Communication method and communication apparatus

By defining resource counts and activation intervals, terminal devices and network devices align with the storage requirements of CSI reports, solving the problem of how terminal devices can efficiently store AI-related CSI reports and improving resource utilization.

WO2026233366A1PCT designated stage Publication Date: 2026-11-12HUAWEI TECH CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
PCT/CN2026/095195
Authority / Receiving Office
WO · WO
Patent Type
Applications
Current Assignee / Owner
Priority Date
2025-05-09
Filing Date
2026-05-07
Publication Date
2026-11-12

AI Technical Summary

Technical Problem

How can terminal devices efficiently store channel state information (CSI) reports related to artificial intelligence, improve resource utilization, and solve the problem of resource waste in existing technologies?

Method used

By defining the count and activation range of resources, terminal devices and network devices align the storage requirements of CSI reports, avoiding resource overlap and improving resource utilization.

Benefits of technology

By aligning resource storage needs, resource waste is reduced, resource utilization is improved, and terminal devices are ensured to efficiently store AI-related CSI reports.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN2026095195_12112026_PF_FP_ABST
    Figure CN2026095195_12112026_PF_FP_ABST
Patent Text Reader

Abstract

Provided in the present application are a communication method and a communication apparatus. The communication method is applied to a first device or a chip of the first device. The method comprises: determining a count of a first resource and / or a first activation interval corresponding to the first resource, wherein the first resource corresponds to a first inference report, or the first resource corresponds to a configuration of the first inference report, or the first resource corresponds to an inference result in the first inference report, or the first resource corresponds to a first monitoring report associated with the first inference report, or the first resource corresponds to a configuration of the first monitoring report associated with the first inference report; and the count of the first resource is A, with A being a natural number greater than or equal to 1. The present application enables a terminal device and a network device to align resources required for storing an AI-related CSI report, as well as a storage duration required therefor, thereby further improving the resource utilization efficiency.
Need to check novelty before this filing date? Find Prior Art

Description

Communication method and communication apparatus

[0001] The present application claims priority to the Chinese patent application No. 202510602354.1, filed on May 9, 2025, and entitled "Communication method and communication apparatus", the content of which is incorporated herein by reference in its entirety. TECHNICAL FIELD

[0002] The present application relates to the field of wireless communication, and in particular to a communication method and a communication apparatus. BACKGROUND

[0003] At present, artificial intelligence (AI) is introduced into wireless communication networks, and has been widely applied to many application scenarios of air interface technology, for example, AI-based channel state information (CSI) prediction, AI-based beam management, and AI-based CSI feedback, and plays an increasingly important role.

[0004] Therefore, how the terminal device stores the CSI report related to AI is a technical problem that needs to be solved by those skilled in the art. SUMMARY

[0005] The present application provides a communication method and a communication apparatus, so that the terminal device and the network device can align the required resources and the required storage duration for storing the CSI report related to AI.

[0006] In a first aspect, a communication method is provided, which can be applied to a first device. The first device can be a terminal device, or can be a component deployed in the terminal device, such as a circuit or a chip (e.g., a modem chip, also known as a baseband chip, or a system on chip (SoC) chip or a system in package (SIP) chip containing a modem core, or a chip for AI or a SoC or SIP chip containing a chip for AI, etc.) inside the terminal device, etc. Alternatively, the method can be applied to a logic module or software, etc. capable of realizing all or part of the functions of the communication device. For the convenience of understanding and description, the terminal device is taken as an example of the first device in the following description of the method.

[0007] Exemplarily, the method comprises determining the count of the first resource and / or the first activation interval corresponding to the first resource. The first resource corresponds to the first inference report, or the first resource corresponds to the configuration of the first inference report, or the first resource corresponds to one inference result in the first inference report, or the first resource corresponds to the first monitoring report associated with the first inference report, or the first resource corresponds to the configuration of the first monitoring report associated with the first inference report. The first resource is counted A times, where A is a natural number greater than or equal to 1.

[0008] Based on the above scheme, by determining the count of the first resource and / or the first activation interval corresponding to the first resource, the terminal device and the network device align the resources required for storing the first inference report and the storage time required, further improving the resource utilization.

[0009] It should be noted that the inference result refers to the information carried by the CSI report / UE report based on AI inference, which can be replaced by at least one of the following: inference information, prediction information, prediction result, prediction instance, result obtained based on an AI model, prediction instance (instance), prediction time instance (time instance), inference information corresponding to a prediction instance, etc. Different AI features are inconsistent, for example, predicted precoding matrix indicator (PMI), predicted reference signal received power (RSRP), predicted Top1 beam, etc.

[0010] It should be noted that the inference report refers to the CSI report / UE report corresponding to the carrying of the above inference result.

[0011] It should be noted that the monitoring report corresponds to the CSI report / UE report carrying the above monitoring result.

[0012] It should be noted that the inference in the present application can be replaced / understood as prediction.

[0013] It should be noted that the first resource corresponding to the first inference report can be understood or replaced as at least one of the following: the first resource corresponding to one first inference report, the first resource corresponding to one potential CSI report sending, and the first resource corresponding to one potential uplink resource. That is, resource counting or storage is performed for each inference report.

[0014] It should be noted that the configuration of the first resource corresponding to the first inference report can be understood or replaced as one or more first inference reports associated with the configuration of the first resource corresponding to the first inference report, wherein the time domain type of the report in the configuration of the first inference report is P or SP. That is, resource counting or storage is performed for one or more inference reports corresponding to the configuration of each inference report.

[0015] It should be noted that one inference report from an SP / P corresponds to one configuration information; one inference report from an AP corresponds to one or more configuration information. Similarly, one monitoring report from an SP / P corresponds to one configuration information; one monitoring report from an AP corresponds to one or more configuration information.

[0016] Compared to the configuration where the first resource corresponds to the first inference report, in the case where the first resource corresponds to the first inference report, since the UE will count or store resources for each inference report, there is no problem of storage overlap in the time dimension for multiple inference reports corresponding to one inference report configuration.

[0017] It should be noted that the first resource corresponding to a reasoning result in the first reasoning report can be understood or replaced by at least one of the following: the first resource corresponds to any reasoning result in the first reasoning report, the first resource corresponds to a reasoning result, or the first resource corresponds to a single reasoning result. That is, resource counting or storage is performed for each reasoning result in each reasoning report, or resource counting or storage is performed for each reasoning result.

[0018] It should be noted that the first resource corresponding to the first monitoring report associated with the first inference report can be understood or replaced with at least one of the following: the first resource corresponds to the first monitoring report. That is, resource counting or storage is performed for each monitoring report.

[0019] It should be noted that the configuration of the first resource corresponding to the first monitoring report associated with the first inference report can be understood or replaced by at least one of the following: the configuration of the first resource corresponding to the first monitoring report. That is, configuring one or more monitoring reports corresponding to each monitoring report for resource counting or storage.

[0020] For example, the method includes determining a count of resources of a first type and / or a first activation interval corresponding to the first resource. The first resource belongs to the first type of resource. Specifically, the first resource corresponds to a first inference report, or the first resource corresponds to configuration information of a first inference report, or the first resource corresponds to an inference result in a first inference report, or the first resource corresponds to a first monitoring report associated with the first inference report, or the first resource corresponds to configuration information of a first monitoring report associated with the first inference report. The count of resources of the first type is based on resource #0 as the granularity, and the first resource includes A resources #0, where A is a natural number greater than or equal to 1.

[0021] It should be noted that a reasoning report includes one or more reasoning results.

[0022] It should be noted that the first resource includes A resources #0, which can be understood as the first resource being counted as A.

[0023] In conjunction with the first aspect, in some possible implementations of the first aspect, the first inference report corresponds to one or more monitoring reports.

[0024] In conjunction with the first aspect, in some possible implementations of the first aspect, the type of the first resource is a channel state information reference signal resource or a corresponding channel state information reference signal resource. The first activation interval corresponds to the activation interval of the channel state information reference signal resource; or the type of the first resource is a storage unit or a corresponding storage unit, and the first activation interval corresponds to the activation interval of the storage unit.

[0025] For example, in conjunction with the first aspect, in some possible implementations of the first aspect, the first type of resource is a channel state information reference signal resource, and the first activation interval corresponds to the activation interval of the channel state information reference signal resource. Alternatively, the first type of resource may correspond to a storage unit, and the first activation interval corresponds to the activation interval of the storage unit.

[0026] In conjunction with the first aspect, in some possible implementations of the first aspect, the start time of the first active interval includes a first start time. The first start time includes one of the following: the time unit corresponding to the first Radio Resource Control (RRC) configuration, the reporting time of the first inference report, the time unit where the Channel State Information (CSI) reference resource of the first inference report is located, the trigger time of the first inference report, and the trigger time of the first monitoring report. The first RRC configuration is used to configure the first inference report, or to configure the first inference report to be associated with one or more monitoring reports.

[0027] It should be noted that the reporting time of the first inference report can be understood or replaced by at least one of the following: the time unit in which the uplink resource carrying the first inference report is located, or the uplink time unit in which the first inference report is located, or the first or last symbol of the PUSCH or PUCCH used to carry the first inference report, etc. This application does not impose any special restrictions on this.

[0028] It should be noted that the CSI reference resource corresponding to the first inference report can be understood or replaced as the CSI reference resource corresponding to the first inference report, or the reference resource corresponding to the first inference report, and this application does not impose any special restrictions on this.

[0029] It should be noted that in the implementation of this application, "time" can be replaced with "time unit," which can be understood as any one or more of the following: slot, subframe, frame, or orthogonal frequency division multiplexing (OFDM) symbol. For example, the time corresponding to A represents the slot, subframe, frame, or OFDM symbol in which A is located, or the first slot, subframe, frame, or OFDM symbol in which A is located, or the last slot, subframe, frame, or OFDM symbol in which A is located.

[0030] In conjunction with the first aspect, in some possible implementations of the first aspect, the end time of the first activation interval is the first end time. The first end time includes one of the following: the time of releasing the first RRC configuration, the end time of the effect of the inference result corresponding to the first inference report, the end time of the prediction window corresponding to the first inference report, the end time of the effect of one or more prediction results corresponding to the first inference report, the completion time of the performance index calculation of the first monitoring report, the reporting time of the first monitoring report, and the time unit where the CSI reference resource of the first monitoring report is located. The first RRC configuration is used to configure the first inference report, or to configure the first inference report to be associated with one or more monitoring reports.

[0031] It should be noted that the end time of the effect of the inference result corresponding to the first inference report can be understood or replaced by at least one of the following: the end time of the prediction window corresponding to the first inference report, or the prediction start time corresponding to the first inference report + the number of prediction results * the prediction result interval. This application does not impose any special restrictions on this.

[0032] It should be noted that the end time of the effect of the inference result corresponding to the first inference report can be replaced by the end time of the effect of all inference results corresponding to the first inference report. For example, if an inference report includes N4 inference results, the first end time can be the end time of the effect of the N5th inference result, or the first time interval after the end time of the effect of the N5th inference result, where N5 is less than or equal to N4, and can be indicated by the UE. Understandably, in the above scheme, the UE indicates the number of inference results that can be cached for a single inference report.

[0033] It should be noted that the calculation time of the performance metric in the first monitoring report can be understood or replaced by at least one of the following: the completion time of the performance metric calculation in the first monitoring report, or Z' symbols of the last symbol of the CSI-RS occasion for channel measurement corresponding to the first monitoring report, wherein Z' can be predefined by the protocol or indicated by the UE. It is understood that Z' includes the time required to calculate the performance metric.

[0034] It should be noted that the reporting time of the first monitoring report can be understood or replaced by at least one of the following: the time unit in which the uplink resource carrying the first monitoring report is located, or the uplink time unit in which the first monitoring report is located, or the first or last symbol of the PUSCH or PUCCH used to carry the first monitoring report, etc. This application does not impose any special restrictions on this.

