DMRS, etc. may be used to measure uplink CSI.

[0160] An FDD communication scenario is used as an example. In the FDD communication scenario, the uplink channel and the downlink channel are not reciprocal or the reciprocity between the uplink channel and the downlink channel cannot be guaranteed. Therefore, a network device generally transmits a downlink reference signal to a terminal device, and the terminal device performs channel measurement and interference measurement based on the received downlink reference signal to estimate downlink CSI. The terminal device generates a CSI report in a manner predefined in a protocol or configured by the network device and feeds the CSI report back to the network device, thereby allowing the network device to obtain downlink CSI.

[0161] For example, the CSI may include at least one of a channel quality indication (CQI), a precoding matrix indicator (PMI), a rank indicator (RI), a CSI-RS resource indicator (CRI), a layer indicator (LI), a reference signal received power (RSRP), a signal-to-interference-plus-noise ratio (SINR), etc. The signal-to-interference-plus-noise ratio may also be referred to as a signal-to-interference-plus-noise ratio.

[0162] The RI indicates the number of layers of downlink transmission recommended by the terminal device. The CQI indicates the modulation and coding scheme that can be supported in the current channel state determined by the terminal device. The PMI indicates the precoding recommended by the terminal device. The number of layers of precoding indicated by the PMI corresponds to the RI. For example, if the RI is n, the PMI indicates n-layer precoding, where n is a positive integer.

[0163] It should be understood that the RI, CQI, PMI, etc. indicated by the CSI report are merely recommended values ​​provided by the terminal device, and the network device may perform downlink transmission based on some or all of the information indicated by the CSI report. Alternatively, the network device may perform downlink transmission without reference to the information indicated by the CSI report.

[0164] AI technology is introduced into a wireless communication network to obtain a CSI feedback mode based on an AI model. A terminal device compresses and feeds back CSI by using the AI ​​model, and a network device restores the compressed CSI by using the AI ​​model. Transmission of a sequence (e.g., a bit sequence) is performed in the AI-based CSI feedback, and overhead is lower than the CSI overhead in conventional CSI feedback.

[0165] FIG. 3 is used as an example. The encoder in FIG. 3 may be a CSI generator, and the decoder may be a CSI reconstructor. The encoder may be located on a terminal device, and the decoder may be located on a network device. The terminal device may generate CSI feedback information z based on raw CSI information V by using the encoder. The terminal device may report a CSI report, and the CSI report may include the CSI feedback information z. The network device may reconstruct the CSI information by using the decoder to obtain restored CSI information V′.

[0166] Raw CSI information V may be acquired by a terminal device through CSI measurements. For example, the raw CSI information V may include a channel response of a downlink channel or an eigenmatrix (a matrix including eigenvectors) of the downlink channel. An encoder processes the eigenmatrix of the downlink channel to obtain CSI feedback information z, i.e., channel state indication information. In other words, the operation of compressing and / or quantizing the eigenmatrix based on a codebook in the related solution is replaced with the operation of processing the eigenmatrix by the encoder to obtain CSI feedback information z. The terminal device reports the CSI feedback information z. The network device processes the CSI feedback information z by using a decoder to obtain reconstructed CSI information V'. For example, pre-coding corresponding to the reconstructed CSI information V' may be reconstructed CSI information V'. In other words, the decoder may process the CSI feedback information z and output the reconstructed CSI information V', i.e., the pre-coding. The precoding corresponding to the recovered CSI information V′ may alternatively be precoding obtained by the network device by further processing V′, e.g., the network device calculates multi-user precoding based on V′.

[0167] The following further describes examples of training and inference processes for AI models in embodiments of the present application.

[0168] The training dataset used to train an AI model includes training samples and sample labels. For example, the training samples are the channel information determined by the terminal device, and the sample labels are the real-world channel information. If the encoder and decoder belong to the same autoencoder, the training data may only contain training samples. In other words, the training samples are the sample labels, i.e., the channel information.

[0169] The specific training process is as follows: The model training node processes channel information, i.e., training samples, using an encoder to obtain channel state indication information, i.e., CSI feedback information, and processes the channel state indication information using a decoder to obtain restored channel information, i.e., restored CSI information. Then, the difference between the restored CSI information and the corresponding sample labels, i.e., the value of the loss function, is calculated, and the parameters of the encoder and decoder are updated based on the value of the loss function to minimize the difference between the restored channel information and the corresponding sample labels, i.e., minimize the loss function. For example, the loss function may be minimum mean square error (MSE) or cosine similarity. The above operations are repeated to obtain an encoder and decoder that meet the target requirements. The model training node may be a terminal device, a network device, or another network element with AI capabilities in a communication system.

[0170] It should be understood that the use of the AI ​​model for CSI compression is used above as an example for illustrative purposes only. In CSI feedback, the AI ​​model may also be used in other scenarios. For example, the AI ​​model may be used for CSI prediction, i.e., channel information at one or more future time points is predicted based on channel information measured at one or more past time points. The specific purpose of the AI ​​model in the CSI feedback scenario is not limited in the embodiments of the present application.

[0171] In the codebook-based CSI feedback mode, for some codebooks with high overhead, such as release (R)15 type II, R16 type II, and R17 type II codebooks, the report content of the CSI report may be divided into two parts: part 1 and part 2. The first part may also be referred to as part 1, and the second part may also be referred to as part 2. The CQI, RI, etc. belong to the first part, and the PMI, etc. belong to the second part. Transmission of the second part is generally performed over a physical uplink shared channel (PUSCH). Because the size of the second part is not fixed and multiple CSI report transmissions may need to be performed on the same resource, the size of the second part to be transmitted may exceed the channel's carrying limit. In a related solution, when the amount of coded modulation symbols (or modulation symbols) of the second portion to be transmitted exceeds a specified threshold, the terminal device discards lower priority portion contents based on the priority of the report contents until the amount of coded modulation symbols of the second portion to be transmitted no longer exceeds the specified threshold. The number of coded modulation symbols of the second portion to be transmitted and the specified threshold can be obtained through calculation by using formulas and parameters defined in the protocol.

[0172] With the development of artificial intelligence (AI) technology, CSI feedback modes based on AI models are emerging. Correspondingly, multiple AI-related CSI report types are involved in the CSI feedback process. AI-related CSI reports and codebook-based CSI reports have different content, and the aforementioned solutions are no longer applicable to AI model-based CSI feedback modes.

[0173] In view of this, the present application provides a communication method and a communication apparatus. The priority of a CSI report, including an AI-related CSI report and an AI-non-related CSI report, is determined based on the type of the CSI report. If the size of a CSI report to be transmitted exceeds a specified threshold, the terminal device can discard partial content of lower priority based on the priority, thereby allowing the network device to obtain high-priority channel information and facilitating subsequent data transmission between the network device and the terminal device. The communication method may be applied to the aforementioned communication system, for example, an FDD communication scenario. Additionally, optionally, the communication method may also be applied to a TDD communication scenario. This is not a limitation of the present disclosure.

[0174] In this application, it should be understood that instruction includes direct instruction (also called explicit instruction) and implicit instruction. Directly indicating information A means including information A. Implicitly indicating information A means indicating information A by directly indicating information B based on the correspondence between information A and information B. The correspondence between information A and information B may be predefined, pre-stored, pre-baked, or pre-configured.

[0175] It should be understood that, in this application, the use of information C to determine information D includes information D being determined based only on information C, and also information D being determined based on information C and other information. In addition, the use of information C to determine information D may further include cases of indirect determination. For example, information D is determined based on information E, and information E is determined based on information C.

[0176] 5 is a schematic flowchart of a communication method according to the present application. As shown in FIG. 5, the method 500 may include the following steps:

[0177] 510: Obtain p groups to be transmitted on a first resource, where the p groups belong to N CSI reports, where p is an integer greater than 1, and N is a positive integer. A type of at least one report among the N CSI reports is an AI-related type.

[0178] 520: Transmit some or all of the p groups on the first resource based on priorities of the p groups, where the priorities of the p groups are based on priorities of the N CSI reports and / or priorities of the p groups in the N CSI reports.

[0179] The first resource may be a resource that can carry a CSI report, for example, a PUSCH resource.

[0180] p groups belonging to N CSI reports means that each CSI report includes at least one of the p groups, and each group belongs to only one CSI report.

[0181] For example, in this embodiment of the present application, the priorities of p groups may be understood as the priorities of p groups in p groups, the priorities of N CSI reports may be understood as the priorities of N CSI reports in N CSI reports, and the priorities of p groups in N CSI reports may be understood as the priorities of p groups in each CSI report to which the p groups belong.

[0182] In one possible implementation, N = 1. The priorities of the p groups are the priorities of the p groups in the CSI report.

[0183] When N=1, p groups belong to the same CSI report. The priority of a group is the priority of that group in the CSI report.

[0184] In one possible implementation, N is an integer greater than 1. The priority of the p groups may be based on the priority of the N CSI reports, which is based on the type of the CSI report.

[0185] In other words, the priority of a group is based on the priority of the CSI report to which it belongs.

[0186] For example, the priority of a CSI report depends on whether the CSI report is an AI-related CSI report. An AI-related CSI report is a CSI report whose type is an AI-related type.

[0187] The number of groups included in different CSI reports may be the same or different.

[0188] For example, each CSI report may include one of p groups, ie, p=N, and the priority of each group may be the priority of the CSI report to which it belongs.

[0189] In one possible implementation, N is an integer greater than 1. The priorities of the p groups may be based on the priorities of the N CSI reports and the priorities of the p groups in each CSI report to which the p groups belong. The priorities of the CSI reports are based on the type of the CSI report.

[0190] In other words, the priority of a group among the p groups may be based on the priority of the CSI report to which the group belongs and the priority of the group in the CSI report.

[0191] For example, the priority of a CSI report depends on whether the CSI report is an AI-related CSI report.

[0192] Optionally, step 520 may include selecting a subset of the p groups based on priorities of the p groups and transmitting the subset of groups on the first resource.

[0193] In some scenarios, the first resource may fail to carry p groups, in which case one or more groups from the p groups may be discarded based on the priorities of the p groups, so that the first resource can carry some of the remaining groups.

[0194] For example, the subset of groups may be subsets of groups with the highest priority among the p groups. In this manner, the network device can obtain channel information within a higher priority group. For example, the priority may be related to the importance of the information within the group. In this manner, the network device can obtain important information or information within an important group.

[0195] Optionally, step 520 may include determining locations of some or all of the p groups on a first resource based on priorities of the p groups, and transmitting some or all of the p groups on the first resource.

[0196] For example, some or all of the p groups are sorted based on the priorities of the p groups, and the sorting of some or all of the groups is consistent with the placement order of some or all of the groups on the first resource.

[0197] It should be understood that the above is merely an example and does not constitute any limitation to the solution in the embodiment of the present application. For example, suitable locations on the first resource may be further selected for the p groups based on the priorities of the p groups.

[0198] According to the solution in this embodiment of the present application, CSI reports to be transmitted on the same resource include AI-related CSI reports, and some or all of the p groups are transmitted based on the priorities of the p groups to be transmitted on the same resource. This helps network devices appropriately use resources to obtain desired information. For example, if the same resource cannot carry all of the p groups, some groups may be selected for transmission based on the priorities of the p groups, with higher-priority groups being transmitted and lower-priority groups being discarded. This allows network devices to obtain channel information within the higher-priority groups, e.g., important groups. For example, the priorities of the p groups may be determined depending on whether the CSI report is AI-related, and then the higher-priority groups are transmitted. In another example, AI-related CSI reports may also be divided into multiple groups. The priorities of the p groups may be determined based on the priorities of the groups in the CSI report, and then the higher-priority groups are transmitted. In this way, the flexibility of information transmission can be improved, and even if it is not possible to transmit the entire AI-related CSI report, the probability that the network device will obtain a partial group of information in the CSI report can be increased, thereby allowing the network device to obtain partial information in the CSI report.

[0199] An AI-related CSI report may be understood as a CSI report related to an AI model. The AI ​​model is used to generate channel information that is fed back to network devices. In other words, the AI ​​model generates some or all of the content of the CSI report, or the content generated by the AI ​​model is processed and used as some or all of the content of the CSI report.

[0200] For example, the AI ​​model may be a two-sided model. A sub-model within the two-sided model and deployed on the terminal device may be used to generate channel information to be fed back to the network device. The sub-model deployed on the network device may be used to reconstruct the channel information. In other words, in the two-sided model, the channel information to be fed back to the network device may be generated by a sub-model of the AI ​​model.

[0201] For example, the AI ​​model may include a CSI generator and a CSI reconstructor. The CSI generator may be located on a terminal device, and the CSI reconstructor may be located on a network device. The terminal device may generate CSI feedback information, i.e., channel information to be fed back to the network device, via the CSI generator and feed back the CSI feedback information to the network device by using a CSI report. The network device may reconstruct the CSI information via the CSI reconstructor to obtain restored CSI information.

[0202] For a detailed description of the CSI generator and the CSI reconstructor, please refer to the above description, and to avoid repetition, the details will not be described again here.

[0203] For example, the AI ​​model may be a one-sided model that may be deployed on a terminal device and used to generate channel information to be fed back to a network device.

[0204] For example, the AI ​​model may be used to predict channel information at one or more future points in time, i.e., channel information to be fed back to the network device, and the channel information may be fed back to the network device by using CSI reports.

[0205] It should be understood that the above is merely an example, and the AI ​​model may alternatively be another type of model, which is not limited in the embodiment of the present application. For ease of explanation, in this embodiment of the present application, an example in which the AI ​​model includes a CSI generator and a CSI reconstructor is mainly used for explanation, and does not constitute any limitation on the solution in the embodiment of the present application.

[0206] In the traditional CSI feedback mode, some channel features have high overhead and some have low overhead. For example, by using an AI model, processing such as compression and / or quantization may be performed on channel features with higher overhead, and the processing results are fed back to the network device by using CSI reports. For channel features with lower overhead, the traditional feedback mode may still be used. For example, feedback is provided to the network device in a codebook-based feedback mode.