[0035] It should be noted that the CSI reference resource corresponding to the first monitoring report can be understood or replaced as the CSI reference resource corresponding to the first monitoring report, or the reference resource corresponding to the first monitoring report. This application does not impose any special restrictions on this.

[0036] It should be noted that the metrics in the first monitoring report include performance metrics. In this application, performance metrics refer to the monitoring results, i.e., the performance metrics calculated based on measurement and inference results. These metrics vary for different AI features, such as square generalized cosine similarity (SGCS), normalized mean square error (NMSE), and Top-1 accuracy. The measurement results can be replaced by at least one of the following: monitoring occasion, CSI-RS transmission occasion for performance monitoring, monitoring time instance, monitoring instance, or measurement instance for performance monitoring. The CSI-RS transmission occasion refers to the time of CSI-RS resource transmission; for beam management (BM) features, multiple CSI-RS resources are typically transmitted within a single CSI-RS occasion; for CSI features, only one CSI-RS resource is typically transmitted within a single CSI-RS occasion, and this is not limited here.

[0037] In conjunction with the first aspect, in some possible implementations of the first aspect, the end time of the first activation interval is a first time offset after the first start time. The first time offset is indicated by the first device, or the first time offset includes a predefined time offset.

[0038] In conjunction with the first aspect, in some possible implementations of the first aspect, the end time of the first activation interval is a first end moment, and there is a first time offset between the first end moment and the first moment. The first time offset is indicated by the first device, or the first time offset includes a predefined time offset. The first moment includes one of the following: the time of releasing the first RRC configuration, the end time of the effect of the inference result corresponding to the first inference report, the end time of the prediction window corresponding to the first inference report, the end time of the effect of one or more prediction results corresponding to the first inference report, the completion time of the performance index calculation of the first monitoring report, the reporting time of the first monitoring report, and the time unit where the CSI reference resource of the first monitoring report is located. The first RRC configuration is used to configure the first inference report, or to configure the first inference report to be associated with one or more monitoring reports.

[0039] In conjunction with the first aspect, in some possible implementations of the first aspect, the reporting time of the first monitoring report is later than the first moment, and the end time of the first activation interval is the reporting time of the first monitoring report. The first moment includes the following: the time of releasing the first RRC configuration, the end time of the effect of the inference result corresponding to the first inference report, the end time of the prediction window corresponding to the first inference report, the end time of the effect of one or more prediction results corresponding to the first inference report, the completion time of the calculation of the monitoring performance index of the first monitoring report, and the time unit in which the CSI reference resource of the first monitoring report is located.

[0040] In conjunction with the first aspect, in some possible implementations of the first aspect, the method further includes sending first indication information. The first indication information is used to indicate A.

[0041] It should be noted that the first indication information can be understood as the UE's capability information, which is used to indicate the UE's capabilities.

[0042] In conjunction with the first aspect, in some possible implementations of the first aspect, the first activation intervals of the first resources corresponding to at least two inference reports or at least two inference results correspond to the same time unit within the first interval, and the first resources within the first interval are not counted repeatedly, or are counted as 1.

[0043] It should be noted that the first resource within the first interval is not counted repeatedly, which can be understood as the count of the first resource within the first interval being independent of the number of overlaps.

[0044] By ensuring that overlapping resources within an interval are not counted repeatedly, and by determining inference reports and / or monitoring reports based on the latest inference results, resource waste is reduced and resource utilization is further improved.

[0045] It should be noted that at least two inference reports can be understood as one or more first inference reports associated with the configuration of the first resource corresponding to the first inference report, wherein the time domain type of the report in the configuration of the first inference report is P or SP. That is, resource counting or storage is performed for one or more inference reports corresponding to the configuration of each inference report.

[0046] In conjunction with the first aspect, in some possible implementations of the first aspect, the first activation intervals of the first resources corresponding to at least two inference reports or at least two inference results correspond to the same time unit within the first interval, the first resources are repeatedly counted within the first interval, and the count of the first resources is correlated with the number of overlaps between the inference reports or inference results within the first interval. Specifically, at least two inference reports correspond to the same inference report configuration, or at least two inference results correspond to the same inference report configuration.

[0047] In conjunction with the first aspect, in some possible implementations of the first aspect, determining the count of the first resource includes determining the count of the first resource within the first time unit.

[0048] For example, in conjunction with the first aspect, in some possible implementations of the first aspect, determining the count of resources of the first type includes determining the count of resources of the first type within a first time unit.

[0049] In conjunction with the first aspect, in some possible implementations of the first aspect, determining the count of the first resource includes: the time of the first downlink control information being earlier than the reporting time of the first inference report, and the first resource being counted A times within the first activation interval. And / or, the time of the first downlink control information being later than the reporting time of the first inference report, and the first resource being counted 0 times within the first activation interval, or not counted, or the first resource being inactive or in a deactivated state within the first activation interval. Wherein, the first downlink control information is used to trigger the first monitoring report.

[0050] For example, in conjunction with the first aspect, in some possible implementations of the first aspect, determining the count of the first type of resource includes: the time of the first downlink control information being earlier than the reporting time of the first inference report, and the first resource being counted A times within the first activation interval. And / or, the time of the first downlink control information being later than the reporting time of the first inference report, and the first resource being counted 0 times within the first activation interval, or not counted, or the first resource being inactive or in a deactivated state within the first activation interval. Wherein, the first downlink control information is used to trigger the first monitoring report.

[0051] In conjunction with the first aspect, in some possible implementations of the first aspect, the first resource is a first inference resource and / or a first monitoring resource. Specifically, the first inference resource is the measurement resource corresponding to the first inference report, and the first monitoring resource is the measurement resource corresponding to the first monitoring report.

[0052] In conjunction with the first aspect, in some possible implementations of the first aspect, the first resource is a first inference resource. The first resource also additionally counts Kp times. Kp is related to the configuration of the first inference report, and / or Kp is indicated by the first device.

[0053] In conjunction with the first aspect, in some possible implementations of the first aspect, the first inference report is associated with one or more monitoring reports, and the first resource is not counted repeatedly.

[0054] In conjunction with the first aspect, in some possible implementations of the first aspect, the time-domain type of the first inference report is one or more of periodic, semi-periodic, or aperiodic.

[0055] In conjunction with the first aspect, in some possible implementations of the first aspect, the first indication information is also used to indicate one or more of the following parameters: the maximum number of supported storage units, and the maximum supported storage time. Wherein, a storage unit corresponds to the storage of the inference report, or a storage unit corresponds to the storage of the inference result.

[0056] Secondly, a communication method is provided. This method can be applied to a second device, which can be a terminal device or a component deployed within the terminal device, such as circuits or chips inside the terminal device (e.g., modem chips, also known as baseband chips, or system-on-chip (SoC) chips or system-in-package (SIP) chips containing modem cores, or chips for AI or SoC or SIP chips containing chips for AI, etc.); or it can be applied to logic modules or software capable of implementing all or part of the functions of the communication device, etc. For ease of understanding and explanation, the following description uses a terminal device as an example of a second device to illustrate this method.

[0057] The second device in the second aspect may be the same device as the first device in the first aspect, or it may be a different device.

[0058] For example, the method includes the following steps: First available processing resources are not used for the occupancy of a first Channel State Information (CSI) report, and second available processing resources are used for the occupancy of a second CSI report. The processing of the first CSI report corresponds to the occupancy of the first processing resources, and the processing of the second CSI report corresponds to the occupancy of the second processing resources. The first processing resources are not used for the processing of the second CSI report.

[0059] It should be noted that the available first resource can be understood as the first resource that is not currently occupied on the symbol.

[0060] It should be noted that the first processing resource is not used for processing the second CSI report; in other words, the second CSI report does not need to occupy the first processing resource.

[0061] It should be noted that the first processing resources are not used for processing the second CSI report. It can also be understood that the second device processes or updates the second CSI report based on the amount of second processing resources required by the second CSI report.

[0062] It should be noted that the second processing resource is used for processing the second CSI report.

[0063] For example, the method includes: if the available first processing resources do not meet the occupancy of a first Channel State Information (CSI) report, the available second processing resources are not used for the occupancy of a second CSI report, regardless of whether the available second processing resources meet the occupancy of a second CSI report. Wherein, the processing of the first CSI report corresponds to the occupancy of the first processing resources, and the processing of the second CSI report corresponds to the occupancy of the second processing resources. The first processing resources are not used for the processing of the second CSI report.

[0064] It should be noted that the first processing resource is not used for processing the second CSI report, which can also be understood as the second CSI report not needing to occupy the first processing resource.

[0065] In conjunction with the second aspect, in some implementations of the second aspect, the processing of the first CSI report also corresponds to the occupation of the second processing resources.

[0066] In conjunction with the second aspect, in some implementations of the second aspect, at least one second CSI report has a lower priority than at least one first CSI report.

[0067] In conjunction with the second aspect, in some implementations of the second aspect, the first processing resource includes the resources corresponding to the channel state information processing unit or the resources corresponding to the channel state information reporting processing unit based on artificial intelligence / machine learning. The second processing resource includes the resources corresponding to the channel state information reporting processing unit based on artificial intelligence / machine learning or the resources corresponding to the channel state information processing unit. It should be noted that, in a specific implementation, the first processing resource includes the resources corresponding to the channel state information processing unit, and the second processing resource includes the resources corresponding to the channel state information reporting processing unit based on artificial intelligence / machine learning.

[0068] It should be noted that, in another specific implementation, the first processing resource includes the resources corresponding to the channel state information reporting (CSI) unit based on artificial intelligence / machine learning (AI / ML), and the second processing resource includes the resources corresponding to the channel state information (CSI) unit. This can also be understood as the first processing resource primarily being an AI / ML-based CSI processing unit, while the second processing resource is a non-AI / ML-based CSI processing unit. As an example, the first processing resource could be an AI / ML-based CSI reporting unit (hereinafter referred to as an APU), and the second processing resource could be a CPU. As another example, the first processing resource is a CPU, and the second processing resource is an APU.

[0069] It should be noted that the APU can be extended to: CSI processing unit, CPU_ML, AI-based CSI processing unit, processing unit for measurement information carried in CSI reports, processing unit for AI information carried in CSI reports, and processing unit for information carried in AI-based CSI reports.

[0070] It should be noted that AI information can be information related to AI inference, AI training, or AI monitoring; this application does not limit the use cases of AI, which can be AI-based CSI prediction, AI-based BM (beam management or beam prediction), AI-based CSI compression, etc.

[0071] It should be noted that when the first processing resources include the resources corresponding to the channel state information reporting processing unit based on artificial intelligence / machine learning, the first CSI report includes the AI-based CSI report; when the second processing resources include the resources corresponding to the channel state information reporting processing unit based on artificial intelligence / machine learning, the second CSI report includes the AI-based CSI report.

[0072] It should be noted that AI-based CSI reports can be understood as / replaced by CSI reports carrying AI-related information. This AI-related information can be information related to AI inference (e.g., predicted Top1 beam identity, predicted top K beam IDs, predicted RSRP, predicted CSI, predicted codebook, etc.) and information related to AI monitoring (e.g., measured Top1 beam ID, measured Top1 beam ID, predicted Top1 beam accuracy, predicted Top K beam accuracy, measured CSI, measured codebook, and performance metrics (e.g., SGCS / NMSE of predicted and measured channel matrices, SGCS / NMSE of predicted and measured precoding matrices)).

[0073] It should be noted that an AI-based CSI report can be understood as / replaced with a CSI report that carries AI-related information. The AI-related information can be understood as information that needs to be obtained through AI model inference or other AI model processing. For example, the AI-related information can be AI inference-related information (such as predicted Top1 beam ID, predicted TopK beam ID, predicted RSRP, predicted CSI, predicted codebook, etc.).

[0074] It should be noted that the CSI report used to carry AI-related information can be understood as follows: the report quantity in the corresponding report configuration (CSI-ReportConfig IE) of the CSI report is configured to be the AI-related information.

[0075] It should be noted that this application does not restrict AI inference information or AI monitoring information; this application does not restrict AI use cases, which can be AI-based CSI prediction, AI-based BM (beam management or beam prediction), AI-based CSI compression, etc.

[0076] Thirdly, a communication method is provided that can be applied to a third device, which can be a terminal device or a component deployed within the terminal device, such as circuits or chips inside the terminal device (e.g., a modem chip, or a SoC chip or SIP chip containing a modem core, or a chip for AI or a SoC or SIP chip containing a chip for AI, etc.); or it can be applied to logic modules or software capable of implementing all or part of the functions of the terminal device, etc. For ease of understanding and explanation, the following description uses a terminal device as an example of a third device to illustrate the method.

[0077] The third device in the third aspect and the first device in the first aspect may be the same device or different devices.

[0078] For example, the method includes determining a count value of a third processing resource corresponding to a second monitoring report, based on the second monitoring report. The second monitoring report is associated with one or more inference reports.

[0079] It should be noted that the third processing resource can be a CPU or an APU, and this application does not impose any special restrictions on it.

[0080] In conjunction with the third aspect, in some implementations of the third aspect, the count value of the third processing resource corresponding to the second monitoring report is correlated with the count value of the third processing resource corresponding to the first measurement resource and / or the first monitoring factor. The first measurement resource is used for channel measurements corresponding to the second monitoring report.

[0081] In conjunction with the third aspect, in some implementations of the third aspect, the count value of the third processing resource corresponding to the second monitoring report is also associated with the first count value. The first count value is indicated by the first device, or the first count value includes a predefined count value, or the first count value is associated with at least one of the following: inference result inverse calculation, monitoring inverse calculation, and squared generalized cosine similarity (SGCS) calculation.

[0082] In conjunction with the third aspect, in some implementations of the third aspect, the first monitoring factor is indicated by the first device, or the first monitoring factor includes predefined monitoring factors, or the first monitoring factor has a corresponding relationship with the first count value.

[0083] Fourthly, an apparatus is provided. This apparatus may include functional modules corresponding to each of the methods / operations / steps / actions described in any one of the first to third aspects, and in any possible implementation of any one aspect. The module may be a hardware circuit, software, or a combination of hardware circuitry and software implementation.

[0084] In one design, the device may include a processing module and a communication module. The communication module is used to perform the sending and receiving actions performed by the first device as described in the first aspect and any possible implementation thereof, while the processing module is used to perform the processing-related actions performed by the first device in the methods described in the first aspect and any possible implementation thereof.

[0085] In one design, the device may include a processing module and a communication module. The communication module is used to perform the sending and receiving actions performed by the second device as described in the second aspect and any possible implementation thereof, while the processing module is used to perform the processing-related actions performed by the second device in the methods described in the second aspect and any possible implementation thereof.

[0086] In one design, the device may include a processing module and a communication module. The communication module is used to perform the sending and receiving actions performed by the third device as described in the third aspect and any possible implementation thereof, while the processing module is used to perform the processing-related actions performed by the third device in the methods described in the third aspect and any possible implementation thereof.

[0087] In one design, the device can be a terminal device, or a device, module, circuit or chip configured in the terminal device, or a device that can be used in conjunction with the terminal device, such as a host or cloud server of an over-the-top (OTT) system.

[0088] In one design, the device can be a network device, or a device, module, circuit, or chip configured in the network device, or a device that can be used in conjunction with the network device, such as an intelligent network element with a radio access network (RAN) intelligent controller (RIC) deployed thereon.

[0089] Fifthly, an apparatus is provided, including a processor and a storage medium storing instructions that, when executed by the processor, cause a method as described in any of the first to third aspects, and in any possible implementation thereof, to be implemented.

[0090] A sixth aspect provides an apparatus including a processing circuit for processing data and / or information such that a method as described in any of the first to third aspects, and in any possible implementation thereof, is implemented.

[0091] The processing circuit may include one or more processors, or all or part of the circuitry in one or more processors used for control or processing functions.

[0092] Optionally, the apparatus may further include a memory for storing programs or instructions, and the processor for running the programs or instructions to implement the methods of any of the first to third aspects, and any possible implementation thereof.

[0093] Optionally, the device may also include the transceiver circuit, or an input / output interface.

[0094] In a seventh aspect, a chip is provided, including processing circuitry for running a program or instructions to implement a method as described in any of the first to third aspects, and in any possible implementation thereof.

[0095] Optionally, the chip may further include a memory for storing programs or instructions.

[0096] Optionally, the chip may also include transceiver circuitry, or input / output interfaces.

[0097] Eighthly, a computer-readable storage medium is provided, the computer-readable storage medium including instructions that, when executed by a processor, cause the method of any one of the first to third aspects, and any possible implementation thereof, to be implemented.

[0098] Ninth aspect, a computer program product is provided, the computer program product including computer program code or instructions, which, when executed, cause a method as described in any of the first to third aspects, and in any possible implementation of any of the aspects, to be implemented.

[0099] It should be understood that the fourth to ninth aspects of this application correspond to the technical solutions of the first to third aspects of this application, and the beneficial effects obtained by each aspect and the corresponding feasible implementation are similar, and will not be repeated here. Attached Figure Description

[0100] Figure 1 is a schematic diagram of a communication system applicable to the communication method of this application embodiment.

[0101] Figure 2 is a schematic diagram of another communication system applicable to the communication method of the embodiments of this application.

[0102] Figure 3 is a schematic diagram of a possible application framework in a communication system.

[0103] Figure 4 is a schematic diagram of another possible application framework in a communication system.

[0104] Figure 5 illustrates the inference process of CSI prediction.

[0105] Figure 6 shows a schematic diagram of the inference process of CSI compression.

[0106] Figure 7 illustrates a schematic diagram of the reasoning process for beam management.

[0107] Figure 8 is a schematic diagram of CSI calculation time.

[0108] Figure 9 shows a schematic diagram of an AI / ML functional framework for wireless air interfaces.

[0109] Figure 10 is a schematic flowchart of a communication method 800 provided in an embodiment of this application. The steps of method 800 are described in detail below.

[0110] Figure 11 is a schematic diagram of determining resource counts for overlapping intervals according to an embodiment of this application.

[0111] Figure 12 is a schematic diagram of determining a resource activation range according to an embodiment of this application.

[0112] Figure 13 is a schematic diagram of determining the resource activation range of a P / SP inference report according to an embodiment of this application.

[0113] Figure 14 is a schematic diagram of determining the resource activation range of an AP inference report according to an embodiment of this application.

[0114] Figure 15 is a schematic diagram of another method for determining the resource activation range of the AP inference report provided in an embodiment of this application.

[0115] Figure 16 is a schematic diagram of another method for determining the resource activation range provided in an embodiment of this application.

[0116] Figure 17 is a schematic diagram of the performance monitoring steps provided in an embodiment of this application.

[0117] Figure 18 is a schematic diagram of the performance monitoring steps provided in an embodiment of this application.

[0118] Figure 19 is a schematic diagram of another performance monitoring step provided in an embodiment of this application.

[0119] Figure 20 is a schematic block diagram of the device provided in an embodiment of this application.

[0120] Figure 21 is a schematic diagram of the structure of a communication device provided in another embodiment of this application.

[0121] Figure 22 is a schematic diagram of the reasoning result storage and determination module provided in an embodiment of this application.

[0122] Figure 23 is a schematic block diagram of the AI ​​processor provided in an embodiment of this application. Detailed Implementation

[0123] The technical solutions in this application will now be described with reference to the accompanying drawings.

[0124] To facilitate understanding of the embodiments of this application, the following points will be explained first:

[0125] First, in this application, the terminal side can also be referred to as the UE (user equipment) side, including: terminal equipment, components deployed in the terminal equipment (such as circuits or chips inside the terminal equipment), equipment deployed outside the terminal equipment (such as over-the-top (OTT) hosts or cloud servers), or components deployed in equipment outside the terminal equipment (such as circuits or chips inside the equipment). The network (NW) side includes: network equipment communicating with the terminal equipment, components deployed in the network equipment (such as circuits or chips inside the network equipment with near real-time radio access network (RAN) intelligent control functions), equipment deployed outside the network equipment (such as intelligent network elements, for example, intelligent network elements with near real-time RAN intelligent control functions), or components deployed in the intelligent network element (such as circuits or chips inside the intelligent network element). The network equipment may include: access network equipment, core network equipment, or operation administration and maintenance (OAM) equipment.

[0126] Second, in this application, the indication includes direct indication (also known as explicit indication) and indirect indication (also known as implicit indication). Directly indicating information A means including information A; indirectly indicating information A can mean indicating information A through the correspondence between information A and information B and by directly indicating information B; or by indicating information A through a preset rule that can be used to determine A based on B and by directly indicating information B. The correspondence between information A and information B, and the preset rule, can be predefined, pre-stored, pre-burned, or pre-configured.

[0127] Third, in this application, "at least one" means one or more, and "more than one" means two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can mean: A alone, A and B simultaneously, or B alone, where A and B can be singular or plural. The character " / " generally indicates an "or" relationship between the preceding and following related objects, but it does not exclude the possibility of indicating an "and" relationship; the specific meaning can be understood in context. "At least one of the following" or similar expressions refer to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can mean: a, b, c; a and b; a and c; b and c; or a and b and c. Here, a, b, and c can be single or multiple.

[0128] Fourth, the use of prefixes such as "first" and "second" in this application is merely for the purpose of distinguishing and describing different things belonging to the same name category, and does not constrain the order, size, or quantity of things. For example, "first resource" and "second resource" are simply different resources, and do not limit the quantity of resources, the order of transmission, or the relationship of priority.

[0129] Fifth, in this application, "send" and "receive" indicate the direction of signal transmission. For example, "send information to XX" can be understood as the destination of the information being XX, which can include direct transmission via the air interface or indirect transmission by other units or modules via the air interface. "Receive information from YY" can be understood as the source of the information being YY, which can include direct reception from YY via the air interface or indirect reception from YY by other units or modules via the air interface. "Send" can also be understood as the "output" of the chip interface, and "receive" can also be understood as the "input" of the chip interface. In other words, sending and receiving can occur between devices, such as between a terminal device and a computing node, or within a device, such as between components, modules, chips, software modules, or hardware modules within the device via a bus, wiring, or interface.

[0130] Sixth, in the embodiments of this application, "when," "if," and "if" all refer to the device making corresponding processing under certain objective circumstances, and are not limited to a time, nor do they require the device to make a judgment action when it is implemented, nor do they mean that there are other limitations.

[0131] Seventh, in this application, the words "example," "exemplary," "for example," or "likely" are used to indicate that something is an example, illustration, or description. Any embodiment or design described as "example," "exemplary," "for example," or "likely" in this application should not be construed as being more preferred or advantageous than other embodiments or designs. Specifically, the use of the words "example," "exemplary," "for example," or "likely" is intended to present the relevant concepts in a specific manner.

[0132] The technical solutions provided in this application can be applied to various communication systems, such as: 5th generation (5G) or new radio (NR) systems, long term evolution (LTE) systems, LTE frequency division duplex (FDD) systems, LTE time division duplex (TDD) systems, wireless local area network (WLAN) systems, satellite communication systems, future communication systems, or integrated systems of multiple systems. The technical solutions provided in this application can also be applied to device-to-device (D2D) communication, vehicle-to-everything (V2X) communication, machine-to-machine (M2M) communication, machine-type communication (MTC), and Internet of Things (IoT) communication systems or other communication systems.