[0207] The codebook-based CSI feedback mode is used as an example. The overhead of the CSI feedback amount in the CSI report is mainly the overhead of the PMI. The PMI is obtained by compressing and / or quantizing the eigenmatrix of the channel based on a codebook. For example, the CSI feedback information can be generated through a CSI generator. The function of the CSI feedback information is similar to that of the PMI. The CSI feedback information may be referred to as AI-PMI. For example, the CSI feedback information may be obtained by processing the eigenmatrix of the channel by the CSI generator. In other words, the operation of compressing and / or quantizing the eigenmatrix based on a codebook is replaced by the operation of processing the eigenmatrix by the CSI generator to obtain the CSI feedback information. The report content of the CSI report may indicate the CSI feedback information. The network device reconstructs the restored CSI information through a CSI reconstructor to determine the precoding. The report content of the CSI report may also indicate the CSI feedback amount in the conventional feedback mode, such as RI or CQI.

[0208] It should be understood that the above is merely an example, and the CSI feedback information generated by the AI ​​model may further indicate other contents, which is not limited in the embodiments of the present application.

[0209] For example, the AI-related types may be classified into multiple types based on the report content of the CSI report.

[0210] Optionally, the AI-related type may include one or more of a first CSI report type, a second CSI report type, or a third CSI report type, where the report content of the first CSI report type includes channel information obtained through inference by an AI model, the report content of the second CSI report type includes performance information of the AI ​​model, or the report content of the third CSI report type includes channel information for performing an operation related to the AI ​​model.

[0211] A CSI report corresponding to the first CSI report type may be referred to as a CSI report based on AI model inference, a CSI report corresponding to the second CSI report type may be referred to as a CSI report for performance monitoring, and a CSI report corresponding to the third CSI report type may be referred to as a CSI report for data collection.

[0212] The following will illustrate the three types mentioned above by using examples.

[0213] A CSI report based on AI model inference is a CSI report whose report content includes information about the output of an AI model. For example, the report content may include the output of an AI model, or the report content includes information obtained from the output of an AI model. For example, the output of an AI model may be channel information obtained through inference by the AI ​​model. For example, the report content of the CSI report may indicate CSI feedback information generated by a CSI generator. Furthermore, the report content of the CSI report may further indicate another CSI feedback quantity, such as RI or CQI.

[0214] The CSI report for data collection includes channel information for implementing operations related to the AI ​​model in its report content. Furthermore, the report content of the CSI report may further indicate another CSI feedback quantity, such as RI or CQI.

[0215] For example, operations related to an AI model may include one or more of the following: AI model training, AI model update, AI model selection, AI model switching, AI model activation, AI model deactivation, AI model performance monitoring, or AI model testing.

[0216] The channel information may be high-precision CSI. High-precision CSI means that the CSI is not compressed and / or quantized, or that the high-precision CSI is compressed and / or quantized to a lower degree than the PMI or AI-PMI. In other words, more CSI is reserved in the high-precision CSI, or the high-precision CSI has higher accuracy.

[0217] For example, high-precision CSI may be fed back in a codebook-based mode, where the raw CSI measured by the terminal device is processed based on an R16 type II codebook, and the processed result is fed back to the network device by using a CSI report.

[0218] For example, high-precision CSI can be fed back in a codebook-based mode, but more raw CSI is reserved compared to the codebook in existing protocols.

[0219] In other words, the raw CSI measured by the terminal device is compressed and / or quantized by using a codebook, but more raw CSI is reserved compared to the codebook in existing protocols.

[0220] For example, high-precision CSI may be fed back in a codebook-based mode, but new parameters may be introduced so that the CSI obtained by the network devices has higher precision.

[0221] A codebook-based feedback mode may be understood as selecting one or more bases from a predetermined codebook for weighting to represent the raw CSI. In general, feedback overhead may be controlled by using the number of selected bases, the number of non-zero weighting coefficients, and the quantization precision of the weighting coefficients. A new parameter is introduced so that the accuracy of the CSI obtained by the network device is higher. It may be understood that the network device can obtain the CSI with higher accuracy by using one or more of a larger number of bases, a larger number of non-zero weighting coefficients, or a higher quantization precision of the weighting coefficients.

[0222] For example, high-precision CSI may be fed back in a scalar quantization mode. The scalar quantization may include W-bit scalar quantization, float16 quantization, float32 quantization, etc. For example, raw CSI measured by a terminal device is quantized in a float16 quantization mode to obtain data CSI, which is fed back to a network device by using a CSI report. W may be a positive integer. For example, W may be any integer from 1 to 8.

[0223] For ease of explanation, the channel information is referred to as data CSI in this embodiment of the present application. The report content of a CSI report for data collection may include data CSI. The data CSI may be used to implement operations related to an AI model.

[0224] For example, the data CSI may be used as training data in the training process of an AI model. For example, the data CSI may be used as training samples input to an AI model and / or as a target output of the AI ​​model. A target output may also be referred to as a label, sample label, truth value (correct answer), target, etc. For example, the data CSI may be used as a target output of the model in the performance monitoring process of the AI ​​model. For example, the data CSI may be used as a target output of the model in the testing process of the AI ​​model.

[0225] For example, the data CSI may be a channel response or an eigenmatrix of the channel.

[0226] The channel response is sometimes called the channel matrix.

[0227] There may be one or more eigenmatrices for the channel.

[0228] A CSI report for performance monitoring is a CSI report whose report content includes performance information of an AI model.

[0229] For example, the performance information may include the accuracy of the output of the artificial intelligence model.

[0230] For example, the AI ​​model includes a CSI generation model and a CSI reconstruction model. The accuracy of the output result of the AI ​​model may be understood as the accuracy of the restored CSI information. For example, the accuracy of the restored CSI information may be reflected by the similarity or error between the raw CSI information and the restored CSI information, such as the cosine similarity between the raw CSI information and the restored CSI information or the mean square error between the raw CSI information and the restored CSI information. The cosine similarity may be generalized cosine similarity (GCS), square generalized cosine similarity (SGCS), etc. The mean square error may be minimum mean square error (MSE), normalized mean square error (NMSE), etc. It should be understood that the above is merely an example, and the accuracy of the restored CSI information may also be reflected by the similarity or error between the restored CSI information and channel information obtained in another manner. For example, the accuracy of the recovered CSI information may be reflected by the similarity or error between the recovered CSI information and the channel information obtained by the network device in the codebook-based feedback mode.

[0231] For example, the performance information may include an AI model's representation of system performance, and may include, for example, at least one of throughput, acknowledgement (ACK) ratio or negative acknowledgment (NACK) ratio, block error rate (BLER), or hypothetical BLER.

[0232] For example, the performance information can include whether the AI ​​model meets pre-set performance requirements.

[0233] For example, the performance requirement may be that the GCS between the raw CSI information and the recovered CSI information is greater than a specified threshold.

[0234] The performance requirements may be predefined, or alternatively, may be configured by the base station, or alternatively, may be determined by the terminal device.

[0235] An AI-related CSI report may alternatively be combined with another CSI report into one CSI report.

[0236] A CSI report for performance monitoring is used as an example. A CSI report for performance monitoring is generally small and may be combined with another CSI report into one CSI report.

[0237] For example, a CSI report for performance monitoring and a CSI report based on AI model inference may be combined into one CSI report. In other words, the report content of the CSI report may indicate channel information obtained through inference by the AI ​​model and performance information of the AI ​​model.

[0238] For example, the AI ​​model includes a CSI generator and a CSI reconstructor. The terminal device may perform channel measurement based on reference signal #1 transmitted by the network device to obtain raw CSI information #1. The terminal device processes the raw CSI information #1 based on the CSI generator to generate CSI feedback information #1 and feeds the CSI feedback information #1 back to the network device using a CSI report. The network device reconstructs the CSI feedback information #1 through the CSI reconstructor to obtain restored CSI information #1. The network device may transmit the restored CSI information #1 to the terminal device, and the terminal device may evaluate the accuracy of the restored CSI information #1 based on the restored CSI information #1 and the raw CSI information #1. The terminal device may perform channel measurement based on reference signal #2 transmitted by the network device to obtain raw CSI information #2. The terminal device processes the raw CSI information #2 based on the CSI generator to generate CSI feedback information #2. When the terminal device provides CSI feedback, the CSI report may include CSI feedback information #2 and the accuracy of the recovered CSI information #1. It should be understood that the above is merely an example and does not constitute any limitation on the solution in the embodiment of the present application.

[0239] For example, the AI-related type may be categorized into multiple types based on the AI ​​model associated with the CSI report.

[0240] For example, when multiple CSI reports are associated with different AI models, the types of the multiple CSI reports may be considered to be different types.

[0241] The above is just an example. The AI-related CSI report may further include other types. This is not limited to the embodiments of the present application.

[0242] For example, an AI-related CSI report may have a higher priority than another CSI report, or alternatively, an AI-related CSI report may have a lower priority than another CSI report.

[0243] Optionally, the priority of a CSI report may alternatively be based on at least one of the following: the periodicity of the CSI report, the channel on which the CSI report is carried, whether the CSI report includes Layer 1 (L1)-RSRP or L1-SINR, the serving cell to which the CSI report belongs, or the ID of the CSI report.

[0244] In one possible implementation, this embodiment may be specific only to N CSI reports carried on the same channel, i.e., p groups belong to N CSI reports carried on the same channel. When CSI reports carried on different channels collide, for example, when a terminal device needs to simultaneously transmit multiple CSI reports and the multiple CSI reports are carried on different channels, processing may be performed in a different manner. For example, only one or more higher priority CSI reports may be transmitted, and the higher priority CSI reports may be one or more CSI reports carried on the same channel.

[0245] In one possible implementation, this embodiment may be specific only to N CSI reports carried on the same channel and having the same periodicity (i.e., the CSI reports are all periodic CSI reports, all semi-persistent CSI reports, or all aperiodic CSI reports), i.e., p groups belong to N CSI reports carried on the same channel and having the same periodicity. When CSI reports carried on different channels or CSI reports with different periodicities collide, for example, when a terminal device needs to simultaneously transmit multiple CSI reports and the multiple CSI reports are carried on different channels or have different periodicities, processing is performed in a different manner. For example, only one or more higher-priority CSI reports may be sent, and the higher-priority CSI reports may be one or more CSI reports carried on the same channel with the same periodicity.

[0246] In the following, we use embodiment 1 and embodiment 2 as examples to explain the priority of CSI reports.

[0247] Aspect 1:

[0248] In one possible implementation, the priority of the CSI report is determined based on one or more parameters, at least one of which indicates a type of the CSI report and the at least one parameter further indicates other information of the CSI report.

[0249] In aspect 1, all or part of the parameters indicating the type of the CSI report further indicate other information of the CSI report.

[0250] Optionally, the priority of the CSI report is determined based on one or more parameters, where the one or more parameters indicate one or more of the following: a periodicity of the CSI report, a channel on which the CSI report is carried, whether the CSI report includes L1-RSRP or L1-SINR, a serving cell to which the CSI report belongs, or an ID of the CSI report. At least one of the at least one or more parameters further indicates a type of the CSI report.

[0251] The types of CSI reports may include an AI-related type and an AI-unrelated type. In this embodiment of the present application, a CSI report corresponding to an AI-unrelated type may also be referred to as an AI-unrelated CSI report.

[0252] That is, the type of the CSI report may be used to distinguish whether the CSI report is an AI-related CSI report, or whether the CSI report is an AI-related CSI report or an AI-non-related CSI report.

[0253] For example, the type of CSI report may include two types: an AI-related type and an AI-unrelated type.

[0254] For example, the AI-related type may further include a first CSI report type, a second CSI report type, or a third CSI report type, in which case the CSI report type may include an AI-non-related type, the first CSI report type, the second CSI report type, or a third CSI report type.

[0255] For a given parameter, different values ​​of the parameter may indicate different information.

[0256] For example, the CSI report priority pri CSI may satisfy the following equation: priCSI (y,k,c,s)=N k N cells M s y+N cells M s k+M s c+s

[0257] pri CSI A smaller value of indicates a higher priority. y is used to distinguish the periodicity of the CSI report and the channel on which the CSI report is carried. For example, y = 0 indicates an aperiodic CSI report carried on the PUSCH, y = 1 indicates a semi-persistent CSI report carried on the PUSCH, y = 2 indicates a semi-persistent CSI report carried on the PUCCH, and y = 3 indicates a periodic CSI report carried on the PUCCH. k is used to distinguish whether the CSI report includes L1-RSRP or L1-SINR. For example, k = 0 indicates a CSI report that includes L1-RSRP or L1-SINR, and k = 1 indicates a CSI report that does not include L1-RSRP or L1-SINR. c is used to distinguish the serving cell to which the CSI report belongs. For example, c may be the index of the serving cell. s is used to distinguish the ID of the CSI report. For example, s may be the ID of the CSI report. N k denotes the number of all possible values ​​of k. For example, if k has a value of 0 or 1, then N k = 2. N cells denotes the maximum number of serving cells, i.e., the number of all possible values ​​of c. s denotes the maximum number of CSI reports, i.e., the number of all possible values ​​of s.

[0258] For example, values ​​for some or all of the above parameters may be added to indicate the type of CSI report. For example, values ​​for one or more parameters in y, k, c, and s may be added to distinguish between AI-related and non-AI-related CSI reports.

[0259] The at least one parameter indicating a type of CSI report may be a single parameter.

[0260] For example, a parameter indicating a type of CSI report may be s. There is a correspondence between the ID of the CSI report and the type of the CSI report. For example, the network device may configure different IDs for AI-related CSI reports and AI-non-related CSI reports. The ID of the AI-related CSI report may be greater than the ID of the AI-non-related CSI report, or the ID of the AI-related CSI report may be less than the ID of the AI-non-related CSI report.

[0261] In this case, no distinction is made as to whether the CSI report is an AI-related CSI report or not.

[0262] For example, the parameter indicating the type of CSI report may be y. The value of y is added to distinguish between AI-related and non-AI-related CSI reports. In other words, y is used to distinguish the periodicity of the CSI report, the channel on which the CSI report is carried, and the type of the CSI report.