[0133] In a communication system, one network element can send signals to or receive signals from another network element. These signals can include information, signaling, or data. The term "network element" can also be replaced by an entity, network entity, device, communication equipment, communication module, node, communication node, etc. This disclosure uses a network element as an example. For instance, a communication system can include at least one terminal device and at least one network device. The network device can send downlink signals to the terminal device, and / or the terminal device can send uplink signals to the network device. It is understood that the terminal device in this disclosure can be replaced by a first network element, and the network device can be replaced by a second network element, both performing the corresponding communication methods described in this disclosure.

[0134] Figure 1 is a schematic diagram of a communication system applicable to the communication method of this application embodiment.

[0135] As shown in Figure 1, the communication system 100A may include at least one access network device, such as access network device 110 shown in Figure 1; the communication system 100A may also include at least one terminal device, such as terminal device 120 and terminal device 130 shown in Figure 1. Access network device 110 and terminal devices (such as terminal devices 120 and 130) can communicate via a wireless link. The communication devices in this communication system, for example, access network device 110 and terminal device 120, can communicate via multi-antenna technology.

[0136] In wireless communication networks, such as mobile communication networks, the services supported by the networks are becoming increasingly diverse, thus requiring increasingly diverse demands. For example, networks need to support ultra-high speeds, ultra-low latency, and / or massive connectivity. This characteristic makes network planning, network configuration, and / or resource scheduling increasingly complex. Furthermore, as network functions become more powerful, such as supporting higher spectrum levels, supporting higher-order multiple-input multiple-output (MIMO) technologies, supporting beamforming (BF), and supporting beam management, network energy efficiency has become a hot research topic. These new demands, new scenarios, and new characteristics bring unprecedented challenges to network planning, operation, and efficient operation. To meet these challenges, artificial intelligence (AI) technology can be introduced into wireless communication networks to achieve network intelligence. To support AI technology in wireless networks, AI nodes may also be introduced.

[0137] Figure 2 is a schematic diagram of another communication system applicable to the communication method of the embodiments of this application.

[0138] Compared to the communication system 100A shown in Figure 1, the communication system 100B shown in Figure 2 also includes an AI network element 140. The AI ​​network element 140 is used to perform AI-related operations, such as building training datasets or training AI models. The AI ​​network element can also be simply referred to as an intelligent network element.

[0139] In one possible implementation, access network device 110 can send data related to the training of the AI ​​model to AI network element 140, which then 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 terminal devices. AI network element 140 can send the results of operations related to the AI ​​model to access network device 110, which then forwards them to the terminal devices. For example, the results of operations related to the AI ​​model may include at least one of the following: a trained AI model, model evaluation results, or test results. Exemplarily, a portion of the trained AI model may be deployed on access network device 110, and another portion on terminal devices 120 and / or 130. Alternatively, the trained AI model may be deployed on access network device 110. Or, the trained AI model may be deployed on terminal devices 120 and / or 130.

[0140] It should be understood that Figure 2 is only used as an example of the AI ​​network element 140 being directly connected to the access network device 110. In other scenarios, the AI ​​network element 140 can also be connected to the terminal device. Alternatively, the AI ​​network element 140 can be connected to both the access network device 110 and the terminal device simultaneously. Alternatively, the AI ​​network element 140 can also be connected to the access network device 110 through a third-party network element. This application embodiment does not limit the connection relationship between the AI ​​network element and other network elements. For example, the AI ​​network element 140 can also be set as a module in the access network device and / or the terminal device, for example, in the access network device 110 or the terminal device shown in Figure 1.

[0141] It should be noted that Figures 1 and 2 are simplified schematic diagrams for ease of understanding. For example, the communication system may also include other devices, such as wireless relay devices and / or wireless backhaul devices, which are not shown in Figures 1 and 2. In practical applications, the communication system may include multiple access network devices and multiple terminal devices. The embodiments of this application do not limit the number of access network devices and terminal devices included in the communication system.

[0142] In the embodiments of this application, the terminal device may also be referred to as UE, access terminal, user unit, user station, mobile station, mobile station, remote station, remote terminal, mobile device, user terminal, terminal, wireless communication device, user agent, or user equipment.

[0143] Terminal devices can be devices that provide voice / data, such as handheld devices with wireless connectivity, in-vehicle devices, etc. Currently, examples of terminals include: mobile phones, tablets, laptops, PDAs, 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 with wireless communication capabilities, computing devices or other processing devices connected to a wireless modem, wearable devices, terminal devices in 5G networks, or future public land mobile communication networks. Terminal devices in a network (PLMN), etc., are not limited to this in the embodiments of this application.

[0144] By way of example and not limitation, in this embodiment, the terminal device can also be a wearable device. Wearable devices, also known as wearable smart devices, are a general term for devices that utilize wearable technology to intelligently design and develop everyday wearables, such as glasses, gloves, watches, clothing, and shoes. Wearable devices are portable devices that are worn directly on the body or integrated into the user's clothing or accessories. Wearable devices are not merely hardware devices, but also achieve powerful functions through software support, data interaction, and cloud interaction. Broadly speaking, wearable smart devices include those that are feature-rich, large in size, and can achieve complete or partial functions without relying on a smartphone, such as smartwatches or smart glasses, as well as those that focus on a specific type of application function and require the use of other devices such as smartphones, such as various smart bracelets and smart jewelry for vital sign monitoring.

[0145] In this embodiment, the device for implementing the functions of the terminal device can be the terminal device itself, or it can be any device capable of supporting the terminal device in implementing those functions, such as a chip system. This device can be installed in or used in conjunction with the terminal device. In this embodiment, the chip system can be composed of chips or may include chips and other discrete components. This embodiment only uses the terminal device as an example to illustrate the device for implementing the functions of the terminal device, and does not constitute a limitation on the solution of this embodiment.

[0146] The access network device in this application embodiment can be a device used to communicate with terminal devices. This access network device can also be called a network device, such as a base station. In this application embodiment, the access network device can refer to a RAN node (or device) that connects the terminal device to the wireless network. A base station can broadly encompass, or be replaced by, various names such as: NodeB, evolved NodeB (eNB), next-generation NodeB (gNB), relay station, access point, transmitting and receiving point (TRP), transmitting point (TP), master station, auxiliary station, motor slide retainer (MSR) node, home base station, network controller, access node, wireless node, access point (AP), transmission node, transceiver node, baseband unit (BBU), remote radio unit (RRU), active antenna unit (AAU), remote radio head (RRH), central unit (CU), distributed unit (DU), radio unit (RU), positioning node, etc. A base station can be a macro base station, micro base station, relay node, donor node, or similar entities, or combinations thereof. A base station can also refer to a communication module, modem, or chip installed within the aforementioned equipment or apparatus. A base station can also be a mobile switching center, a device that performs base station functions in D2D, V2X, and M2M communications, or a device that performs base station functions in future communication systems. A base station can support networks using the same or different access technologies. Optionally, a RAN node can also be a server, wearable device, vehicle, or in-vehicle equipment. For example, the access network equipment in V2X technology can be a roadside unit (RSU). The embodiments of this application do not limit the specific technology or equipment form used in the access network equipment.

[0147] Base stations can be fixed or mobile. For example, a helicopter or drone can be configured to act as a mobile base station, and one or more cells can move depending on the location of the mobile base station. In other examples, a helicopter or drone can be configured as a device to communicate with another base station.

[0148] In some deployments, the access network equipment mentioned in the embodiments of this application may be a device including a CU, or a DU, or a device including both CU and DU, or a device with a control plane CU node (central unit-control plane (CU-CP)) and a user plane CU node (central unit-user plane (CU-UP)) and a DU node. For example, the access network equipment may include gNB-CU-CP, gNB-CU-UP, and gNB-DU.

[0149] In some deployments, multiple RAN nodes collaborate to assist terminals in achieving wireless access, with different RAN nodes each implementing some of the base station's functions. For example, RAN nodes can be CUs, DUs, CU-CPs, CU-UPs, or RUs. CUs and DUs can be configured separately or included in the same network element, such as a BBU. RUs can be included in radio frequency equipment or radio frequency units, such as RRUs, AAUs, or RRHs.

[0150] RAN nodes can support one or more types of fronthaul interfaces, each corresponding to a DU and RU with different functions. If the fronthaul interface between the DU and RU is a Common Public Radio Interface (CPRI), the DU is configured to implement one or more baseband functions, and the RU is configured to implement one or more radio frequency functions. If the fronthaul interface between the DU and RU is another type of interface, relative to CPRI, it moves some downlink and / or uplink baseband functions—for example, for downlink, precoding, digital beamforming, or one or more of inverse fast Fourier transform (IFFT) / cyclic prefix addition (CP)—from the DU to the RU; and for uplink, digital beamforming, or one or more of fast Fourier transform (FFT) / cyclic prefix removal (CP)—from the DU to the RU. In one possible implementation, this interface can be an enhanced common public radio interface (eCPRI). Under the eCPRI architecture, the splitting methods between DU and RU are different, corresponding to different types (category, Cat) of eCPRI, such as eCPRI Cat A, B, C, D, E, F.

[0151] Taking eCPRI Cat A as an example, for downlink transmission, layer mapping is used as the dividing line. The DU is configured to implement one or more functions preceding layer mapping (i.e., coding, rate matching, scrambling, modulation, and layer mapping itself), while other functions following layer mapping (e.g., resource element (RE) mapping, digital beamforming, or one or more of IFFT / CP addition) are implemented in the RU. For uplink transmission, de-RE mapping is used as the dividing line. The DU is configured to implement one or more functions preceding de-mapping (i.e., decoding, rate matching de-matching, descrambling, demodulation, inverse discrete Fourier transform (IDFT), channel equalization, and de-RE mapping itself), while other functions following de-mapping (e.g., digital BF or FFT / CP removal) are implemented in the RU. It is understood that descriptions of the functions of the DU and RU corresponding to various types of eCPRI can be found in the eCPRI protocol and will not be elaborated upon here.

[0152] In one possible design, the processing unit in the BBU used to implement baseband functions is called the baseband high (BBH) unit, and the processing unit in the RRU / AAU / RRH used to implement baseband functions is called the baseband low (BBL) unit.

[0153] In different systems, CU (or CU-CP and CU-UP), DU, or RU may have different names, but those skilled in the art will understand their meaning. For example, in an open RAN (ORAN) architecture, CU can also be called open CU (open-CU, O-CU), DU can also be called open DU (open-DU, O-DU), CU-CP can also be called open CU-CP (open-CU-CP) O-CU-CP, CU-UP can also be called open CU-UP (open-CU-UP, O-CU-UP), and RU can also be called open RU (open-RU, O-RU). Any of the units among CU (or CU-CP, CU-UP), DU, and RU in this application can be implemented through software modules, hardware modules, or a combination of software modules and hardware modules.

[0154] In this embodiment, the apparatus for implementing the functions of a network device can be a network device itself; it can also be an apparatus capable of supporting the network device in implementing those functions, such as a chip system, hardware circuit, software module, or a hardware circuit plus a software module. This apparatus can be installed in the network device or used in conjunction with the network device. In this embodiment, the example of a network device being used to implement the functions of a network device is provided only and does not constitute a limitation on the solutions described in this embodiment.

[0155] Network devices and / or terminal devices can be deployed on land, including indoors or outdoors, handheld or vehicle-mounted; they can also be deployed on water; and they can also be deployed in the air on airplanes, balloons, and satellites. This application does not limit the scenario in which the network devices and terminal devices are located. Furthermore, terminal devices and network devices can be hardware devices, or software functions running on dedicated hardware or general-purpose hardware, such as virtualization functions instantiated on a platform (e.g., a cloud platform), or entities that include dedicated or general-purpose hardware devices and software functions. This application does not limit the specific form of the terminal devices and network devices.

[0156] Optionally, the AI ​​node can be deployed in one or more of the following locations within the communication system: access network equipment, terminal equipment, or core network elements. Alternatively, the AI ​​node can also be deployed independently, for example, in a location other than any of the aforementioned devices, such as in the host or cloud server of an OTT system. The AI ​​node can communicate with other devices in the communication system, which can be one or more of the following: access network equipment, terminal equipment, or core network elements.

[0157] It is understood that this application does not limit the number of AI nodes. For example, when there are multiple AI nodes, these nodes can be divided based on function, such as different AI nodes being responsible for different functions.

[0158] It can also be understood that AI nodes can be independent devices, or they can be integrated into the same device to achieve different functions. Alternatively, they can be network elements in hardware devices, software functions running on dedicated hardware, or virtualization functions instantiated on a platform (e.g., a cloud platform). This application does not limit the specific form of the aforementioned AI nodes.

[0159] AI nodes can be AI network elements or AI modules.

[0160] Figure 3 is a schematic diagram of a possible application framework in a communication system.

[0161] As shown in Figure 3, network elements in the communication system are connected via interfaces (e.g., next-generation (NG) interfaces, Xn interfaces) or air interfaces. The NG interface is the interface between the radio access network and the 5G core network. The Xn interface is the interface between access network devices, and the air interface is the interface between access network devices and terminal devices. These network element nodes, such as core network devices, access network nodes (RAN nodes), terminals, or one or more devices in the OAM, are equipped with one or more AI modules (only one is shown in Figure 3 for clarity). The access network node can be a single RAN node or can include multiple RAN nodes, for example, including CU and DU. The CU and / or DU can also be equipped with one or more AI modules. Optionally, the CU can be further divided into CU-CP and CU-UP. One or more AI models are configured in the CU-CP and / or CU-UP.

[0162] The AI ​​module is used to implement corresponding AI functions. AI modules deployed in different network elements can be the same or different. Depending on the parameter configuration, the AI ​​module can implement different functions. The AI ​​module model can be configured based on one or more of the following parameters: structural parameters (e.g., at least one of the following: number of neural network layers, neural network width, inter-layer connections, neuron weights, neuron activation function, or bias in the activation function), input parameters (e.g., type and / or dimension of input parameters), or output parameters (e.g., type and / or dimension of output parameters). The bias in the activation function can also be referred to as the neural network bias.

[0163] An AI module can have one or more models. A model can infer an output, which includes one or more parameters. The learning, training, or inference processes of different models can be deployed on different nodes or devices, or they can be deployed on the same node or device.

[0164] Network devices can be network devices equipped with one or more AI modules. These network devices can include one or more devices in the core network, access network nodes (RAN nodes), or OAM as shown in Figure 3. For example, the AI ​​module can be a RAN intelligent controller (RIC) as shown in Figure 4, such as a near-real-time RIC (near-RT RIC) or a non-real-time RIC (non-RT RIC). For instance, a near-real-time RIC is located in a RAN node (e.g., in a CU or DU), while a non-real-time RIC is located in the OAM, a cloud server, a core network device, or other access network devices. The RIC can obtain subsets from multiple terminal devices from RAN nodes (e.g., CU, CU-CP, CU-UP, DU, and / or RU), reassemble them into a training dataset, and train based on the training dataset. Exemplarily, near-real-time RICs and non-real-time RICs can also be set up as separate network elements, and the access network device can be either a near-real-time RIC or a non-real-time RIC.

[0165] Figure 4 is a schematic diagram of another possible application framework in a communication system.

[0166] The communication system shown in Figure 4 includes access network nodes (CU, DU, and RU are shown in the figure) and terminals, as well as a Resource Interchange (RIC). For example, the RIC can be an AI module as shown in Figure 3, used to implement AI-related functions. The RIC includes near-real-time RICs and non-real-time RICs. Non-real-time RICs primarily process non-real-time information, such as data that is not sensitive to latency, with latency in the order of seconds. Real-time RICs primarily process near-real-time information, such as data that is relatively sensitive to latency, with latency in the order of tens of milliseconds.

[0167] The near real-time RIC is used for model training and inference. For example, it can be used to train an AI model and then use that AI model for inference. The near real-time RIC can obtain network-side and / or terminal-side information from RAN nodes (e.g., CU, CU-CP, CU-UP, DU, and / or RU) and / or terminals. This information can be used as training data or inference data. Optionally, the near real-time RIC can deliver inference results to RAN nodes and / or terminals. Optionally, inference results can be exchanged between CU and DU, and / or between DU and RU. For example, the near real-time RIC delivers the inference result to the DU, and the DU sends it to the RU.

[0168] The non-real-time RIC is also used for model training and inference. For example, it can be used to train an AI model and then use that model for inference. The non-real-time RIC can obtain network-side and / or terminal-side information from RAN nodes (e.g., one or more of CU, CU-CP, CU-UP, DU, or RU) and / or terminals. This information can be used as training data or inference data, and the inference results can be delivered to the RAN nodes and / or terminals. Optionally, inference results can be exchanged between CU and DU, and / or between DU and RU; for example, the non-real-time RIC delivers the inference results to the DU, which then forwards them to the RU.

[0169] The near real-time RIC and non-real-time RIC can also be set up as separate network elements. Optionally, the near real-time RIC and non-real-time RIC can also be part of other devices. For example, the near real-time RIC can be set in the RAN node (e.g., in the CU or DU), while the non-real-time RIC can be set in the OAM, cloud server, core network device, or other network device.

[0170] To facilitate understanding of the embodiments of this application, the terms involved in this application will be briefly explained below.

[0171] (1) Artificial Intelligence. This refers to enabling machines to learn, accumulate experience, and solve problems that humans can solve through experience, such as natural language understanding, image recognition, and chess. Artificial intelligence can be understood as the intelligence exhibited by machines created by humans. Generally, artificial intelligence refers to the technology of presenting human intelligence through computer programs. The goals of artificial intelligence include understanding intelligence by constructing computer programs that demonstrate symbolic reasoning or reasoning.

[0172] (2) Machine Learning (ML). This is one implementation method of artificial intelligence. Machine learning is a method that endows machines with the ability to perform functions that cannot be directly programmed. In practical terms, machine learning is a method that uses data to train a model and then uses the model to make predictions. There are many machine learning methods, such as neural networks (NN), decision trees, and support vector machines. Machine learning theory mainly involves designing and analyzing algorithms that allow computers to learn automatically. Machine learning algorithms are a class of algorithms that automatically analyze data to obtain patterns and use these patterns to predict unknown data.

[0173] (3) Channel information.

[0174] In communication systems, network devices determine one or more of the following configurations for scheduling downlink data channels of terminal devices based on channel information: resources, modulation and coding scheme (MCS), and precoding. Channel information, also known as channel environment information (CSI) or channel characteristics information, reflects channel features and quality. For example, CSI can be represented by a channel matrix; for instance, CSI may include the channel matrix, or it may include the channel's eigenvectors.

[0175] CSI measurement refers to the process by which the receiver calculates channel information based on a reference signal transmitted by the transmitter, i.e., estimating channel information using channel estimation methods. The propagation of a wireless signal in a channel can be represented as Y = HX + N_noise, where H is the CSI, X is the reference signal, N_noise is the noise, and Y is the received signal. The reference signal X is known information defined by the terminal device and network device. After acquiring the received signal Y, channel estimation algorithms, such as least squares or least mean square error, can be used for channel estimation. For example, the reference signal may include one or more of the following: channel state information reference signal (CSI-RS), synchronizing signal / physical broadcast channel block (SSB), sounding reference signal (SRS), or demodulation reference signal (DMRS). One or more of CSI-RS, SSB, and DMRS can be used to measure downlink channel information. SRS and DMRS can be used to measure uplink channel information.

[0176] Taking FDD communication as an example, since uplink and downlink channels lack reciprocity or cannot guarantee reciprocity, network devices need to obtain downlink CSI through uplink feedback from terminal devices. Network devices typically send a downlink reference signal to the terminal device, which receives this signal. Since the terminal device knows the transmission information of the downlink reference signal, it can perform channel measurements and interference measurements based on the received signal to estimate the downlink channel it traverses. The terminal device then generates the downlink CSI based on this measurement and the resulting downlink channel matrix. Finally, the terminal device generates a CSI report according to a predefined protocol method or network device configuration and feeds it back to the network device so that it can obtain the downlink CSI.

[0177] In this embodiment, the meaning of CSI is broader than that in traditional schemes. It is not limited to channel quality indication (CQI), precoding matrix indicator (PMI), rank indicator (RI), layer indicator (LI), or CSI-RS resource indicator (CRI). It can also be one or more of the following: channel response information (such as channel response matrix), channel matrix, channel feature matrix, precoding matrix, reference signal receiving power (RSRP), signal to interference plus noise ratio (SINR), time domain channel properties (TDCP), the identity (ID) of the optimal beam, or the ID of the top K beams. The signal to interference plus noise ratio can also be called the signal-to-interference-plus-noise ratio. The optimal beam refers to the beam that maximizes the received or transmitted energy. For example, if the receiver uses different receiving beams to receive signals, the optimal beam may include the beam with the largest RSRP among the signals received from multiple different receiving beams. For example, if the transmitter uses different transmit beams to send signals, the optimal beam can include the beam with the highest RSRP (Receiving Power Ratio) of the signal when it arrives at the receiver. Similarly, the optimal top K beams refer to the top K beams that maximize the received or transmitted energy.

[0178] In this system, RI indicates the number of downlink transmission layers suggested by the terminal device, CQI indicates the modulation and coding schemes supported by the current channel conditions as determined by the terminal device, and PMI indicates the precoding suggested by the receiver of the reference signal, such as the terminal device. The number of precoding layers indicated by PMI corresponds to RI. The channel response and channel matrix represent the channel itself, while the channel feature matrix and precoding matrix are matrices composed of features extracted from the channel.

[0179] It should be understood that network devices may perform downlink transmissions in accordance with some or all of the information indicated in the CSI report. Alternatively, network devices may perform downlink transmissions without following the information indicated in the CSI report.

[0180] In codebook-based CSI feedback, for some codebooks with high overhead, such as R15type II, R16type II, and R17type II, the CSI report content can be divided into two parts: Part 1 and Part 2. CQI and RI belong to Part 1, while PMI belongs to Part 2. Part 2 is typically transmitted via the Physical Uplink Shared Channel (PUSCH). Since the size of Part 2 is not fixed, and multiple CSI reports may need to be transmitted on the same resource, the size of Part 2 to be transmitted may exceed the channel's capacity. In related schemes, when the number of coded modulation symbols (or modulation symbols) in Part 2 to be transmitted exceeds a set threshold, the terminal device will discard lower-priority content according to the priority order of the report content until the number of coded modulation symbols in Part 2 to be transmitted does not exceed the set threshold. The number of coded modulation symbols in Part 2 to be transmitted and the set threshold can be calculated using formulas and parameters defined in the protocol.

[0181] Taking the CSI feedback method based on the R16 codebook as an example, part 2 of a CSI report can be divided into three groups. In descending order of priority, these three groups are group 0, group 1, and group 2. Group 0 includes oversampling selection (or phase rotation selection), spatial basis indication, and strongest coefficient indication; group 1 includes frequency basis indication, partial coefficient indication, partial coefficient amplitude, and phase; group 2 includes remaining partial coefficient indication, remaining partial coefficient amplitude, and phase.

[0182] (4) Channel Reporting.

[0183] Channel reports can be used to reflect channel measurement information or channel information corresponding to a reference signal (which can be used for channel measurement or channel estimation). In other words, channel reports are information generated based on the information obtained from measuring the reference signal, and they can reflect channel environment information, etc.

[0184] Channel report can also be replaced by channel measurement report, or measurement report, or CSI report, or CSI feedback information, or CSI compression information, etc., without limiting other terms that may be used.

[0185] (5) AI-based CSI-RS feedback enhancement.

[0186] AI-based CSI-RS feedback enhancement can perform CSI prediction and CSI compression.