[0263] For example, y=0 indicates an aperiodic CSI report not associated with an AI carried on the PUSCH, y=1 indicates a semi-persistent CSI report not associated with an AI carried on the PUSCH, y=2 indicates a semi-persistent CSI report not associated with an AI carried on the PUCCH, y=3 indicates a periodic CSI report not associated with an AI carried on the PUCCH, y=4 indicates a CSI report for performance monitoring, y=5 indicates a CSI report based on AI model inference, and y=6 indicates a CSI report for data collection.

[0264] It should be understood that the above-described values ​​of y and their corresponding meanings are merely examples. In another implementation, there may be another correspondence between the value of y and the meaning of that value. Alternatively, in another implementation, the correspondence between the value of y and the meaning of that value may be a part of the above-described correspondence. For example, the meanings of y being 0, 1, 2, and 3 are as described above, and y=4 indicates an AI-related CSI report. In this case, the specific type of AI-related CSI report is not distinguished. In another example, the meanings of y being 0, 1, 2, and 3 are as described above, and y=4 indicates a CSI report for performance monitoring and a CSI report based on AI model inference, and y=5 indicates a CSI report for data collection. In this case, the CSI report for performance monitoring and the CSI report based on AI model inference may be the same CSI report.

[0265] For example, the parameter indicating the type of CSI report may be k. The value of k is added to distinguish between AI-related and non-AI-related CSI reports. That is, k is used to distinguish the type of CSI report and whether the CSI report includes L1-RSRP or L1-SINR.

[0266] For example, k=0 indicates an AI-non-associated CSI report including L1-RSRP or L1-SINR, k=1 indicates an AI-non-associated CSI report without L1-RSRP or L1-SINR, k=2 indicates a CSI report for performance monitoring, k=3 indicates a CSI report based on AI model inference, and k=4 indicates a CSI report for data collection.

[0267] It should be understood that the above values ​​of k and their corresponding meanings are merely examples. In another implementation, there may be a different correspondence between the values ​​of k and their corresponding meanings. Alternatively, in another implementation, the correspondence between the values ​​of k and their corresponding meanings may be a subset of the above correspondences.

[0268] The at least one parameter indicating the type of CSI report may be a plurality of parameters.

[0269] For example, the parameters indicating the type of the CSI report may be y and k. The values ​​of y and k are added to distinguish between an AI-related CSI report and an AI-non-related CSI report. Specifically, y is used to distinguish whether the CSI report is an AI-related CSI report, the channel on which the CSI report is carried, and the type of the CSI report, and k is used to distinguish the type of the AI-related CSI report and whether the CSI report includes L1-RSRP or L1-SINR.

[0270] For example, y=0 indicates an aperiodic CSI report not associated with an AI carried on the PUSCH, y=1 indicates a semi-persistent CSI report not associated with an AI carried on the PUSCH, y=2 indicates a semi-persistent CSI report not associated with an AI carried on the PUCCH, y=3 indicates a periodic CSI report not associated with an AI carried on the PUCCH, and y=4 indicates a CSI report associated with an AI. k=0 indicates that the CSI report includes performance monitoring information to reflect L1-RSRP, L1-SINR, or an AI model, k=1 indicates that the CSI report does not include performance monitoring information to reflect L1-RSRP, L1-SINR, or an AI model and is a CSI report based on AI model inference, and k=2 indicates that the CSI report does not include performance monitoring information to reflect L1-RSRP, L1-SINR, or an AI model and is a CSI report for data collection. A CSI report that includes performance monitoring information to reflect an AI model is a CSI report for performance monitoring.

[0271] It should be understood that the above values ​​of y and k and the corresponding meanings of those values ​​are merely examples. In another implementation, there may be other correspondences between the values ​​of y and k and the meanings of those values. Alternatively, in another implementation, the correspondences between the values ​​of y and k and the meanings of those values ​​may be a subset of the above correspondences.

[0272] The above formulas and related parameters are merely examples and do not constitute limitations on the solutions in the embodiments of the present application.

[0273] Aspect 2:

[0274] Optionally, the priority of the CSI report is determined based on one or more parameters, at least one of which indicates a type of the CSI report, and the at least one parameter does not indicate other information of the CSI report.

[0275] For a related description of the types of CSI reports, please refer to aspect 1. The details will not be described again herein.

[0276] In aspect 2, the at least one parameter indicates only the type of CSI report and does not indicate other information.

[0277] Optionally, another parameter of the one or more parameters indicates at least one of the following: a periodicity of the CSI report, a channel on which the CSI report is carried, whether the CSI report includes L1-RSRP or L1-SINR, a serving cell to which the CSI report belongs, or an ID of the CSI report.

[0278] For example, based on four parameters y, k, c, and s, at least one parameter may be added to indicate the type of CSI report. For example, the at least one parameter may be l. The added parameter l is used to distinguish between AI-related CSI reports and AI-non-related CSI reports. l may be located at the most significant bit, the least significant bit, or any position among y, k, c, and s. The parameter at the most significant bit has the largest coefficient, and the parameter at the least significant bit has the smallest coefficient. The formula in aspect 1 is used as an example. The parameters from the most significant bit to the least significant bit are y, k, c, and s, respectively.

[0279] For example, l may be located in the most significant bit, and the priority pri of the CSI report CSI may satisfy the following formula: pri CSI (l,y,k,c,s)=N y N k N cells M s l+N k N cells M s y+N cells M s k+M s c+s

[0280] pri CSI A smaller value of indicates a higher priority.

[0281] l is used to distinguish the type of CSI report. For example, l=0 indicates a CSI report not related to AI, l=1 indicates a CSI report for performance monitoring, l=2 indicates a CSI report based on AI model inference, and l=3 indicates a CSI report for data collection. y denotes the quantity of all possible values ​​of y. For example, if y=0, 1, 2, or 3, then N y= 4. y is used to distinguish the periodicity of the CSI report and the channel on which the CSI report is carried. For example, y = 0 indicates an aperiodic CSI report carried on the PUSCH, y = 1 indicates a semi-persistent CSI report carried on the PUSCH, y = 2 indicates a semi-persistent CSI report carried on the PUCCH, and y = 3 indicates a periodic CSI report carried on the PUCCH. k is used to distinguish whether the CSI report includes L1-RSRP or L1-SINR. For example, k = 0 indicates a CSI report that includes L1-RSRP or L1-SINR, and k = 1 indicates a CSI report that does not include L1-RSRP or L1-SINR. c is used to distinguish the serving cell to which the CSI report belongs. s is used to distinguish the ID of the CSI report. N k denotes the quantity of all possible values ​​of k. cells denotes the maximum number of serving cells. M s indicates the maximum amount of CSI reports.

[0282] The above values ​​of l and their corresponding meanings are merely examples. In another implementation, there may be a different correspondence between the values ​​of l and their corresponding meanings. Alternatively, in another implementation, the correspondence between the values ​​of l and their corresponding meanings may be a subset of the correspondences described above.

[0283] For example, l may be located between the position y and the position k, and the priority pri of the CSI report CSI may satisfy the following formula: pri CSI (l,y,k,c,s)=N l N k N cells M s y+N k N cells M s l+N cells M s k+M s c+s

[0284] pri CSIA smaller value indicates a higher priority.

[0285] l is used to distinguish the type of CSI report. For example, l=0 indicates a CSI report not related to AI, l=1 indicates a CSI report for performance monitoring, l=2 indicates a CSI report based on AI model inference, and l=3 indicates a CSI report for data collection. l denotes the number of all possible values ​​of l. For example, l=0, 1, 2, or 3, and N l = 4. The meanings of the other parameters are the same as those described above, and the details will not be explained again here.

[0286] The above values ​​of l and their corresponding meanings are merely examples. In another implementation, there may be another correspondence between the values ​​of l and their meanings. Alternatively, in another implementation, the correspondence between the values ​​of l and their meanings may be a part of the above correspondence. In another implementation, multiple parameters may be added to indicate the type of CSI report. This is not limited to the embodiments of the present application.

[0287] The above formulas and related parameters are merely examples and do not constitute limitations on the solutions in the embodiments of the present application.

[0288] In addition, the above-described aspects 1 and 2 may alternatively be used in combination. The at least one parameter indicating the type of CSI report may include multiple parameters, some of which indicate only the type of CSI report, and other parameters which indicate the type of CSI report and other information.

[0289] For example, some parameters are used to distinguish some AI-related CSI reports from non-AI-related reports, and other parameters are used to distinguish other AI-related CSI reports from non-AI-related CSI reports, and indicate other information.

[0290] For example, the CSI report priority pri CSI may satisfy the following formula: pri CSI (l,y,k,c,s)=N l N k N cells M s y+N k N cells M s l+N cells M s k+M s c+s

[0291] pri CSI A smaller value indicates a higher priority.

[0292] l is used to distinguish the type of CSI report. For example, l=0 indicates a CSI report not related to AI, and l=1 indicates an AI-related CSI report. k is used to distinguish the type of CSI report and whether the CSI report includes L1-RSRP or L1-SINR. For example, k=0 indicates that the CSI report includes performance monitoring information to reflect L1-RSRP, L1-SINR, or an AI model. k=1 indicates that the CSI report does not include performance monitoring information to reflect L1-RSRP, L1-SINR, or an AI model and is a CSI report based on AI model inference. k=2 indicates that the CSI report does not include performance monitoring information to reflect L1-RSRP, L1-SINR, or an AI model and is a CSI report for data collection. l denotes the quantity for all possible values ​​of l. For example, if l=0 or 1, then N l= 2. For example, y = 0 indicates an aperiodic CSI report carried on the PUSCH, y = 1 indicates a semi-persistent CSI report carried on the PUSCH, y = 2 indicates a semi-persistent CSI report carried on the PUCCH, and y = 3 indicates a periodic CSI report carried on the PUCCH. c is used to distinguish the serving cell to which the CSI report belongs. s is used to distinguish the ID of the CSI report. N k denotes the number of all possible values ​​of k. cells denotes the maximum number of serving cells. M s indicates the maximum number of CSI reports.

[0293] Optionally, the priority of the CSI report may be based on an AI model associated with the CSI report.

[0294] In the following, aspects 3 and 4 are used as examples to explain the priority of CSI reports.

[0295] Aspect 3:

[0296] In one possible implementation, the priority of the CSI report is determined based on one or more parameters, at least one of which indicates an AI model associated with the CSI report, and the at least one parameter further indicates other information of the CSI report.

[0297] In aspect 3, all or part of the parameters indicating the AI ​​model associated with the CSI report further indicate other information of the CSI report.

[0298] Optionally, the priority of the CSI report is determined based on one or more parameters, where the one or more parameters indicate one or more of the following: a periodicity of the CSI report, a channel on which the CSI report is carried, whether the CSI report includes L1-RSRP or L1-SINR, a serving cell to which the CSI report belongs, or an ID of the CSI report. At least one of the at least one or more parameters further indicates an identifier of an AI model associated with the CSI report.

[0299] The priority calculation formula in aspect 1 is used as an example, and the priority of a CSI report is determined based on four parameters y, k, c, and s. In aspect 3, some or all of the values ​​of the parameters y, k, c, and s may be added to distinguish the identifier of the AI ​​model associated with the CSI report.

[0300] The at least one parameter indicating an identifier of an AI model associated with the CSI report may be a single parameter.

[0301] For example, a parameter indicating an identifier of an AI model associated with a CSI report may be y. The value of y is added to distinguish the identifier of an AI model associated with a CSI report. In other words, y is used to distinguish the periodicity of the CSI report, the channel on which the CSI report is carried, and the identifier of the AI ​​model associated with the CSI report.

[0302] For example, the CSI report priority pri CSI may satisfy the following formula: pri CSI (y,k,c,s)=N k N cells M s y+N cells M s k+M s c+s

[0303] priCSI A smaller value of indicates a higher priority. For example, y=0 indicates an aperiodic CSI report not associated with an AI carried on the PUSCH, y=1 indicates a semi-persistent CSI report not associated with an AI carried on the PUSCH, y=2 indicates a semi-persistent CSI report not associated with an AI carried on the PUCCH, y=3 indicates a periodic CSI report not associated with an AI carried on the PUCCH, y=4 indicates that the AI ​​model identifier associated with the CSI report is 0, y=5 indicates that the AI ​​model identifier associated with the CSI report is 1, etc. k is used to distinguish whether the CSI report includes L1-RSRP or L1-SINR. For example, k=0 indicates a CSI report including L1-RSRP or L1-SINR, and k=1 indicates a CSI report not including L1-RSRP or L1-SINR. c is used to distinguish the serving cell to which the CSI report belongs. For example, c may be the index of the serving cell. s is used to distinguish the ID of the CSI report. For example, s may be the ID of the CSI report. N k denotes the number of all possible values ​​of k. For example, if k has a value of 0 or 1, then N k = 2. N cells denotes the maximum number of serving cells, i.e., the number of all possible values ​​of c. s denotes the maximum number of CSI reports, i.e., the number of all possible values ​​of s.

[0304] It should be understood that the above values ​​of y and their corresponding meanings are merely examples. In another implementation, there may be another correspondence between the values ​​of y and the meanings of those values. Alternatively, in another implementation, the correspondence between the values ​​of y and the meanings of those values ​​may be a subset of the above correspondences.

[0305] Aspect 4:

[0306] Optionally, the priority of the CSI report is determined based on one or more parameters, at least one of which indicates an AI model associated with the CSI report, and the at least one parameter does not indicate other information of the CSI report.

[0307] In aspect 4, the at least one parameter indicates only the AI ​​model associated with the CSI report and no other information.

[0308] Optionally, another parameter of the one or more parameters indicates at least one of the following: a periodicity of the CSI report, a channel on which the CSI report is carried, whether the CSI report includes L1-RSRP or L1-SINR, a serving cell to which the CSI report belongs, or an ID of the CSI report.

[0309] For example, at least one parameter may be added to indicate an AI model associated with a CSI report based on four parameters y, k, c, and s. For example, the at least one parameter may be m. The added parameter m is used to distinguish the identifier of the AI ​​model associated with the CSI report. m may be located at the most significant bit, the least significant bit, or any position between y, k, c, and s. The coefficient of the parameter in the most significant bit is the largest, and the coefficient of the parameter in the least significant bit is the smallest. The formula of aspect 1 is used as an example. The parameters from the most significant bit to the least significant bit are y, k, c, and s, respectively.