[0187] CSI prediction: For example, it can include time-domain CSI prediction based on the UE side model.

[0188] Figure 5 illustrates the inference process of CSI prediction.

[0189] As shown in Figure 5, the AI / ML-based CSI prediction model is used to predict future CSI based on historical CSI. The historical CSI can be of various types, such as the original channel matrix or a precoding matrix. The input data of the CSI prediction model is related to the channel measurement results; for example, the input data can be obtained by preprocessing (e.g., normalizing) the channel measurement results. The output data of the CSI prediction model may also require some processing.

[0190] For example, the training data for training the CSI prediction model can be acquired or generated by the UE. This training data may include the target CSI within the observation window and / or the target CSI within the prediction window. For the UE-side model inference process, the UE internally performs model inference using the corresponding input data. The output data of the CSI prediction model's inference process may include the predicted CSI. For the network-side performance monitoring process, the performance metrics used for model monitoring or the data used to calculate these metrics are acquired or generated by the UE and fed back to the base station, which then performs performance monitoring. The data related to the CSI prediction model's model monitoring process may include the ground-truth CSI, the calculated performance metrics, and the performance monitoring results.

[0191] CSI compression can include, for example, spatial-frequency domain CSI compression or spatial-temporal domain CSI compression based on a two-side model. Spatial-frequency domain CSI compression can be understood as the joint compression of CSI in both the spatial and frequency domains. Spatial-temporal domain CSI compression can also be understood as the joint compression of CSI in both the spatial and temporal domains.

[0192] Figure 6 shows a schematic diagram of the inference process of CSI compression.

[0193] The AI / ML-based CSI generation component generates CSI feedback information at the UE side. The AI / ML-based CSI reconstruction component reconstructs the CSI at the base station side based on the received CSI feedback information. The UE can select a CSI generation model compatible with the CSI reconstruction model used by the gNB based on model pairing information. The input (for the CSI generation component) or output (for the CSI reconstruction component) of the AI / ML model used for CSI compression can be of the type of raw channel matrix or precoding matrix, etc. Further preprocessing of the channel measurement results may be required to generate the input to the CSI generation model. Further processing may also be required for the output of the CSI reconstruction model.

[0194] For the NW portion of two-side model inference, input data can be generated or acquired by the UE and terminates at the base station. For the UE portion of two-side model inference, input data is available within the UE. Data related to the CSI-compressed model inference process may include CSI feedback information.

[0195] For model training, the training data used for model training can be generated or acquired by the UE or base station. Data related to the CSI-compressed model training process may include target CSI, CSI feedback information, or gradients of CSI feedback information, etc.

[0196] For NW-side performance monitoring, the calculated performance metrics or the data used for performance metric calculation can be generated or acquired by the UE and terminate at the base station. Data related to the model monitoring process of CSI compression may include target CSI or calculated performance metrics, etc.

[0197] For CSI feedback enhancement use cases, the monitoring method can be:

[0198] 1) NW side monitoring, such as base stations estimating the performance of AI / ML models based on the target CSI reported by the UE (the actual channel estimation results associated with the CSI report) and further generating monitoring decisions.

[0199] 2) UE-side monitoring, for example, the UE performs model monitoring based on the output of the CSI reconstruction model indicated by the base station (the UE needs the output of the CSI reconstruction model to be associated with the CSI report), or, based on the output of the CSI reconstruction model of the UE-side agent, or, directly estimates intermediate KPIs, or estimates the monitoring output. The base station can configure thresholds to the UE to instruct the UE to perform model monitoring.

[0200] Data collection for CSI compression use cases, from the perspectives of training, inference, and monitoring, may include the following information:

[0201] - Training: Target CSI, CSI feedback, gradient of CSI feedback.

[0202] - Reasoning: CSI feedback.

[0203] - Monitoring: Target CSI, calculated performance metrics.

[0204] (6) AI-based beam management enhancement

[0205] AI-based beam management enhancement can include sub-scenarios such as beam scanning matrix prediction and optimal beam prediction. AI / ML-based sparse beam prediction aims to improve accuracy, and one possible workflow is as follows:

[0206] 1) Generation of the initial model. By having a certain number of UEs report the results of the SSB full-beam scan, a sparse scan matrix is ​​trained. This matrix is ​​usually unique to each cell.

[0207] 2) The base station sends the sparse model to the UE (which can be done via SIB messages, etc.), and the UE performs beam scanning in the P1 phase based on this matrix;

[0208] 3) Based on the sparse scan results of the UE, the base station infers the optimal CSI-RS beam and starts P2 scanning of the UE, and the UE feeds back the optimal CSI-RS beam ID.

[0209] Figure 7 illustrates a schematic diagram of the reasoning process for beam management.

[0210] The BM case is divided into two types: BM case1 and BM case2. The BM case model is a one-side model, meaning the model is deployed on either the NW side or the UE side. For BM-Case1 and BM-Case2, the UE can report the prediction results to the NW based on the output of the UE-side model, or the NW can predict the Top-1 / N beams based on the reported measurements of Set B for the NW-side model.

[0211] BM Case 1: Downlink beam prediction based on measurement results from Set B, predicting set A (Set A). One possible workflow is as follows:

[0212] 1) The gNB scans the Set B beam, and the gNB / UE acquires the measurement results of Set B;

[0213] 2) The AI ​​model on the gNB / UE side uses the measurement results of SetB as the model input to predict the TopK beam on Set A.

[0214] BM Case 2: Predicting future downlink beamforming under Set A based on historical measurement results of Set B. One possible procedure is as follows:

[0215] 1) The gNB scans the Set B beam, and the gNB / UE acquires the measurement results of Set B;

[0216] 2) The AI ​​model on the gNB / UE side uses the measurement results of SetB as the model input to predict the TopK beam on Set A at future times.

[0217] For the UE side model, model monitoring methods are divided into the following types:

[0218] Method 1 (Type 1) Network-side monitoring:

[0219] Network-side monitoring (NW side monitoring): The UE reports the NW's labels and inference outputs to calculate metrics.

[0220] UE-assisted monitoring: UEs report performance metrics or events based on performance metrics.

[0221] Type 2 UE-side monitoring: UE reports monitoring decisions (e.g., model selection / activation / deactivation / switching / fallback operations).

[0222] A performance metric can be one or more of the following:

[0223] Key performance indicators (KPIs) related to beam prediction accuracy, such as Top-K / 1 beam prediction accuracy. Link quality-related KPIs, such as throughput, Layer 1 reference signal received power (L1-RSRP), L1-SINR, and assumed block error rate (BLER). Performance metrics for AI / ML-based input / output data distribution. The difference between measured RSRP and predicted RSRP, i.e., the difference in L1-RSRP.

[0224] Data collection for BM use cases, from the perspectives of training, inference, and monitoring, may include the following information:

[0225] - Training: L1-RSRPs and / or beam IDs;

[0226] -Inference: Predicted L1-RSRPs and / or beam IDs;

[0227] - Monitoring: L1-RSRPs and / or beam IDs, and calculating performance metrics (e.g., prediction accuracy, beam accuracy indicator (BAI)).

[0228] (7) AI-based positioning accuracy enhancements

[0229] AI-based location enhancement primarily aims to improve location accuracy. The possible process for enhancing location accuracy is as follows:

[0230] 1) Collect raw data using a reference UE controlled by the operator.

[0231] 2) The location management function (LMF) node (non-RAN side node) and gNB are trained separately. The LMF model can infer the final positioning (latitude and longitude, etc.), and the gNB model can infer the line of sight (LOS) / non-line of sight (NLOS) judgment results.

[0232] (8) CSI processing rules.

[0233] The NR protocol specifies the CSI processing guidelines on the terminal device side. The terminal device reports the number N of CSI calculations it can process simultaneously, based on its own capabilities. CPU That is, the UE is configured with N CPU Each CSI processing unit (CPU) is used to process all CSI reports configured on all configuration component carriers (CCs).

[0234] At a given symbol, if the computation reported by the CSI uses L CPUs, then the terminal device has N CPUs. CPU -L represents unused CPU resources. For a given symbol S, there are N... CPUIf -L CPUs are not occupied, and there are N CSI reports that need to occupy their respective CPUs starting from symbol S, then the number of CPUs corresponding to the nth CSI report among the N CSI reports is... Then the terminal device does not need to update the (NM) lowest priority CSI reports out of N CSI reports, where n = 1, 2, ..., N, 0 ≤ M ≤ N, and M is a subset of N CSI reports. The maximum value.

[0235] When there is insufficient idle CPU, the UE does not need to update the CSI report, but rather does not need to provide feedback. When there is insufficient CPU, the CSI report provided by the UE can be a cached previous CSI report or anything else, entirely depending on the UE implementation.

[0236] The amount of CPU used varies depending on the different types of CSI reports.

[0237] When performing time-frequency tracking of the tracking reference signal (TRS), i.e., when the report quantity is configured as 'none' and the CSI-RS-ResourceSet contains high-level parameter TRS information (trs-Info), no CPU is used, i.e., OCPU=0.

[0238] For Layer 1 Reference Signal Received Power (L1-RSRP) measurement, i.e., when reportQuantity is configured as 'cri-RSRP', 'ssb-index-RSRP', or 'none', and CSI-RS-ResourceSet does not contain the higher-layer parameter trs-Info, the CPU usage is 1, i.e., OCPU = 1.

[0239] For a CSI report with the high-level parameter reportQuantity set to 'tdcp' and the latency Y configured by the high-level parameter Y, OCPU = (Y+1)÷X, where the value of X is reported by the UE capability.

[0240] When reportQuantity is configured as 'cri-RI-PMI-CQI', 'cri-RI-i1', 'cri-RI-i1-CQI', 'cri-RI-CQI', or 'cri-RI-LI-PMI-CQI', OCPU = Ks, where Ks represents the number of non-zero power (NZP) CSI-RS resources in CSI-RS ResourceConfigMobility (CMR).

[0241] When an aperiodic (AP) CSI report is triggered, if there is no physical uplink shared channel (PUSCH) transmission and no CPU is occupied, and this CSI report is wideband, Type I codebook or has no PMI feedback, and there is only one CSI-RS resource with 4 or fewer ports in the CMR, then all CPUs are occupied, i.e., OCPU = NCPU.

[0242] For each CSI report processed, the CPU will continuously occupy a certain number of symbols. The protocol specifies that when the reporting type (reportConfigType) is not set to 'none', the number of CPU symbols occupied is determined according to the following rules.

[0243] The CPU time occupied by periodic (P) or semi-persistent (SP) CSI reports (excluding the initial semi-persistent CSI report reported on the PUSCH after the physical downlink control channel (PDCCH) triggers the CSI report) is as follows: starting from the first orthogonal frequency division multiplexing (OFDM) symbol of the earliest of the latest resources in the CSI-RS / CSI interference measurement (CSI-IM) / SSB occasions that are earlier than the CSI reference resource, until the last symbol of the PUSCH / physical uplink control channel (PUCCH) carrying the report.

[0244] CPU time consumed by non-periodic CSI reports: from the first symbol after the PDCCH that triggers the CSI report to the last symbol of the PUSCH that carries the report.

[0245] The CPU time occupied by the initial semi-persistent CSI report reported on the PUSCH after the PDCCH triggers the CSI report is from the first symbol after the PDCCH until the last symbol of the PUSCH carrying the report.

[0246] The CPU time consumed by a semi-static CSI report on the PUSCH configured for the R18 Doppler codebook is calculated from the first symbol of the latest consecutive Kp P / SP CSI-RS Occasions earlier than the CSI reference resource, up to the last symbol of the PUSCH carrying the report. Here, Kp ∈ {1, 2, 4} is indicated by the UE capability and can be understood as the observation window supported by the UE.