[0310] For example, m may be located in the most significant bit, and the priority pri of the CSI report CSI may satisfy the following formula: pri CSI (m,y,k,c,s)=N y N k N cells M s m+N k N cellsM s y+N cells M s k+M s c+s

[0311] pri CSI A smaller value of indicates a higher priority.

[0312] m is used to distinguish the identifier of the AI ​​model associated with the CSI report. For example, m=0 indicates a CSI report that is not AI-related, m=1 indicates that the identifier of the AI ​​model associated with the CSI report is 0, m=2 indicates that the identifier of the AI ​​model associated with the CSI report is 1, etc. y denotes the number of all possible values ​​of y. For example, if y=0, 1, 2, or 3, then N y = 4. y is used to distinguish the periodicity of the CSI report and the channel on which the CSI report is carried. For example, y = 0 indicates an aperiodic CSI report carried on the PUSCH, y = 1 indicates a semi-persistent CSI report carried on the PUSCH, y = 2 indicates a semi-persistent CSI report carried on the PUCCH, and y = 3 indicates a periodic CSI report carried on the PUCCH. k is used to distinguish whether the CSI report includes L1-RSRP or L1-SINR. For example, k = 0 indicates a CSI report that includes L1-RSRP or L1-SINR, and k = 1 indicates a CSI report that does not include L1-RSRP or L1-SINR. c is used to distinguish the serving cell to which the CSI report belongs. s is used to distinguish the ID of the CSI report. N k denotes the number of all possible values ​​of k. cells denotes the maximum number of serving cells. M s indicates the maximum number of CSI reports.

[0313] It should be understood that the above values ​​of m and their corresponding meanings are merely examples. In another implementation, there may be a different correspondence between the values ​​of m and their corresponding meanings. Alternatively, in another implementation, the correspondence between the values ​​of m and their corresponding meanings may be a subset of the above correspondences.

[0314] Additionally, the above-mentioned embodiments 3 and 4 may alternatively be used in conjunction with embodiments 1 and 2.

[0315] For example, the priority of a CSI report may be determined based on one or more parameters, at least one of which indicates a type of the CSI report, and at least one of which indicates an AI model associated with the CSI report.

[0316] A combination of the first and third aspects is used as an example. For example, the priority of a CSI report is determined based on one or more parameters, where at least one of the one or more parameters indicates a type of the CSI report, the at least one parameter further indicates an AI model associated with the CSI report, and the at least one parameter further indicates other information of the CSI report.

[0317] A combination of Aspect 1 and Aspect 3 is used as an example. For example, the priority of a CSI report is determined based on one or more parameters, where at least one of the one or more parameters indicates a type of the CSI report, and the at least one of the one or more parameters indicates an AI model associated with the CSI report. The at least one parameter indicating the type of the CSI report further indicates other information of the CSI report. The at least one parameter indicating the AI ​​model associated with the CSI report further indicates other information of the CSI report. The at least one parameter indicating the type of the CSI report is completely different from or partially the same as the at least one parameter indicating the AI ​​model associated with the CSI report.

[0318] A combination of aspect 1 and aspect 4 is used as an example. For example, the priority of a CSI report is determined based on one or more parameters, where at least one of the one or more parameters indicates a type of the CSI report, and the at least one of the one or more parameters indicates an AI model associated with the CSI report. The at least one parameter indicating the type of the CSI report further indicates other information of the CSI report. The at least one parameter indicating the AI ​​model associated with the CSI report does not indicate other information of the CSI report. The at least one parameter indicating the type of the CSI report is completely different from the at least one parameter indicating the AI ​​model associated with the CSI report.

[0319] A combination of aspects 2 and 3 is used as an example. For example, the priority of a CSI report is determined based on one or more parameters, where at least one of the one or more parameters indicates a type of the CSI report, and the at least one of the one or more parameters indicates an AI model associated with the CSI report. The at least one parameter indicating the type of the CSI report does not indicate other information of the CSI report. The at least one parameter indicating the AI ​​model associated with the CSI report further indicates other information of the CSI report. The at least one parameter indicating the type of the CSI report is completely different from the at least one parameter indicating the AI ​​model associated with the CSI report.

[0320] A combination of aspects 2 and 4 is used as an example. For example, the priority of a CSI report is determined based on one or more parameters, where at least one of the one or more parameters indicates a type of the CSI report, and the at least one of the one or more parameters indicates an AI model associated with the CSI report. The at least one parameter indicating the type of the CSI report does not indicate other information of the CSI report. The at least one parameter indicating the AI ​​model associated with the CSI report does not indicate other information of the CSI report. The at least one parameter indicating the type of the CSI report is completely different from the at least one parameter indicating the AI ​​model associated with the CSI report.

[0321] It should be understood that the above is merely an example, and the priority of the CSI report may be determined by another combination of aspects 1 to 4. This is not limited to the embodiments of the present application.

[0322] A CSI report may be divided into Part 1 and Part 2, or may be used as a whole, i.e., not divided. A CSI report that is not divided into Part 1 or Part 2 may be considered to be entirely Part 1 or Part 2.

[0323] The CSI report may be carried on the PUSCH or may be carried on the PUCCH. Alternatively, a portion of the CSI report may be carried on the PUSCH and another portion of the CSI report may be carried on the PUCCH. For example, part 1 is carried on the PUCCH and part 2 is carried on the PUSCH.

[0324] For example, a CSI report based on AI model inference may be divided into Part 1 and Part 2. The report content of Part 1 may indicate one or more of the following: RI, CQI, ID of the AI ​​model, original total size of the CSI report, actual total size of the CSI report, original size of Part 2, actual size of Part 2, original size of CSI at each layer, actual size of CSI at each layer, etc. The sizes may be expressed by the amount of bits. The report content of Part 2 may indicate AI-PMI. The original size is the size before discarding, and the actual size is the actual size for transmission. A CSI report is used as an example. The original total size of the CSI report is the size of the CSI report before discarding, and the actual total size of the CSI report is the size of the report content in the CSI report that is actually transmitted to the network device.

[0325] For example, a CSI report for data collection may be divided into Part 1 and Part 2. The report content of Part 1 may indicate one or more of the quantization mode of the data CSI, the codebook parameters, the CSI information type, the number of non-zero coefficients, etc. The quantization mode of the data CSI may include W-bit scalar quantization, float16 quantization, etc. The report content of Part 2 may indicate the data CSI. Float16 quantization means that floating-point numbers are quantized by using 16 bits (bits). W is a positive integer. For example, W may be any integer within the range of [1, 8].

[0326] For example, a CSI report for performance monitoring may be divided into Part 1 and Part 2. The report content of Part 1 may indicate the overall accuracy of the reconstructed CSI information. The overall accuracy of the reconstructed CSI information may be determined based on the accuracy of each layer and / or each subband in the reconstructed CSI information. For example, the overall accuracy of the reconstructed CSI information may be expressed as an average value of SGCS between the reconstructed information of each layer and / or each subband and the raw information of each layer and / or each subband. The report content of Part 2 may indicate the accuracy of the reconstructed information of each layer and / or each subband in the reconstructed CSI information.

[0327] As described above, the CSI report for performance monitoring is generally small, and the CSI report for performance monitoring may not need to be divided. In other words, the CSI report for performance monitoring may be considered as part 1 or part 2 of the CSI report as a whole.

[0328] Part 2 may be divided into multiple groups.

[0329] The CSI feedback mode based on the R16 codebook is used as an example. Part 2 of the CSI report can be divided into three groups. In descending order of priority, the three groups are group 0, group 1, and group 2. Group 0 includes oversampling selection (or phase rotation selection), spatial-domain basis indication, and strongest coefficient indication. Group 1 includes frequency-domain basis indication, part of coefficient indication, and part of coefficient amplitude and phase. Group 2 includes remaining coefficient indication, and remaining coefficient amplitude and phase.

[0330] It should be understood that the above is just an example, and the number of groups obtained by dividing part 2 may be 3 or another value, which is not limited in the embodiment of the present application.

[0331] Part 2 may not be divided into groups, in other words, part 2 may be considered as a group as a whole.

[0332] Part 1 is generally not divided into groups, in other words, part 1 is considered as a group as a whole. Alternatively, part 1 may be divided into groups. This is not limited to the embodiment of the present application.

[0333] As mentioned above, one of the N CSI reports may include one of the p groups.

[0334] For example, part 2 of the CSI report may not be divided, and part 2 may be used as one of the p groups included in the CSI report. In another example, the CSI report may not be divided, and the CSI report may be used as one of the p groups included in the CSI report. In another example, part 2 of the CSI report may be divided into group 0, group 1, and group 2. Any of group 0, group 1, and group 2 may be used as one of the p groups included in the CSI report.

[0335] Alternatively, one of the N CSI reports may include multiple groups among the p groups.

[0336] For example, the CSI report may be divided into Part 1 and Part 2, and Part 1 and Part 2 may be used as two of the p groups included in the CSI report. As another example, Part 2 of the CSI report may be divided into Group 0, Group 1, and Group 2. Any multiple of Group 0, Group 1, and Group 2 may be used as multiple groups of the p groups included in the CSI report.

[0337] When a CSI report includes multiple groups, the priority of the groups may be determined in multiple ways.

[0338] For example, for the same CSI report, the priority of part 1 may be higher than the priority of part 2.

[0339] For example, part 2 of the CSI report is N g +1 groups, and the priority of the groups in descending order is group 0, group 1, ..., and group N. g N g is a non-negative integer.

[0340] If part 2 is not divided, part 2 may be used as any one of the aforementioned groups, as needed.

[0341] For ease of explanation, this order is only used as an example for explanation in this embodiment of the present application, and does not constitute any limitation on the scope of this embodiment of the present application.

[0342] For different CSI reports, g The values ​​of may be the same or different.

[0343] Whether the N CSI reports should be split and the specific splitting scheme may be independent of each other.

[0344] Optionally, the N CSI reports include a first CSI report, and the type of the first CSI report is a first CSI report type. The first CSI report includes multiple groups among the p groups.

[0345] The first CSI report may be any one of the N CSI reports.

[0346] The type of the first CSI report is a first CSI report type, that is, the report content of the first CSI report includes channel information obtained through inference by an AI model. The first CSI report is a CSI report based on AI model inference.

[0347] For example, N = 1. In this case, the priority of the multiple groups may be based on the priority of the multiple groups in the first CSI report.

[0348] For example, N>1. In this case, the priority of the multiple groups may be based on the priority of the first CSI report and the priorities of the multiple groups in the first CSI report.

[0349] For example, the plurality of groups includes p groups, and the priorities of the p groups among the p groups are based on the priority of a first CSI report among the N CSI reports and the priorities of the p groups in the first CSI report, where p is an integer greater than 1.

[0350] In this embodiment of the present application, the first CSI report may be divided into multiple groups, thereby improving the flexibility of information transmission and increasing the probability that a network device will acquire a certain group in the first CSI report. This helps the network device acquire partial information in the CSI report. For example, the first CSI report is divided into multiple groups, and the groups correspond to different priorities. When the first resource cannot carry p groups, even if the entire first CSI report cannot be transmitted, the probability that the network device will acquire some groups in the first CSI report can be increased. This helps the network device acquire partial information in the CSI report.

[0351] The following describes the priorities in CSI reports based on AI model inference with reference to Examples 1 to 3.

[0352] Example 1:

[0353] Optionally, the channel information obtained through inference by the AI ​​model includes channel information of multiple layers, each of the p1 groups includes channel information of at least one layer among the multiple layers, and priorities of the p1 groups in the first CSI report match priorities of the channel information of the at least one layer in the p1 groups.

[0354] The priority of the channel information of the at least one layer is the priority of the channel information of the at least one layer among the channel information of the plurality of layers.

[0355] For example, the priority of the channel information of the at least one layer is related to the index of the at least one layer.

[0356] In other words, for a CSI report based on AI model inference, the priority of the channel information obtained through inference by the AI ​​model may be classified based on layers. The priority of the channel information of a layer is related to the index of the layer. In this embodiment of the present application, the priority of the channel information of a layer may also be referred to as the layer priority.

[0357] For example, a smaller index of a layer indicates a higher priority of the channel information of the layer. In this case, the channel information of layer 1 has the highest priority.

[0358] In another example, a larger layer index indicates a higher priority for the channel information of the layer, where the channel information of Layer 1 has the lowest priority.

[0359] For example, the priorities of the channel information of a layer or the layer indices are related to the eigenvalues ​​or singular values ​​corresponding to the eigenvectors of the layer. The priorities of the channel information of a layer or the layer indices are sorted in ascending order of the singular values ​​corresponding to the eigenvalues ​​or the eigenvectors of the layer, or the priorities of the channel information of a layer or the layer indices are sorted in descending order of the singular values ​​corresponding to the eigenvalues ​​or the eigenvectors of the layer. Alternatively, the layer indices may be determined in another manner.

[0360] In this embodiment of the present application, the priority of channel information acquired through AI inference may be classified based on layers, and the priority of the channel information of a layer may correspond to the priority of the group to which the layer belongs in the first CSI report. The beamforming gains of different layers may differ. Generally, a larger eigenvalue or singular value corresponding to the eigenvector of a layer indicates a larger beamforming gain of that layer. Therefore, the beamforming gains of different layers may be distinguished through layer-based priority classification. This is useful for implementing priority classification based on different beamforming gains. For example, a layer with a larger beamforming gain may be used as a higher-priority layer, and the higher-priority layer belongs to a higher-priority group. In this way, it may be ensured that this portion of CSI can be fed back preferentially. In other words, the probability that a network device acquires channel information of a layer with a larger beamforming gain is increased. For example, the layer indices may be arranged in descending order of the eigenvalues ​​or singular values ​​corresponding to the eigenvector of the layer. A smaller layer index indicates a higher priority of the channel information of the layer. In this way, layers with greater beamforming gain may be used as higher priority layers.