[0247] The NR protocol also specifies the CSI calculation time. Network devices must allow sufficient time for terminal devices when triggering CSI reporting. For CSI reports triggered by DCI on the PUSCH, the protocol stipulates that a terminal device will only report a valid CSI report if the following two conditions are met:

[0248] Condition 1: The first uplink symbol carrying the corresponding CSI report (including the impact of timing advance) must not start earlier than symbol Zref;

[0249] Condition 2: The first uplink symbol carrying the nth CSI report (including the effects of timing advance) must start no earlier than symbol Z'ref(n).

[0250] Here, Z ref An uplink symbol is defined as one in which the interval between the start time of its cyclic prefix (CP) and the end time of the last symbol of the PDCCH that triggers the CSI report is greater than or equal to T. proc,CSI =(Z)(2048+144)·κ2 -μ ·T c And it is the earliest uplink symbol to satisfy this condition. When aperiodic CSI-RS is used for channel measurements of the nth triggered CSI report, Z' ref (n) is defined as an uplink symbol whose CP start time is greater than or equal to the interval between the start time of the CP and the end time of the last symbol of the latest-ending resource among the resources used for channel measurements. proc,CSI =(Z′)(2048+144)·κ2 -μ ·T c Furthermore, it must be the earliest uplink symbol to meet this condition. The above provision can be understood as the time interval between the end time of the first symbol of the PUSCH carrying the CSI report and the end time of the last symbol of the PDCCH triggering the CSI report being greater than or equal to the specified time parameter T.proc,CSI Furthermore, the time interval between the first symbol of the PUSCH carrying the CSI report and the end time of all reference resources used for channel measurements must be greater than or equal to the specified time parameter T′. proc,CSI As shown in Figure 8.

[0251] Figure 8 is a schematic diagram of CSI calculation time.

[0252] When the above conditions are not met, the terminal device does not need to update the reported CSI. The values ​​of Z and Z' in the above formula are determined according to the tables and principles given in the protocol, and the constant κ is 64. c Represents time units, where T c =1 / (Δf) max ·N F ), where Δf max =480×10 3 Hz, N f =4096. μ represents the value corresponding to the subcarrier spacing, for example, 0 represents a subcarrier spacing of 15kHz, 1 represents a subcarrier spacing of 30kHz, 2 represents a subcarrier spacing of 60kHz, and 3 represents a subcarrier spacing of 120kHz.

[0253] Furthermore, for CSI reporting not triggered by DCI (periodic and semi-persistent reporting), the protocol defines a CSI reference resource to limit the CSI calculation time, ensuring that the terminal device only needs to update the reported CSI when there is sufficient CSI calculation time. The CSI reference resource is defined as a time-frequency resource. In the frequency domain, the CSI reference resource is defined by a set of downlink physical resource blocks corresponding to the frequency band related to CSI calculation. In the time domain, the CSI reference resource is defined as a valid timeslot preceding the uplink timeslot for CSI reporting, and the symbol interval between the timeslot containing the CSI reference resource and the CSI reporting timeslot must be greater than the value specified by the protocol. The CSI-RS used to calculate the CSI report cannot be later than the CSI reference resource. If there is no valid downlink timeslot corresponding to a certain CSI reporting configuration, the terminal device may not report the CSI. The above provisions can be understood as the time interval between the CSI-RS used to calculate the CSI report and the CSI reporting timeslot being greater than or equal to the specified time parameter.

[0254] NR specifies the activation and counting rules for CSI-RS reports / resources.

[0255] In any time slot, the UE will not have more active CSI-RS ports or active CSI-RS resources in the active bandwidth part (BWP) than reported in the capability report. NZP CSI-RS resources are active for the duration defined below:

[0256] For AP CSI-RS: It starts from the end of the PDCCH containing the request and ends at the end of the scheduling PUSCH containing the report associated with the non-periodic CSI-RS.

[0257] For SP CSI-RS: It starts from the end of the activation command and ends at the end of the deactivation command.

[0258] For P CSI-RS: It starts when the higher-level signaling configuration period CSI-RS is configured and ends when the period CSI-RS configuration is released.

[0259] If a CSI-RS resource is referenced N times by one or more CSI Reporting Settings that do not have the higher-layer parameter "csi-ReportSubConfigToAddModList" configured, then the CSI-RS resource and the CSI-RS ports within the CSI-RS resource are counted N times. This can be understood as the UE needing to store N measurement results for the CSI-RS resource for different processing purposes if a CSI-RS resource is referenced N times by one or more CSI Reporting Settings. Therefore, the CSI-RS resource count must be multiplied by N, and the corresponding CSI-RS port count within each CSI-RS resource must also be multiplied by N.

[0260] For CSI-ReportConfig links linked with a higher-layer codebook type parameter configured as 'typeII-Doppler-r18' or 'typeII-Doppler-PortSelection-r18', the CSI-RS resources in the channel measurement CSI-RS resource set are either periodic or semi-static. The CSI-RS resources and the CSI-RS ports within those resources are counted Kp times, where Kp ∈ {1, 2, 4} is indicated by the UE capability. This can be understood as follows: for Doppler prediction, the UE needs to store measurement results for Kp time intervals for the CSI-RS resources using a sliding window. Therefore, for the P / SP CSI-RS resources, the resource count is multiplied by Kp, and correspondingly, the CSI-RS port count within the CSI-RS resources is also multiplied by Kp.

[0261] (9) Doppler codebook

[0262] For R18 Doppler codebook prediction, the standard enhancements for CPU count, CPU usage, and CSI calculation time are as follows:

[0263] -CPU count:

[0264] If the CSI-RS resource set corresponding to the channel measurement is aperiodic (AP) and K CSI-RS resources are configured, then O_CPU = 8 (for K = 12) and O_CPU = Y1 * K (for K < 12), where Y1 ∈ {1, 2, 3} is reported by the UE capability indication.

[0265] If the corresponding CSI-RS Resource Set used for channel measurement is periodic (P) or semi-static (SP) and a CSI-RS resource is configured, then O_CPU = 4 (for N_4 = 1), O_CPU = max(Y2*N_4,4) (for N_4>1), where the value of N4 is configured by the higher-layer parameter N4, and Y2∈{1,2,3} is reported by the UE capability indication.

[0266] - CPU timeline usage:

[0267] For AP CSI reports, consistent with legacy, CPU usage starts from the first symbol of the PDCCH that triggered the CSI report and ends at the last symbol of the scheduled PUSCH.

[0268] For SP CSI reports, CPU usage starts from the first symbol of the most recent consecutive Kp cycles / semi-static CSI-RS occasion of the CSI reference resource and ends at the last symbol of the PUSCH carrying the report, where the value of Kp∈{1,2,4} is indicated by the UE capability.

[0269] -CSI calculation time:

[0270] If the CSI report is configured with N_4=1, and the non-zero power CSI-RS resource set (NZP-CSI-RS-ResourceSet) corresponding to the channel measurement is aperiodic and has K CSI-RS resources, then the CSI calculation delay is: (Z2+14(K-1)m,Z2');

[0271] If the CSI report is configured with N_4=1, and the NZP-CSI-RS-ResourceSet corresponding to the channel measurement is a periodic or semi-static CSI-RS resource, then the CSI calculation delay is: (Z2+w,Z2'), where w=56*(Kp-1) or 56*Kp symbols, and Kp∈{1,2,4} is indicated by the UE capability.

[0272] If the CSI report is configured as N4>1, and the NZP-CSI-RS-ResourceSet corresponding to the channel measurement is aperiodic and has K CSI-RS resources, then the CSI calculation delay is: (Z2+14(K-1)m,Z2') or (Z2+14(K-1)m+Z2',2*Z2').

[0273] If the CSI report is configured as N4>1, and the NZP-CSI-RS-ResourceSet corresponding to the channel measurement is a periodic or semi-static CSI-RS resource, then the CSI calculation delay is: (Z2+w,Z2') or (Z2+w+Z2',2*Z2').

[0274] Therefore, for monitoring purposes, corresponding inference reports need to be stored. For P / SP inference reports, cached inference results may be overwritten / updated before the corresponding monitoring performance metrics are calculated, potentially leading to errors in performance metric calculation. Furthermore, if a monitoring report is associated with inference reports from multiple periods, the activation end time to the end of the monitoring report is unreasonable because an excessively long activation time means the UE doesn't need to save so many historical inference report results, and inference results that have already calculated performance metrics don't need to be saved, reducing storage resource utilization. Similarly, if a inference report is associated with multiple monitoring reports, the activation end time to the end of the monitoring report is also unreasonable because the inference report will be saved multiple times, also reducing storage resource utilization.

[0275] The communication method provided in the embodiments of this application will be described below with reference to the accompanying drawings, taking the interaction between a terminal device and a network device as an example.

[0276] Terminal devices can be the terminal device itself, the communication module within the terminal device, or the circuits or chips within the terminal device responsible for communication functions (such as modem chips, also known as baseband chips, or SoC chips or SIP chips containing modem cores, etc.), or AI entities on the terminal device side. AI entities on the terminal device side can be the terminal device itself, or AI entities serving the terminal device, such as servers, such as OTT servers or cloud servers.

[0277] The network device side can be the network device itself, the communication module within the network device, or the circuits or chips within the network device responsible for communication functions (such as a modem chip, also known as a baseband chip, or a system-on-a-chip (SoC) chip or SIP chip containing a modem core, etc.), or it can be an AI entity on the network device side. The AI ​​entity on the network device side can be the network device itself, or it can be an AI entity serving the network device, such as a RIC, OAM, or a server, such as an OTT server or a cloud server.

[0278] Figure 9 shows a schematic diagram of an AI / ML functional framework for wireless air interfaces.

[0279] The diagram shows the data collection unit, model training unit, management unit, inference unit, and model storage unit.

[0280] The data collection unit can collect data from network devices and / or terminal devices. AI-based air interface or RAN systems require different types of data at various stages of the AI ​​model development process. For example, training the model requires a large amount of diverse data, model inference requires real-time data, and model monitoring requires near real-time data. For wireless or air interface tasks, the data collected by the data collection unit via the wireless air interface includes, but is not limited to, one or more of the following: measurement signals transmitted by network devices, measurement results measured and reported by terminal devices, reference signals transmitted by terminal devices, and measurement results measured by network devices. The data collected by the data collection unit, after preprocessing and other operations, can be used for one or more of the following: model training, model inference, model monitoring, or analysis. The data collected by the data collection unit can be applied to different stages, such as training data for model training, performance monitoring data for monitoring model performance, and so on.

[0281] The model training unit can train or retrain (or update) the model based on the training data from the data collection unit, and can send the trained model or the updated model to the model storage unit for storage.

[0282] The management unit can monitor model performance based on monitoring data from the data collection unit, and perform actions such as activating / deactivating, switching, selecting, and rolling back the model (or model function) based on performance. For example, it can send performance feedback or retraining requests to the model training unit. The management unit can also send model transfer requests to the model storage unit, requesting that the model storage unit send the model to the inference unit. The management unit can also generate monitoring reports based on the monitoring data.

[0283] The inference unit can perform model inference based on the model from the model storage unit and the inference data from the data collection unit, and output an inference report. The inference unit can also receive instructions from the model management unit.

[0284] In one specific implementation, the inference unit includes a device on which the AI ​​model is deployed. For example, a terminal device.

[0285] In one specific implementation, the management unit includes a device that sends configuration information. For example, a network device.

[0286] Figure 10 is a schematic flowchart of a communication method 800 provided in an embodiment of this application. The steps of method 800 are described in detail below.

[0287] S810 allows terminal devices and network devices to configure storage parameters.

[0288] In one specific implementation, S810 includes the network device sending configuration information 1 to the terminal device.

[0289] In one specific implementation, configuration information 1 is used to instruct terminal resource storage inference report 1.

[0290] In one specific implementation, configuration information 1 is used to instruct the network device to configure monitoring report 1 for the terminal device. Monitoring report 1 is determined based on inference report 1.