[0361] In addition, the sparsity of the eigenvectors of different layers may be different, i.e., the feedback accuracy may be different. Generally, the larger the eigenvalue or singular value corresponding to the eigenvector of a layer, the stronger the sparsity of the eigenvector of the layer, i.e., the higher the feedback accuracy for the same feedback overhead. Therefore, the feedback accuracy of different layers can be distinguished through layer-based priority classification. This is useful for implementing priority classification based on different feedback accuracy. For example, a layer with higher feedback accuracy may be used as a higher-priority layer, and the higher-priority layer belongs to a higher-priority group. In this way, it can be ensured that this portion of CSI can be fed back preferentially. In other words, the probability that a network device acquires channel information for a layer with higher feedback accuracy increases. For example, the layer indices may be arranged in descending order of the eigenvalues ​​or singular values ​​corresponding to the eigenvectors of those layers. A smaller layer index indicates a higher priority for the channel information of the layer. In this way, a layer with higher feedback accuracy may be used as a higher-priority layer.

[0362] In this case, the CSI information obtained through inference by the AI ​​model may include channel information corresponding to different layers. The CSI information obtained through inference by the AI ​​model is a string of bits output by the AI ​​model. For example, a correspondence between the channel information of those layers and the bits output by the AI ​​model may be preset. For example, the AI ​​model outputs 20 bits, and the channel information of multiple layers is the AI-PMI of two layers, with the first 10 bits being the AI-PMI of layer 1 and the last 10 bits being the AI-PMI of layer 2.

[0363] The fact that the priorities of the p1 groups in the first CSI report match the priorities of the channel information of the at least one layer included in the p1 groups can be understood as follows: Among the p1 groups, the priority of Group #1 is higher than the priority of Group #2, and the priority of the channel information of the at least one layer included in Group #1 is higher than the priority of the channel information of the at least one layer included in Group #2. For example, a smaller index of a layer indicates a higher priority of the channel information of that layer. In this case, the priority of Group #1 is higher than the priority of Group #2, and the index of the at least one layer included in Group #1 is lower than the index of the at least one layer included in Group #2.

[0364] In the embodiment of the present application, the p1 groups may be all or some of the groups of the CSI report. For example, among all the groups of the CSI report, some groups include channel information of a layer, and these some groups are the above p1 groups, and other groups do not include channel information of a layer. Alternatively, among all the groups of the CSI report, each group includes channel information of a layer, and all the groups are the above p1 groups. This is not a limitation in the embodiment of the present application.

[0365] For the sake of explanation, the following uses an example in which the channel information obtained through inference by the AI ​​model is AI-PMI. The channel information of the multiple layers is the AI-PMI of the multiple layers. For example, the report content of part 2 may indicate AI-PMI, and part 2 may be divided into three groups, i.e., p1=3. In descending order of priority, the three groups are group 0, group 1, and group 2, respectively. The smaller the layer index, the higher the priority of the layer.

[0366] For example, group 0 contains the AI-PMI for layer 1, group 1 contains the AI-PMI for layer 2, and group 2 contains the AI-PMI for the remaining layers.

[0367] In another example, group 0 includes the AI-PMI of layer 1, group 1 includes the AI-PMI of layer 2 and layer 3, and group 2 includes the AI-PMI of the remaining layers.

[0368] In another example, Group 1 contains the AI-PMI for Layer 1, and Group 2 contains the AI-PMI for the remaining layers.

[0369] It should be understood that the above is merely an example, and there may be other correspondences between layers and groups, which does not constitute any limitation on the solution in the embodiments of the present application.

[0370] Example 2:

[0371] Optionally, the channel information obtained through inference by the AI ​​model includes channel information of a plurality of segments, where p1 groups each include channel information of at least one segment among the plurality of segments, and priorities of the p1 groups in the first CSI report match priorities of the channel information of the at least one segment in the p1 groups.

[0372] The priority of the channel information of the at least one segment is the priority of the channel information of the at least one segment in the channel information of the plurality of segments.

[0373] For example, the priority of the channel information for a segment is related to the index of that segment.

[0374] In other words, for a CSI report based on AI model inference, the priority of the channel information obtained through the AI ​​model inference may be classified based on the segments. The channel information obtained through the AI ​​model inference may be divided into the channel information of the multiple segments, and the priority of the channel information of the segments is related to the index of the segment. In this embodiment of the present application, the priority of the channel information of a segment may also be referred to as the priority of the segment.

[0375] For example, a smaller index for a segment indicates a higher priority for the channel information for that segment.

[0376] In another example, a larger index for a segment indicates a higher priority for the channel information for that segment.

[0377] If the AI ​​model supports segmentation, the priorities can be classified based on the segments.

[0378] For example, CSI feedback information (e.g., AI-PMI) generated by a CSI generator is divided into multiple segments, and the segments are input to a CSI reconstructor, which can reconstruct reconstructed CSI information. Compared to reconstructed CSI information reconstructed based on the complete CSI feedback information, reconstructed CSI information reconstructed based on the segments may have lower accuracy. For example, the CSI generator outputs 20 bits, with segment 1 including the first 10 bits and segment 2 including the last 10 bits. The CSI reconstructor can reconstruct reconstructed CSI information based on the first 10 bits. Compared to reconstructed CSI information reconstructed based on 20 bits, reconstructed CSI information reconstructed based on the first 10 bits has lower accuracy.

[0379] If the CSI reconstructor can reconstruct the reconstructed CSI information based on segments, the AI ​​model may be considered to support segmentation. If the CSI reconstructor can reconstruct the reconstructed CSI information based only on complete CSI feedback information but cannot reconstruct the reconstructed CSI information based on segments, the AI ​​model may be considered not to support segmentation.

[0380] For example, the priority of a segment's channel information or its index is related to whether CSI can be independently restored based on the segment. For illustrative purposes, the following uses an example in which a smaller segment index indicates a higher priority of the segment's channel information. The segment from which CSI can be independently restored has the highest priority channel information or the lowest segment index. For example, channel information acquired through inference using an AI model may be divided into three segments, with the first segment being used to independently restore CSI, the second segment and the first segment being used jointly to restore CSI, and the third segment, the first segment, and the second segment being used jointly to restore CSI. In this case, the first segment has the highest priority or the lowest index, the second segment has the second highest priority or the second lowest index, and the third segment has the lowest priority or the highest index.

[0381] In this embodiment of the present application, priority classification is performed based on segments, whereby segments based on which CSI can be independently reconstructed have higher priorities, and higher priority segments belong to higher priority groups. In this way, it can be preferentially guaranteed that this portion of CSI can be fed back. In other words, the probability that a network device will obtain a segment based on which CSI can be independently reconstructed increases. This helps the network device reconstruct CSI. In this way, even if the entire first CSI report cannot be transmitted, the probability that the network device will obtain a higher priority group can be increased. Even if a segment with a lower priority is discarded, the network device can still reconstruct CSI based on the remaining segments with a higher priority.

[0382] The fact that the priorities of the p1 groups in the first CSI report match the priority of the channel information of the at least one segment in the p1 groups can be understood as follows: Among the p1 groups, the priority of group #1 is higher than the priority of group #2, and the priority of the channel information of at least one segment included in group #1 is higher than the priority of the channel information of at least one segment included in group #2. For example, a smaller segment index indicates a higher priority of the channel information of the segment. In this case, the priority of group #1 is higher than the priority of group #2, and the index of the at least one segment included in group #1 is smaller than the index of the at least one segment included in group #2.

[0383] In the embodiment of the present application, the p1 groups may be all or some of the groups of the CSI report. For example, among all the groups of the CSI report, some groups include channel information of the segments, and these some groups are the p1 groups, and other groups do not include channel information of the segments. Alternatively, among all the groups of the CSI report, each group includes channel information of the segments, and all the groups are the p1 groups. This is not limited to the embodiment of the present application.

[0384] For the sake of explanation, the following uses an example in which the channel information obtained through inference by the AI ​​model is AI-PMI. The channel information of the plurality of segments is the AI-PMI of the plurality of segments. For example, the report content of Part 2 may indicate AI-PMI, and Part 2 may be divided into three groups, i.e., p1=3. In descending order of priority, the three groups are Group 0, Group 1, and Group 2, respectively. The smaller the segment index, the higher the segment priority.

[0385] For example, group 0 contains the AI-PMI of segment 1, group 1 contains the AI-PMI of segment 2, and group 2 contains the AI-PMI of the remaining segments.

[0386] In another example, group 0 includes the AI-PMI of segment 1 and the AI-PMI of segment 2, group 1 includes the AI-PMI of segment 3 and the AI-PMI of segment 4, and group 2 includes the AI-PMI of the remaining segments.

[0387] In another example, group 1 contains the AI-PMI for segment 1, and group 2 contains the AI-PMI for the remaining segments.

[0388] It should be understood that the above is merely an example, and there may be other correspondences between segments and groups, which does not constitute any limitation on the solution in the embodiments of the present application.

[0389] The number of segments, the number of bits included in each segment, the bits included in each segment, and the correspondence between segments and groups may be predefined, configured by the network device, or partially configured by the network device and partially predefined. Alternatively, some or all of the above information may be determined by the terminal device, and the terminal device may report related information to the network device. For example, the terminal device may report to the network device the number of segments supported by each AI model on the terminal device, the number of bits included in each segment, and the bits included in each segment. The bits included in each segment indicate the bits included in that segment.

[0390] Example 3:

[0391] Optionally, the channel information obtained through inference by the AI ​​model includes channel information of a plurality of segments, where p1 groups each include channel information of at least one segment among the plurality of segments, and priorities of the p1 groups in the first CSI report correspond to priorities of the channel information of the at least one segment in the p1 groups, and the priority of the at least one segment is related to a layer in which the at least one segment is located and a position of the at least one segment in the layer.

[0392] The difference between Example 3 and Example 2 is that in Example 2, the channel information obtained through inference by the AI ​​model is segmented as a whole, each segment may include multiple layers, and the priority of each segment is independent of the layers. Example 3 can be understood as follows: The channel information obtained through inference by the AI ​​model includes channel information of the multiple layers. Some or all of the layers are segmented to obtain the channel information of the multiple segments. The priority of a segment is related to the layer in which the segment is located and the position of the segment in that layer. For further explanation, please refer to Example 2. To avoid repetition, some explanations will be omitted as appropriate in the explanation of Example 3.

[0393] In Example 3, the number of segments in a layer may be the same or different. For a non-segmented layer, the number of segments in a layer may be considered to be 1. The number of bits included in the segments in each layer may be the same or different. The bits included in the segments in each layer may be the same or different.

[0394] The number of segments obtained by dividing each layer, the number of bits included in each segment of each layer, the bits included in each segment of each layer, the correspondence between segments and groups in each layer, etc. may be predefined, configured by the network device, or partially configured by the network device and partially predefined. Alternatively, some or all of the above information may be determined by the terminal device, and the terminal device may report related information to the network device. For example, the terminal device may report to the network device the number of bits included in each segment of each layer and the bits included in each segment of each layer. The bits included in each segment of each layer indicate the bits included in the segments of that layer.

[0395] For example, the priority of the channel information for a segment is related to the index of that segment.

[0396] FIG. 6 is a diagram showing how the priority of a segment is determined.

[0397] For example, the priority can be determined by first dividing the segments and then layering them, as shown in Figure 6. In other words, the priority of each segment is adjacent to the priority of the adjacent segments in the layer in which it is located.

[0398] For example, the priority pri(i,j) of a segment may satisfy the following formula: pri(i,j)=N layer i+j

[0399] i indicates the index of the segment in the layer, i.e., the position of the segment in the layer; j indicates the index of the layer in which the segment is located, i.e., the layer in which the segment is located. N layer indicates the total number of layers.

[0400] FIG. 7 is a diagram of another aspect of determining segment priorities.

[0401] For example, the priority can be determined by layering first and then segmenting, as shown in Figure 7. For a given segment, a segment with a priority adjacent to the priority of the given segment is a segment in an adjacent layer that is located at the same position as the given segment in the layer in which the given segment is located.

[0402] For example, the priority pri(i,j) of a segment may satisfy the following formula: pri(i,j)=N sec j+i

[0403] i indicates the index of the segment in the layer, i.e., the position of the segment in the layer; j indicates the index of the layer in which the segment is located, i.e., the layer in which the segment is located. N sec indicates the number of segments in each layer.

[0404] In Figures 6 and 7, each layer is sec Only the example of dividing into segments is used, but does not constitute any limitation to the solution in the embodiments of the present application.

[0405] For example, a smaller value of pri(i,j) indicates a higher priority of the segment.

[0406] For example, a larger value of pri(i,j) indicates a higher priority for the segment.

[0407] The fact that the priorities of the p1 groups in the first CSI report match the priority of the channel information of the at least one segment in the p1 groups can be understood as follows: In the p1 groups, the priority of group #1 is higher than the priority of group #2, and the priority of the channel information of at least one segment included in group #1 is higher than the priority of the channel information of at least one segment included in group #2.

[0408] For example, part 2 may be divided into three groups, i.e., p1 = 3. In descending order of priority, the three groups are group 0, group 1, and group 2, respectively.

[0409] For example, group 0 contains the first a segments with the highest priority, group 1 contains the b segments with the highest priority after the first a segments, and group 2 contains the remaining segments, where a is a positive integer and b is a positive integer.

[0410] In another example, group 1 contains the first c segments with the highest priority, and group 2 contains the AI-PMI of the remaining segments, where c is a positive integer.

[0411] It should be understood that the above is merely an example, and there may be other correspondences between segments and groups, which does not constitute any limitation on the solution in the embodiments of the present application.

[0412] In the three examples above, the groups may also contain other information.

[0413] For example, among the plurality of groups within the p groups included in the first CSI report, the group with the highest priority may further include auxiliary information indicating, for example, one or more contents of the ID of the AI ​​model, the original total size of the AI-PMI, the actual total size of the AI-PMI, the original size of the AI-PMI for each layer, and the actual size of the AI-PMI for each layer.

[0414] For example, in the plurality of groups within p groups included in the first CSI report, group 0 may be the group with the highest priority, and group 0 may include the auxiliary information.

[0415] Optionally, the N CSI reports include a second CSI report, and the type of the second CSI report is a second CSI report type. The second CSI report includes multiple groups among the p groups.

[0416] The second CSI report may be any one of the N CSI reports.

[0417] The second CSI report type is a second CSI report type, that is, the report content of the second CSI report includes performance information of the AI ​​model. The second CSI report is a CSI report for performance monitoring.

[0418] For example, N = 1. In this case, the priorities of the multiple groups may be based on the priorities of the multiple groups in the second CSI report.

[0419] For example, N>1. In this case, the priorities of the groups may be based on the priorities of the second CSI report and the priorities of the groups in the second CSI report.

[0420] For example, the plurality of groups includes p groups, where the priorities of the p groups are based on the priority of a first CSI report in the N CSI reports and the priorities of the p groups in the second CSI report, where p is an integer greater than 1.

[0421] Optionally, the performance information of the AI ​​model includes overall performance information, performance information of the plurality of layers, or performance information of a plurality of subbands, and each of the p2 groups includes at least one piece of performance information in the performance information of the AI ​​model, and the priority of the p2 groups in the second CSI report matches the priority of the at least one piece of performance information in the p2 groups.

[0422] The priority of the at least one piece of performance information is the priority of the at least one piece of performance information among the plurality of pieces of performance information.

[0423] In the following, the overall performance information, the performance information of the multiple layers, and the performance information of the multiple subbands will be described using an example in which the performance information of an AI model is the accuracy of the recovered CSI information.

[0424] For example, the overall performance information may be the overall accuracy of the reconstructed CSI information. The performance information of a layer may be the accuracy of the reconstructed information of that layer. The performance information of a subband may be the accuracy of the reconstructed information of that subband. For example, the accuracy may be expressed as the SGCS between the reconstructed information and the raw information. The performance information of a layer may be the SGCS between the reconstructed CSI information of the layer and the raw CSI information of the layer. The overall accuracy of the reconstructed CSI information may be determined based on the accuracy of each layer and / or each subband, which is the accuracy within the reconstructed CSI information. For example, the overall performance information may be the average value of the SGCS of each layer and / or each subband. The performance information of a subband may be the SGCS between the reconstructed CSI information of the subband and the raw CSI information of the subband.

[0425] For a related description of the performance information of the AI ​​model, please refer to the above description, and the details will not be described again here.

[0426] Optionally, when the AI ​​performance information includes overall performance information, the priority of the overall performance information may be the highest priority of the performance information in the AI ​​performance information.

[0427] Optionally, the AI ​​performance information may include overall performance information and performance information of the multiple layers, and the priority of the overall performance information is higher than the priority of the performance information of the multiple layers.

[0428] Optionally, the AI ​​performance information may include overall performance information and performance information of the plurality of subbands, and the priority of the overall performance information is higher than the priority of the performance information of the plurality of subbands.

[0429] The performance priority of a layer or subband may be determined in the order of layer first, then subband. Specifically, for a given subband, a subband with a priority adjacent to the priority of the subband is a subband in an adjacent layer that is at the same position as the subband in the layer in which the subband is located.

[0430] Alternatively, the performance priorities of layers or subbands may be determined in the order of subband first and then layer, specifically, the priority of each subband is adjacent to the priorities of adjacent subbands in the layer in which it is located.

[0431] For a detailed description, see Example 3 above. The channel information of the segment needs to be replaced with the performance information of the subband. To avoid repetition, the details will not be described again here.

[0432] Note that in this embodiment of the present application, the p2 groups may be all or some of the groups of the CSI reports. For example, among all the groups of the CSI reports, some groups include performance information, and these some groups are the above p2 groups, while other groups do not include performance information. Alternatively, among all the groups of the CSI reports, each group includes performance information, and all groups are the above p2 groups. This is not limited to the embodiment of the present application.

[0433] The fact that the priorities of the p2 groups in the second CSI report match the priority of at least one piece of performance information included in the p2 groups can be understood as follows: Among the p2 groups, the priority of group #3 is higher than the priority of group #4, and the priority of at least one piece of performance information included in group #3 is higher than the priority of at least one piece of performance information included in group #4.

[0434] For example, the performance information of the AI ​​model may be placed in part 2 of the second CSI report, and part 2 of the second CSI report may be divided into three groups, i.e., p2=3. In descending order of priority, the three groups are group 0, group 1, and group 2, respectively. A smaller layer index indicates a higher priority of the layer.

[0435] For example, group 0 may include overall performance, group 1 may include performance for some layers or subbands, and group 2 may include performance for the remaining layers or subbands. For example, each layer may include five subbands, for a total of two layers. In this case, the performance information of the AI ​​model may include overall performance and performance for 10 subbands, with group 0 including overall performance, group 1 including performance for the first four subbands with the highest priority, and group 2 including performance for the remaining six subbands.

[0436] It should be understood that the above is merely an example, and there may be other correspondences between performance information and groups, which does not constitute any limitation on the solution in the embodiment of the present application.

[0437] Optionally, at least two of the N CSI reports have the same priority.

[0438] For example, the prioritization scheme may be applied to a scenario in which multiple CSI reports need to be used together, which may be a scenario in which multiple CSI reports need to be present together to be valid.

[0439] For example, multiple CSI reports with the same priority may be used in a scenario where performance is monitored. For example, the multiple CSI reports with the same priority may include a CSI report based on AI model inference and a CSI report in a traditional feedback mode (e.g., a codebook-based feedback mode). A network device may compare the accuracy of the channel information fed back by using the two CSI reports to evaluate the performance of the AI ​​model.

[0440] For multiple CSI reports having the same priority, when the multiple CSI reports need to be discarded, the multiple CSI reports may be discarded simultaneously, or some of the groups having the same priority may be discarded based on the priorities in the multiple CSI reports. In step 520, determining some groups from the p groups may be understood as discarding other groups from the p groups.

[0441] For example, the priorities of the N CSI reports may be determined by the terminal device.

[0442] Optionally, the method 500 further includes receiving first indication information from the network device, the first indication information indicating report priorities of some or all of the N CSI reports.

[0443] In other words, the report priorities of some or all of the N CSI reports may be dynamically configured or adjusted by the network device.

[0444] For example, the network device may determine report priorities for some or all of the N CSI reports in the manner described above.

[0445] For example, a network device may dynamically configure or adjust the priority of a CSI report by using one or more of RRC signaling, MAC CE, or DCI.

[0446] In other words, the first indication information may be carried in one or more of RRC signaling, MAC CE, or DCI.

[0447] For example, the network device may configure the priority of the CSI report in the RRC signaling for configuring the CSI report.

[0448] In another example, the network device may configure a priority of the CSI report in the MAC CE for activating the CSI report.

[0449] In another example, a network device may configure or adjust the priority of one or more reports in a DCI.

[0450] In this way, the priority of the CSI reports can be controlled more flexibly to facilitate adjustment based on actual requirements.

[0451] For example, a network device may configure the same priority for multiple CSI reports. For a detailed description, see the previous description. The details will not be described again here.

[0452] The following describes the priority of a group of part 2 of N CSI reports: g +1 groups. For example, in part 2 of each CSI report, the descending priorities are, in order, Group 0, Group 1, ..., and Group N. g N gis an integer equal to or greater than 0. The priorities of the N CSI reports are, in descending order, CSI report 1, CSI report 2, ..., and CSI report N. The process of determining the numbers 1 to N is the process of determining the priorities of the N CSI reports. For the determination manner, refer to the above-mentioned manner 1 and manner 2. The details will not be described again here.

[0453] For example, the priority of groups in part 2 of N CSI reports is, in descending order, group 0 of CSI report 1 to CSI report N, group 1 of CSI report 1, group 2 of CSI report 1, ..., group N of CSI report 1. g , CSI Report 2 Group 1, CSI Report 2 Group 2, …, CSI Report 2 Group N g , …, Group 1 of CSI report N, Group 2 of CSI report N, and Group N of CSI report N g may be.

[0454] Group 0 of CSI report 1 to CSI report N is group 0 of CSI report 1, group 0 of CSI report 2, ..., and group 0 of CSI report N. Group 0 of CSI report 1 to CSI report N may have the same priority. Alternatively, group 0 of CSI report 1 to CSI report N may have N different priorities, and may be group 0 of CSI report 1, group 0 of CSI report 2, ..., and group 0 of CSI report N, in descending order of priority.

[0455] For example, N g = 3, and the priority of the groups of part 2 of the N CSI reports may be, in descending order, group 0 of CSI report 1 to group 1 of CSI report 1, group 2 of CSI report 1, group 3 of CSI report 1, group 1 of CSI report 2, group 2 of CSI report 2, group 3 of CSI report 2, ..., group 1 of CSI report N, group 2 of CSI report N, and group 3 of CSI report N.

[0456] For example, the priority of groups in part 2 of N CSI reports is, in descending order, group 0 of CSI report 1 to group 1 of CSI report 1, group 2 of CSI report 1, ..., group N of CSI report 1. g -1, Group 1 of CSI Report 2, Group 2 of CSI Report 2, …, Group N of CSI Report 2 g -1, Group 1 of CSI report N, Group 2 of CSI report N, …, Group N of CSI report N g -1, Group N of CSI Report 1 through CSI Report N g may be.

[0457] For related descriptions of Group 0 of CSI Report 1 to CSI Report N, please refer to the above description.

[0458] Group N of CSI Report 1 to CSI Report N g Group N of CSI Report 1 g , Group N of CSI Report 2 g , …, Group N of CSI Report N g Group N of CSI Report 1 to CSI Report N g Alternatively, group N of CSI report 1 to CSI report N may have the same priority. g may have N different priorities, and in descending order of priority, group N of CSI report 1 g , Group N of CSI Report 2 g , …, Group N of CSI Report N g may be.

[0459] For example, N g= 3, and the priorities of the groups of part 2 of the N CSI reports may be, in descending order, group 0 of CSI report 1 to CSI report N, group 1 of CSI report 1, group 2 of CSI report 1, group 1 of CSI report 2, group 2 of CSI report 2, ..., group 1 of CSI report N, group 2 of CSI report N, and group 3 of CSI report 1 to CSI report N.

[0460] It should be understood that the above is merely an example and the priorities of the groups of part 2 of the N CSI reports may alternatively be in another order.

[0461] In the above example, it is not limited that each CSI report needs to be divided into part 1 and part 2, and part 2 of each CSI report is N g It is not limited to being divided into +1 groups.

[0462] For example, if a CSI report is not divided into part 1 or part 2, the CSI report may be considered as part 1 or part 2 as a whole.

[0463] For example, if part 2 of the CSI report is not divided into multiple groups, part 2 of the CSI report may be considered as a group as a whole.

[0464] The priority sorting of the groups of part 2 of N CSI reports may also be applicable when part 2 of a CSI report is considered as a group as a whole, and only other groups of part 2 of a CSI report need to be removed from the sorting. For example, part 2 of CSI report 3 may be considered as group 0 of CSI report 3. In this case, group 1 to group N of CSI report 3 in the above sorting may be removed. g may be removed. The remaining sorting may be used as a priority sorting for part 2 groups of N CSI reports.

[0465] As described above, in step 520, a portion of the p groups to be transmitted on the first resource can be determined based on the priorities of the p groups.

[0466] In other words, based on the priorities of the p groups, the unselected groups among the p groups are discarded. In this embodiment of the present application, the discarding may be performed on a group-by-group basis among the p groups.

[0467] For example, the third CSI report in the N CSI reports includes one of the p groups.

[0468] The third CSI report may be any one of the N CSI reports.

[0469] In other words, for p groups, some CSI reports include only one group.

[0470] For example, for one of the N CSI reports, if the CSI report is not divided into part 1 or part 2, the CSI report may be used as part 1 or part 2 as a whole, i.e., as one of the p groups as a whole. If a group needs to be discarded, the entire CSI report is discarded. If a group needs to be reported, the entire CSI report is reported.

[0471] In another example, for one of the N CSI reports, if part 2 of the CSI report is not divided into multiple groups, part 2 of the CSI report as a whole may be considered as one of the p groups. If the group needs to be discarded, the entire part 2 of the CSI report is discarded. If the group needs to be reported, the entire part 2 of the CSI report is reported.

[0472] According to the solution in this embodiment of the present application, for any information with a high integrity requirement, if part of the information is discarded, subsequent configuration may be affected. The information with a high integrity requirement may be used as one of p groups. In this manner, all of the information with a high integrity requirement may be discarded or uploaded. For example, the information with a high integrity requirement may be located in a CSI report, and the CSI report may not be divided into Part 1 and Part 2. Alternatively, the information with a high integrity requirement may be located in Part 2 of the CSI report, and Part 2 may not be divided into groups.

[0473] A CSI report for data collection is used as an example. If partial information in the data CSI is discarded, the accuracy of the data CSI will be reduced, and the accuracy of the operation related to the AI ​​model will be affected. For example, the training performance of the AI ​​model or the performance monitoring results of the AI ​​model will be affected. The CSI report for data collection may not be divided into part 1 and part 2. Alternatively, the CSI report for data collection may be divided into part 1 and part 2, the data CSI may be located in part 2, and part 2 will not be divided into multiple groups.

[0474] For example, in step 520, a subset of the p groups may be transmitted on the first resource based on the priorities of the p groups, where the total size of the subset of groups is less than or equal to a specified threshold. The specified threshold may be the size of a resource within the first resource that may be used to perform the transmission of the CSI report.

[0475] If the total size of the p groups exceeds a specified threshold, the terminal device may sequentially discard the groups with the lowest priority among the p groups until the total size of the part of the groups for which transmission needs to be performed on the first resource is equal to or less than the specified threshold.

[0476] As described above, some or all of the p groups may be sorted based on the priorities of the p groups in step 520. In other words, step 510 may be replaced by obtaining a CSI report to be transmitted on a first resource, where the CSI report includes all or some of the N CSI reports, and step 520 may be replaced by transmitting the obtained CSI report on the first resource, where the transmission format of the obtained CSI report satisfies the descriptions in the following embodiments.

[0477] The following describes the format of the CSI report by using an example.

[0478] For example, the p groups are sorted from front to back in descending order of the priority of the p groups.

[0479] For example, in any two of the N CSI reports, the priority of each group in the lower-priority CSI report is lower than the priority of the corresponding group in the higher-priority CSI report. For example, the N CSI reports include CSI report #1 and CSI report #2. CSI report #1 includes p1 groups, where the p1 groups include groups p1-1 and p1-2, and CSI report #2 includes p1' groups, where the p1' groups include groups p1'-1 and p1'-2. The priority of group p1-1 is higher than the priority of group p1-2, and the priority of group p1'-1 is higher than the priority of group p1'-2, and the priority of group p1-1 is higher than the priority of group p1'-1, and the priority of group p1'-1 is higher than the priority of group p1-2, and the priority of group p1-2 is higher than the priority of group p1'-2.

[0480] Groups that have the same priority in two CSI reports are corresponding groups.

[0481] In other words, during priority determination, the priority of the group is considered first, and then the priority of the CSI report to which the corresponding group belongs is considered.

[0482] In the following, the format of the CSI report when N=1 is explained by using an example.

[0483] For example, p groups are those in part 1 and CSI report, N in part 2 g +1 groups, i.e., p=N g +2. The priority of part 1 is higher than the priority of part 2. Part 2 is N g +1 groups, and the priorities of those groups in descending order are, in order, Group 0, Group 1, ., and Group N. g This becomes:

[0484] The internal structure of the CSI report is shown in Figure 8, which, from front to back, is Part 1, Group 0, Group 1, …, and Group N. g This becomes:

[0485] It should be understood that Figure 8 is merely an example and does not constitute any limitation on the solution in the embodiments of the present application. Channel coding may be performed on Part 1 and Part 2 as a whole. For example, as shown in Figure 8, Part 2 may be concatenated after Part 1 for channel coding, or channel coding may be performed on Part 1 and Part 2 separately. Part 1 and Part 2 may be transmitted in the same message or in different messages. A CSI report may alternatively have only Part 1 or only Part 2.

[0486] For example, p groups are N in part 2 of the CSI report. g +1 groups, in which case the CSI report shown in Figure 8 does not include part 1.

[0487] For example, the p groups may include part 1 and part 2 of the CSI report, and part 2 may not be divided into multiple groups. In this case, part 2 of the CSI report shown in FIG. 8 is one group.

[0488] The following describes the format of the CSI report when N>1 by using an example: N>1, that is, multiple CSI reports are transmitted on the same resource.

[0489] For example, p groups are N of part 1 and part 2 of N CSI reports. g +1 group, i.e., p=N*(N g +2). In the CSI report, the priority of part 1 is higher than the priority of part 2. N of part 2 in descending order g The priority of the +1 groups is, in order, Group 0, Group 1, ..., Group N g This becomes:

[0490] The internal structure of the CSI report is shown in FIG. 9. Part 1 of the N CSI reports is arranged from front to back in descending order of the priorities of the N CSI reports. Part 1 of the N CSI reports may be considered as Part 1 as a whole. After Part 1 of the N CSI reports, for Part 2 of the N CSI reports, Group 0 of the N CSI reports is arranged from front to back in descending order of the priorities of the N CSI reports. For each CSI report, other groups other than Group 0 in Part 2 of the CSI report are arranged based on the priorities of the groups in the CSI report, and other groups of the N CSI reports are arranged based on the priorities of the N CSI reports. Part 2 of the N CSI reports may be considered as Part 2 as a whole.

[0491] 9 is merely an example and does not constitute any limitation on the solution in the embodiments of the present application. For example, the priority of the entire part 2 may alternatively be the above-mentioned priority order of the part 2 of the N CSI reports, or may be another priority order.

[0492] Channel coding may be performed on the entire part 1 and the entire part 2 as a whole. For example, the entire part 1 may be concatenated followed by the entire part 2 for channel coding, or channel coding may be performed separately on the entire part 1 and the entire part 2. The entire part 1 and the entire part 2 may be transmitted in the same message or in different messages. Channel coding may also be performed separately on part 1 and part 2 of each CSI report.

[0493] For example, in p groups, the partial CSI may include only part 1 or part 2. Alternatively, part 2 of the partial CSI may not be divided into multiple groups. In this case, the CSI report shown in FIG. 9 may be adaptively modified.

[0494] For example, CSI report 1 is considered as part 2 as a whole, i.e., the p groups do not include part 1 of CSI report 1. The arrangement order of the p groups may be shown in Figure 10, and only part 1 of CSI report 1 needs to be deleted from Figure 9.

[0495] In another example, CSI report 2 is considered as part 1 as a whole, i.e., the p groups do not include part 2 of CSI report 2. The arrangement order of the p groups may be shown in FIG. 11, and only part 2 of CSI report 2 needs to be deleted from FIG. 9.

[0496] In another example, part 2 of CSI report N is not divided and is considered as group 0 as a whole, i.e., the p groups do not include other groups in CSI report N. The arrangement order of the p groups may be shown in FIG. 12, and the other groups in part 2 of CSI report N need to be deleted from FIG. 9.

[0497] Furthermore, if the size of the p groups does not satisfy the transmission requirements of the first resource, e.g., if the size of the p groups is greater than a specified threshold, the p groups are discarded one by one in back-to-front order until the transmission requirements of the first resource are satisfied, e.g., until the size of the remaining groups is less than or equal to a specified threshold, and the remaining groups are part of the groups for which transmission needs to be performed on the first resource.

[0498] It can be understood that the information names in some of the above embodiments are merely examples and do not constitute limitations on the protection scope of the embodiments of the present application.

[0499] It can be further understood that the formulas used in the embodiments of the present application are merely examples for explanation, and do not constitute limitations on the protection scope of the embodiments of the present application. In the process of calculating the above related parameters, calculations may be performed based on the above formulas, or based on modifications of the above formulas, or calculations may be performed in other ways to satisfy the calculation results of the formulas.

[0500] It may further be understood that some optional features in the embodiments of the present application may be independent of other features in some scenarios or may be combined with other features in some scenarios, without limitation.

[0501] Furthermore, it can be understood that the solutions in the embodiments of the present application may be appropriately combined for use, and the explanations or descriptions of terms in the embodiments may be mutually referenced or explained in the embodiments. This is not limited thereto.

[0502] It may be further understood that the sequence numbers of various numerals in the embodiments of the present application do not imply an execution sequence, but are merely for distinction to facilitate explanation, and therefore should not constitute any limitation on the implementation process of the embodiments of the present application.

[0503] It may further be understood that in the method embodiments described above, the methods and operations performed by a device may alternatively be performed by a component (eg, a chip or circuit) of a device.

[0504] Corresponding to the methods provided in the above-mentioned method embodiments, an embodiment of the present application further provides a corresponding apparatus. The apparatus includes corresponding modules configured to perform the above-mentioned method embodiments. The modules may be software, hardware, or a combination of software and hardware. It may be understood that the technical features described in the method embodiments are also applicable to the following apparatus embodiments.

[0505] 13 is a diagram of a communications device 1700 according to an embodiment of the present application. The device 1700 includes a transceiver unit 1710 and a processing unit 1720. The transceiver unit 1710 may be configured to implement corresponding communications functions. The transceiver unit 1710 may also be referred to as a communications interface, a communications unit, etc. The processing unit 1720 may be configured, for example, to configure resources to implement corresponding processing functions. The processing unit 1720 may also be referred to as a processor, etc.

[0506] Optionally, the apparatus 1700 further includes a storage unit. The storage unit may be configured to store instructions and / or data. The processing unit 1720 may read the instructions and / or data in the storage unit, thereby causing the apparatus to perform the actions of the device or network element in the method embodiments described above.

[0507] The apparatus 1700 may be a terminal device or a communication device used in or with a terminal device and capable of implementing a communication method executed on the terminal device side. Alternatively, the apparatus 1700 may be a network device or a communication device used in or with a network device and capable of implementing a communication method executed on the network device side.

[0508] When the apparatus 1700 is used in a terminal device, the apparatus 1700 may perform steps or processes performed by the terminal device in the aforementioned method embodiments. The transceiver unit 1710 may be configured to perform transmission / reception-related operations of the terminal device in the aforementioned method embodiments. The processing unit 1720 may be configured to perform processing-related operations of the terminal device in the aforementioned method embodiments.

[0509] When the apparatus 1700 is used in a network device, the apparatus 1700 may perform steps or processes performed by the network device in the aforementioned method embodiments. The transceiver unit 1710 may be configured to perform transmission / reception-related operations of the network device in the aforementioned method embodiments. The processing unit 1720 may be configured to perform processing-related operations of the network device in the aforementioned method embodiments.

[0510] It should be understood that the specific processes of the units performing the aforementioned corresponding steps have been described in detail in the aforementioned method embodiments, and for the sake of brevity, the details will not be described here.

[0511] It should be further understood that the apparatus 1700 herein is embodied in the form of a functional unit. The term "unit" herein may refer to an application-specific integrated circuit (ASIC), an electronic circuit, a processor (e.g., a shared processor, a dedicated processor, or a group processor) configured to execute one or more software or firmware programs, a memory, a merge logic circuit, and / or another suitable component supporting the described functionality. In an optional example, those skilled in the art can understand that the apparatus 1700 may specifically be a terminal device in the aforementioned embodiments and may be configured to perform procedures and / or steps corresponding to the terminal device in the aforementioned method embodiments. Alternatively, the apparatus 1700 may specifically be a network device in the aforementioned embodiments and may be configured to perform procedures and / or steps corresponding to the network device in the aforementioned method embodiments. To avoid repetition, details will not be described again here.

[0512] The apparatus 1700 in the above solutions has a function for implementing corresponding steps performed by a device (e.g., a terminal device, or in another example, a network device) in the above methods. The functions may be implemented by hardware or by hardware executing corresponding software. The hardware or software may include one or more modules corresponding to the above functions. For example, a transceiver unit may be replaced by a transceiver (e.g., a transmitting unit in a transceiver unit may be replaced by a transmitter, and a receiving unit in a transceiver unit may be replaced by a receiver), and another unit, such as a processing unit, may be replaced by a processor to separately perform receiving and transmitting operations and related processing operations in the method embodiments.

[0513] Additionally, the transceiver unit 1710 may alternatively be a transceiver circuit (eg, may include a receiving circuit and a transmitting circuit), and the processing unit may be a processing circuit.

[0514] It should be noted that the apparatus in Figure 13 may be a network element or device in the aforementioned embodiments, or may be a chip or a chip system, such as a system-on-chip (SoC). The transceiver unit may be an input / output circuit or a communication interface. The processing unit is a processor, a microprocessor, or an integrated circuit integrated on a chip. This is not limited herein.

[0515] 14 is a diagram of another communication device 1800 according to an embodiment of the present application. The device 1800 includes a processor 1810. The processor 1810 is configured to execute computer programs or instructions stored in a memory 1820, or to read data or signaling stored in the memory 1820, to perform the method in the above-described method embodiments. Optionally, there are one or more processors 1810.

[0516] Optionally, as shown in Figure 14, the device 1800 further includes a memory 1820 configured to store computer programs or instructions and / or data. The memory 1820 and the processor 1810 may be integrated or may be located separately. Optionally, there are one or more memories 1820.

[0517] 14, the apparatus 1800 further includes a transceiver 1830. The transceiver 1830 is configured to receive and / or transmit signals. For example, the processor 1810 is configured to control the transceiver 1830 to receive and / or transmit signals.

[0518] In one solution, the apparatus 1800 may be used in a terminal device. Specifically, the apparatus 1800 may be a terminal device or a device capable of supporting a terminal device in implementing the functions of the terminal device in any one of the above examples. The apparatus 1800 is configured to implement the operations performed by the terminal device in the above method embodiments.

[0519] For example, the processor 1810 is configured to execute computer programs or instructions stored in the memory 1820 to perform the relevant operations of the terminal device in the above method embodiments.

[0520] In another solution, the apparatus 1800 may be used in a network device. Specifically, the apparatus 1800 may be a network device or may be a device capable of supporting a network device in implementing the functionality of the network device in any one of the aforementioned examples. The apparatus 1800 is configured to implement the operations performed by the network device in the aforementioned method embodiments.

[0521] For example, the processor 1810 is configured to execute computer programs or instructions stored in the memory 1820 to perform the relevant operations of the network device in the method embodiments described above.

[0522] It should be understood that the processor referred to in the embodiments of the present application may be a central processing unit (CPU), or may be another general-purpose processor, a digital signal processor (DSP), an ASIC, a field programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component, etc. The general-purpose processor may be a microprocessor, or the processor may be any conventional processor, etc.

[0523] Furthermore, it should be understood that the memory referred to in the embodiments of the present application may be volatile memory and / or nonvolatile memory. The nonvolatile memory may 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 may be random access memory (RAM). For example, the RAM may be used as an external cache. By way of example and not limitation, RAM includes multiple forms such as static random access memory (static RAM, SRAM), dynamic random access memory (dynamic RAM, DRAM), synchronous dynamic random access memory (synchronous DRAM, SDRAM), double data rate synchronous dynamic random access memory (double data rate SDRAM, DDR SDRAM), enhanced synchronous dynamic random access memory (enhanced SDRAM, ESDRAM), synchlink dynamic random access memory (synchlink DRAM, SLDRAM), and direct rambus random access memory (direct rambus RAM, DR RAM).

[0524] It should be noted that when the processor is a general-purpose processor, a DSP, an ASIC, an FPGA, or another programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component, the memory (storage module) may be integrated into the processor.

[0525] Additionally, it should be noted that memory as referred to herein is intended to comprise, without being limited to, these and any other suitable types of memory.

[0526] An embodiment of the present application further provides a computer-readable storage medium, which stores computer instructions for implementing a method performed by a communication device in an embodiment of the method described above.

[0527] For example, when the computer program is executed by a computer, the computer is enabled to perform the method executed by the terminal device in the above-described method embodiments.

[0528] In another example, a computer program, when executed by a computer, enables the computer to perform the method performed by the network device in the aforementioned method embodiments.

[0529] Certain embodiments of the present application further provide a computer program product including instructions that, when executed by a computer, perform the method performed by a device (e.g., a terminal device or a network device) in the aforementioned method embodiments.

[0530] An embodiment of the present application further provides a communication system including the terminal device and the network device described above, wherein the terminal device and the network device are capable of implementing the communication method described in any one of the above examples.

[0531] Optionally, the system further includes a device that communicates with the terminal device and / or the network device.

[0532] For the description of the relevant contents and beneficial effects in any one of the above-provided devices, please refer to the corresponding method embodiments provided above, and the details will not be described again in this specification.

[0533] In some embodiments provided herein, it should be understood that the disclosed devices and methods may be implemented in other ways. For example, the described device embodiments are merely examples. For example, the division into units is merely a logical division of function, and other divisions may be used in actual implementation. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not implemented. In addition, the shown or discussed mutual couplings or direct couplings or communication connections may be implemented through some interfaces. Indirect couplings or communication connections between devices or units may be implemented in electronic, mechanical, or other forms.

[0534] All or part of the above-described embodiments may be implemented using software, hardware, firmware, or any combination thereof. When software is used to implement an embodiment, all or part of the embodiment may be implemented in the form of a computer program product. A computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, they generate all or part of the procedures or functions according to the embodiments of the present application. The computer may be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. For example, the computer may be a personal computer, a server, a network device, etc. The computer instructions may be stored in a computer-readable storage medium or transmitted from a computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions may be transmitted from a website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber, or digital subscriber line (DSL)) or wireless (e.g., infrared, radio, microwave, etc.) methods. The computer-readable storage medium may be any available medium that can be accessed by a computer, or a data storage device that integrates one or more available media, such as a server or a data center. The available medium may be a magnetic medium (e.g., a floppy disk, a hard disk, or a magnetic tape), an optical medium (e.g., a DVD), a semiconductor medium (e.g., a solid-state drive (SSD)), etc. For example, the available medium may include, but is not limited to, any medium that can store program code, such as a USB flash disk, a removable hard disk, a ROM, a RAM, a magnetic disk, or an optical disk.

[0535] The above description is merely a specific implementation of the present application and is not intended to limit the scope of protection of the present application. Any modifications or replacements that can be easily conceived by those skilled in the art within the technical scope disclosed in the present application shall fall within the scope of protection of the present application. Therefore, the scope of protection of the present application is subject to the scope of protection of the claims.

Claims

1. 1. A communication method comprising: obtaining p groups to be transmitted on a first resource, the p groups belonging to N channel state information (CSI) reports, where p is an integer greater than 1 and N is an integer greater than 1; transmitting some or all of the p groups on the first resource based on priorities of the p groups, wherein the priorities of the p groups are based on priorities of the N CSI reports, the priorities of the CSI reports are based on types of the CSI reports, and the type of at least one of the N CSI reports is an artificial intelligence (AI)-related type. method.

2. 2. The method of claim 1, wherein the priority of the CSI report is based on one or more parameters indicating at least one of: a periodicity of the CSI report; a channel on which the CSI report is carried; whether the CSI report includes a Layer 1 reference signal received power (L1-RSRP) or a Layer 1 signal-to-interference-and-noise ratio (L1-SINR); a serving cell to which the CSI report belongs; or an identity (ID) of the CSI report.

3. 2. The method of claim 1, wherein the priority of the CSI report is based on one or more parameters, at least one of which indicates a type of the CSI report, the type of the CSI report including the AI-related type or the AI-non-related type, and another of which indicates at least one of: a periodicity of the CSI report, a channel on which the CSI report is carried, whether the CSI report includes L1-RSRP or L1-SINR, a serving cell to which the CSI report belongs, or an ID of the CSI report.

4. 4. The method of claim 1, wherein the AI-related type includes one or more of a first CSI report type, a second CSI report type, or a third CSI report type, and report content of the first CSI report type includes channel information acquired through inference by an AI model, report content of the second CSI report type includes performance information of the AI ​​model, or report content of the third CSI report type includes channel information for implementing an operation related to the AI ​​model.

5. 5. The method of claim 4, wherein the N CSI reports include a first CSI report, a type of the first CSI report is the first CSI report type, the first CSI report includes p1 groups among the p groups, and priorities of the p1 groups are based on a priority of the first CSI report and priorities of the p1 groups in the first CSI report, where p1 is an integer greater than 1.

6. 6. The method of claim 5, wherein the channel information acquired through inference by the AI ​​model includes channel information of multiple layers, the p1 groups each include channel information of at least one layer among the multiple layers, and priorities of the p1 groups in the first CSI report match priorities of the channel information of the at least one layer in the p1 groups.

7. 6. The method of claim 5, wherein the channel information acquired through inference by the AI ​​model includes channel information of a plurality of segments, the p1 groups each include channel information of at least one segment among the plurality of segments, and a priority of the p1 groups in the first CSI report matches a priority of the channel information of the at least one segment in the p1 groups.

8. The method of claim 7 , wherein the priority of the at least one segment is related to a layer in which the at least one segment is located and a position of the at least one segment in the layer in which the at least one segment is located.

9. 9. The method of claim 4, wherein the N CSI reports include a second CSI report, a type of the second CSI report is the second CSI report type, the second CSI report includes p2 groups among the p groups, and priorities of the p2 groups are based on a priority of the second CSI report and priorities of the p2 groups in the second CSI report, where p2 is an integer greater than 1.

10. 10. The method of claim 9, wherein the performance information of the AI ​​model includes overall performance information, performance information of the plurality of layers, or performance information of a plurality of subbands, each of the p2 groups includes at least one piece of performance information in the performance information of the AI ​​model, and a priority of the p2 groups in the second CSI report matches a priority of the at least one piece of performance information in the p2 groups.

11. The method of claim 10 , wherein the priority of the overall performance information is higher than the priority of the performance information of the plurality of layers, or the priority of the overall performance information is higher than the priority of the performance information of the plurality of subbands.

12. 12. The method of claim 1, wherein p=N.

13. 13. The method of claim 1, wherein at least two of the N CSI reports have the same priority.

14. The method comprises: receiving first indication information from a network device, the first indication information indicating priorities of some or all of the N CSI reports; 14. The method according to any one of claims 1 to 13.

15. The method of claim 14 , wherein the first indication information is carried in one or more of a radio resource control (RRC) signaling, a medium access control (MAC) control element (CE), or a downlink control information (DCI).

16. 16. The method of claim 1, wherein transmitting some or all of the p groups on the first resource based on the priorities of the p groups comprises selecting a subset of the p groups based on the priorities of the p groups, and transmitting the subset of groups on the first resource.

17. 16. The method of claim 1, wherein transmitting some or all of the p groups on the first resource based on the priorities of the p groups comprises: determining positions of some or all of the p groups on the first resource based on the priorities of the p groups; and transmitting some or all of the p groups on the first resource.

18. 1. A method of communication comprising: obtaining p groups of CSI reports to be transmitted on a first resource, where p is an integer greater than 1; transmitting some or all of the p groups on the first resource based on priorities of the p groups in the CSI report, wherein a type of the CSI report is an AI-related type. method.

19. 19. The method of claim 18, wherein the AI-related type includes one or more of a first CSI report type, a second CSI report type, or a third CSI report type; report content of the first CSI report type includes channel information obtained through inference by an AI model, report content of the second CSI report type includes performance information of the AI ​​model, or report content of the third CSI report type includes channel information for implementing an operation related to the AI ​​model.

20. 20. The method of claim 19, wherein the CSI report type is the first CSI report type, the channel information acquired through inference by the AI ​​model includes channel information of multiple layers, each of the p groups includes channel information of at least one layer among the multiple layers, and priorities of the p groups in the CSI report match priorities of the channel information of the at least one layer in the p groups.

21. 20. The method of claim 19, wherein the channel information acquired through inference by the AI ​​model includes channel information of a plurality of segments, each of the p groups includes channel information of at least one segment among the plurality of segments, and a priority of the p groups in the CSI report matches a priority of the channel information of the at least one segment in the p groups.

22. 22. The method of claim 21, wherein the priority of the at least one segment is related to a layer in which the at least one segment is located and a position of the at least one segment in the layer in which the at least one segment is located.

23. 20. The method of claim 19, wherein the CSI report type is the second CSI report type, the performance information of the AI ​​model includes overall performance information, performance information of the plurality of layers, or performance information of a plurality of subbands, each of the p groups includes at least one piece of performance information among the performance information of the AI ​​model, and a priority of the p groups in the CSI report matches a priority of the at least one piece of performance information among the p groups.

24. 24. The method of claim 23, wherein the priority of the overall performance information is higher than the priority of the performance information of the plurality of layers, or the priority of the overall performance information is higher than the priority of the performance information of the plurality of subbands.

25. 1. A method of communication comprising: receiving q groups on a first resource, the q groups being some or all of p groups, the p groups belonging to N CSI reports, q being a positive integer, p being an integer greater than 1, and N being an integer greater than 1, priorities of the p groups being based on priorities of the N CSI reports, priorities of the CSI reports being based on types of the CSI reports, and a type of the at least one report among the N CSI reports being an AI-related type; communicating with a terminal device based on the q groups; method.

26. 26. The method of claim 25, wherein the priority of the CSI report is based on one or more parameters indicating at least one of: a periodicity of the CSI report; a channel on which the CSI report is carried; whether the CSI report includes a Layer 1 reference signal received power (L1-RSRP) or a Layer 1 signal-to-interference-and-noise ratio (L1-SINR); a serving cell to which the CSI report belongs; or an identity ID of the CSI report, wherein at least one of the one or more parameters further indicates the type of the CSI report, wherein the type of the CSI report comprises the AI-related type or an AI-non-related type.

27. 26. The method of claim 25, wherein the priority of the CSI report is based on one or more parameters, at least one of which indicates a type of the CSI report, wherein the type of the CSI report includes the AI-related type or the AI-non-related type, and another parameter of the one or more parameters includes at least one of a periodicity of the CSI report, a channel on which the CSI report is carried, whether the CSI report includes L1-RSRP or L1-SINR, a serving cell to which the CSI report belongs, or an ID of the CSI report.

28. 28. The method of claim 25, wherein the AI-related type includes one or more of a first CSI report type, a second CSI report type, or a third CSI report type, and report content of the first CSI report type includes channel information obtained through inference by an AI model, report content of the second CSI report type includes performance information of the AI ​​model, or report content of the third CSI report type includes channel information for implementing an operation related to the AI ​​model.

29. 29. The method of claim 28, wherein the N CSI reports include a first CSI report, a type of the first CSI report is the first CSI report type, the first CSI report includes p1 groups among the p groups, and priorities of the p1 groups are based on a priority of the first CSI report and priorities of the p1 groups in the first CSI report, where p1 is an integer greater than 1.

30. 30. The method of claim 29, wherein the channel information obtained through inference by the AI ​​model includes channel information of multiple layers, each of the p1 groups includes channel information of at least one layer among the multiple layers, and a priority of the p1 groups in the first CSI report matches a priority of the channel information of the at least one layer in the p1 groups.

31. 30. The method of claim 29, wherein the channel information obtained through inference by the AI ​​model includes channel information of a plurality of segments, each of the p1 groups includes channel information of at least one segment among the plurality of segments, and a priority of the p1 groups in the first CSI report matches a priority of the channel information of the at least one segment in the p1 groups.

32. 32. The method of claim 31 , wherein the priority of the at least one segment is related to a layer in which the at least one segment is located and a position of the at least one segment in the layer in which the at least one segment is located.

33. 33. The method of claim 28, wherein the N CSI reports include a second CSI report, a type of the second CSI report is the second CSI report type, the second CSI report includes p2 groups among the p groups, and priorities of the p2 groups are based on a priority of the second CSI report and priorities of the p2 groups in the second CSI report, where p2 is an integer greater than 1.

34. 34. The method of claim 33, wherein the performance information of the AI ​​model includes overall performance information, performance information of the plurality of layers, or performance information of a plurality of subbands, each of the p2 groups includes at least one piece of performance information in the performance information of the AI ​​model, and a priority of the p2 groups in the second CSI report matches a priority of the at least one piece of performance information in the p2 groups.

35. 35. The method of claim 34, wherein the priority of the overall performance information is higher than the priority of the performance information of the plurality of layers, or the priority of the overall performance information is higher than the priority of the performance information of the plurality of subbands.

36. 36. The method of any one of claims 25 to 35, wherein p=N.

37. 37. The method of claim 25, wherein at least two of the N CSI reports have the same priority.

38. 38. The method of claim 25, further comprising sending a first indication indicating priorities of some or all of the N CSI reports.

39. 1. A method of communication comprising: transmitting first indication information indicating priorities of some or all of the N CSI reports, where N is an integer greater than 1, and a report type of at least one of the N CSI reports is an AI-related type; receiving some or all of the p groups on a first resource, where some or all of the p groups are based on priorities of the p groups, the p groups belonging to the N CSI reports, and the priorities of the p groups are based on priorities of the N CSI reports, where p is an integer greater than 1; method.

40. 1. A method of communication comprising: receiving first indication information indicating priorities of some or all of N CSI reports, where N is an integer greater than 1, and a report type of at least one of the N CSI reports is an AI-related type; transmitting some or all of the p groups on a first resource, where some or all of the p groups are based on priorities of the p groups, the p groups belonging to the N CSI reports, and the priorities of the p groups are based on priorities of the N CSI reports, where p is an integer greater than 1. method.

41. 41. The method of claim 39 or 40, wherein the first indication information is carried in one or more of RRC signaling, MAC CE, or DCI.

42. 42. The method of claim 39, wherein at least two of the N CSI reports have the same priority.

43. 43. The method of claim 39, wherein the priority of the CSI report is based on a type of the CSI report, the type of the CSI report comprising the AI-related type or an AI-non-related type.

44. 44. The method of claim 43, wherein the AI-related type includes one or more of a first CSI report type, a second CSI report type, or a third CSI report type, and report content of the first CSI report type includes channel information obtained through inference by an AI model, report content of the second CSI report type includes performance information of the AI ​​model, or report content of the third CSI report type includes channel information for implementing an operation related to the AI ​​model.

45. A communications device comprising a unit or module configured to perform the method of any one of claims 1 to 17, 18 to 24, 25 to 38 or 39 to 44.

46. 10. A computer-readable storage medium comprising instructions that, when executed by a processor, perform a method according to any one of claims 1 to 17, 18 to 24, 25 to 38, or 39 to 44.

47. 10. A communications device comprising: a processor; the processor coupled to a storage medium; the storage medium storing instructions that, when executed by the processor, perform a method according to any one of claims 1 to 17, 18 to 24, 25 to 38, or 39 to 44.

Images