[0291] In one specific implementation, configuration information 1 is also used to indicate the correlation between the reasoning report and the monitoring report.

[0292] S820, the terminal device determines the count of the first resource and / or the first activation interval corresponding to the first resource.

[0293] This can also be understood as the terminal device determining the count of the first type of resource, and the first resource belonging to the first type of resource.

[0294] The first resource corresponds to the first inference report, or the configuration information of the first resource corresponds to the first inference report, or the first resource corresponds to a inference result in the first inference report, or the first resource corresponds to the first monitoring report associated with the first inference report, or the configuration information of the first resource corresponds to the first monitoring report associated with the first inference report.

[0295] The count of Type I resources and / or the count of the first resource is based on resource #0. The first resource includes A resources #0, where A is a natural number greater than or equal to 1.

[0296] The first resource includes A resources #0, which can also be understood as the first resource count being A.

[0297] This can also be understood as increasing the count of the first resource within the first activation interval by A units. A is a natural number greater than or equal to 1.

[0298] In one specific implementation, the first type of resource includes channel state information reference signal (CSI-RS) resources.

[0299] In one specific implementation, the first resource includes a CSI-RS resource corresponding to an inference report.

[0300] In one specific implementation, resource #0 includes one CSI-RS resource.

[0301] It should be understood that in some specific implementations, the first resource is also called the activation resource, and this application does not make any special limitation on this.

[0302] In one specific implementation, the value of A is 1. That is, storing one inference report requires one first resource from the terminal device.

[0303] It should be noted that activated resources include CSI-RS resources.

[0304] It should be noted that the inference report includes inference reports based on CSI-RS measurement results, which include, but are not limited to, codebooks, precoding matrices, and raw channel matrices.

[0305] It should be noted that the first resource count indicates the amount of resources required to store the inference report. The active interval indicates the duration of storage for the inference report.

[0306] In one specific implementation, the first resource counting rule includes the following:

[0307] Counting Rule 1: When an inference report is configured, activated, triggered, or reported, the CSI-RS resource count value increases by A. A is a natural number greater than or equal to 1.

[0308] This can be understood as the UE consuming A activation resources or CSI-RS resources to store an inference report. The value of A can be indicated by the UE, predefined by the protocol, or related to the configuration of the monitoring report.

[0309] This can be understood as the UE consuming A units of first resources to store the first inference report. The value of A can be indicated by the UE, predefined by the protocol, or related to the configuration of the monitoring report.

[0310] This can be understood as the UE storing the first inference report, with the first resource count being A. The value of A can be indicated by the UE, predefined by the protocol, or related to the configuration of the monitoring report.

[0311] In one specific implementation, the value of A is indicated by the UE to include the value of A as part of the UE's capability to report.

[0312] In one specific implementation, the value of A is related to the configuration of the monitoring report, and the value of A is less than or equal to N4, or the value of A is less than or equal to the product of N4 and the number of inference reports that need to be monitored.

[0313] In one specific implementation, CSI-RS resources can be replaced or understood as: primary resources.

[0314] It should be noted that counting rule one also applies when an inference report is associated with a monitoring report, in which case the CSI-RS resource count value is increased by A. A is a natural number greater than or equal to 1. The value of A is consistent with that in counting rule one, and for the sake of brevity, this application will not elaborate further here.

[0315] If an inference report is associated with a monitoring report, then an additional CSI-RS resource is counted, or A CSI-RS resources are counted.

[0316] If an inference report is associated with a monitoring report, then the first CSI-RS resource is considered, and this first CSI-RS resource is counted 1 time, or A times. This can be understood as the first CSI-RS resource corresponding to the storage of that inference report.

[0317] In the implementation of this application, the inference report includes, but is not limited to: the inference result corresponding to the inference report, the inference result, the Doppler codebook, the predicted CSI-RS measurement result, the predicted codebook, the prediction result, etc.

[0318] Counting Rule 2: When an inference report is associated with one or more monitoring reports, the CSI-RS resource count is incremented by A. A is a natural number greater than or equal to 1.

[0319] In one specific implementation, CSI-RS resources can be replaced or understood as: primary resources.

[0320] In one specific implementation, the value of A is 1. This means the UE stores only one copy of the inference report, i.e., one copy for the first resource count. Since the UE does not store a duplicate inference report, the utilization rate of the UE's storage resources can be further improved.

[0321] If an inference report is associated with one or more monitoring reports, then an additional CSI-RS resource is counted, or A CSI-RS resources are counted.

[0322] If an inference report is associated with one or more monitoring reports, then the first CSI-RS resource is considered, and this first CSI-RS resource is counted 1 time, or A times. This can be understood as the first CSI-RS resource corresponding to the storage of that inference report.

[0323] It should be noted that the counting rules one and two described above are applicable to all time-domain inference reports. Time-domain inference reports include periodic (P) inference reports, semi-persistent (SP) inference reports, or aperiodic (AP) inference reports.

[0324] It should be noted that the above counting rules one and two can apply simultaneously to P reasoning reports, SP reasoning reports, and AP reasoning reports. Alternatively, the above counting rules one and two can apply only to AP reasoning reports. Or, the above counting rules one and two can apply only to P reasoning reports and / or SP reasoning reports.

[0325] The following description uses the example that Rules 1 and 2 apply only to AP inference reports.

[0326] In one specific implementation, the inference report is an AP inference report. The corresponding first resource counting rules include the following:

[0327] Counting Rule 1: When an AP inference report is configured, activated, triggered, or reported, the CSI-RS resource count value increases by A. A is a natural number greater than or equal to 1. Storing an AP inference report requires A first resources from the UE. The value of A can be indicated by the UE, predefined by the protocol, or related to the configuration of the monitoring report.

[0328] In one specific implementation, the value of A is indicated by the UE to include the value of A as part of the UE's capability to report.

[0329] In one specific implementation, the value of A is related to the configuration of AP monitoring reports. The value of A is less than or equal to N4, or the value of A is less than or equal to the product of N4 and the number of AP inference reports that need to be monitored.

[0330] In one specific implementation, CSI-RS resources can be replaced or understood as: primary resources.

[0331] It should be noted that counting rule one also applies when an AP inference report is associated with an AP monitoring report, in which case the CSI-RS resource count value is increased by A. A is a natural number greater than or equal to 1. The value of A is consistent with that in counting rule one, and for the sake of brevity, this application will not elaborate further here.

[0332] If an AP inference report is associated with an AP monitoring report, then an additional CSI-RS resource is counted, or A CSI-RS resources are counted.

[0333] If an AP inference report is associated with an AP monitoring report, then the first CSI-RS resource is considered, and this first CSI-RS resource is counted 1 time, or A times. This can be understood as the first CSI-RS resource corresponding to the storage of that AP inference report.

[0334] In the implementation of this application, the AP inference report includes, but is not limited to: the inference result corresponding to the AP inference report, the AP inference result, the Doppler codebook, the predicted CSI-RS measurement result, the predicted codebook, and the prediction result.

[0335] Counting Rule 2: When an AP inference report is associated with one or more AP monitoring reports, the CSI-RS resource count is incremented by A. A is a natural number greater than or equal to 1.

[0336] In one specific implementation, CSI-RS resources can be replaced or understood as: primary resources.

[0337] In one specific implementation, the value of A is 1. This means the UE stores only one copy of the AP inference report, i.e., only one resource count. Since the UE does not store a duplicate AP inference report, the UE's storage resource utilization can be further improved.

[0338] If an AP inference report is associated with one or more AP monitoring reports, then an additional CSI-RS resource is counted, or A CSI-RS resources are counted.

[0339] If an AP inference report is associated with one or more AP monitoring reports, then the first CSI-RS resource is considered, and this first CSI-RS resource is counted 1 time, or A times. This can be understood as the first CSI-RS resource corresponding to the storage of that inference report.

[0340] It should be noted that counting rules one and two apply only to the implementation of P reasoning reports and / or SP reasoning reports, which is similar to the implementation of counting rules one and two applying only to AP reasoning reports. For the sake of brevity, this application will not repeat them here.

[0341] In some specific implementations, the prediction instances of multiple inference reports overlap. The following describes the counting rules for overlapping intervals.

[0342] Figure 11 is a schematic diagram of determining resource counts for overlapping intervals according to an embodiment of this application.

[0343] As an example and not a limitation, the figure shows a schematic diagram of overlapping prediction instances of multiple inference reports. This application does not specifically limit the specific overlapping relationship of the prediction instances.

[0344] In one specific implementation, the first resource corresponds to the configuration of the first inference report.

[0345] As an example and not a limitation, the figure illustrates the prediction instances contained in Inference Report 1 and Inference Report 2, as well as the process of monitoring the prediction instances on the reference signal (RS) resource and generating a monitoring report.

[0346] The figure shows predicted instances #1, #2, #3, and #4 of inference report 1 from the terminal device at times t1, t2, t3, and t4, and predicted instances #1, #2, #3, and #4 of inference report 2 from the terminal device at times t3, t4, t5, and t6. The inference results corresponding to predicted instances #3 and #4 of inference report 1 are overwritten by the inference results corresponding to predicted instances #1 and #2 of inference report 2.

[0347] Counting rule three: The first resource in the overlapping interval is counted as 1.

[0348] This can be understood as not performing an additional first resource count, i.e., updating based on the latest result. For cases where part of the inference result of Inference Report 1 is overwritten by part of the inference result of Inference Report 2, the first resource count for the overlapping interval is 1. Inference Report 2 includes inference reports that occur after Inference Report 1 in the time domain.

[0349] This counting rule requires minimal modification to the protocol, but some results in the inference report are updated by the next inference report in the time domain. Therefore, the terminal device cannot accurately and completely monitor the inference report.

[0350] If, in certain implementations, the UE requires accurate or complete monitoring and inference reports, then counting rule four can be used.

[0351] Counting rule four: Multiply the first resource count of the overlapping interval by (1+B), where B is the number of overlaps.

[0352] This can be understood as performing additional resource counting, meaning it will not overwrite or update based on the latest inference results.

[0353] Under this counting rule, the inference report will not be overwritten or updated, and the UE can accurately monitor the inference report.

[0354] In one specific implementation, the count of the first resource in the overlapping interval is C*(1+B). Here, C is the CSI-RS resource count corresponding to at least one inference result within the overlapping interval. Alternatively, C is the CSI-RS resource count corresponding to the time unit where at least one inference result within the overlapping interval is located. B is the number of overlaps of at least one inference result within the overlapping interval, and both B and C are natural numbers greater than or equal to 1.

[0355] For example, in the implementation shown in the attached figure, the overlapping interval includes the time intervals where t3 and t4 are located. The value of the first resource count C in the overlapping interval is 2. Inference report 1 and inference report 2 overlap once in the overlapping interval, so the value of B is 1.

[0356] It should be noted that the overlapping interval includes at least two inference reports or at least two inference results whose corresponding CSI-RS resources correspond to the same time unit within the current interval. This can also be understood as the resource activation intervals of at least two inference reports or at least two inference results corresponding to the same time unit within the overlapping interval.

[0357] It should be noted that counting rules three and four are not parallel counting rules to counting rules one and two, but rather further refinements of counting rules one and two when there is overlap in the predicted instances among multiple inference reports.

[0358] It should be noted that counting rules three and four can be combined with counting rules one and two to form new implementation methods, which will not be elaborated here.

[0359] The first resource counting has been described above using different counting rules. The first activation interval will be described below.

[0360] In one specific implementation, the rules for the first activation interval of the first resource include the following:

[0361] Activation Interval Rule 1: The start time of the first activation interval is the time when a Radio Resource Control (RRC) configuration is received, transmitted, or activated. This RRC configuration is used to configure inference reports. The end time of the first activation interval is the time when this RRC configuration is released.

[0362] In one specific implementation, if the network side does not configure a monitoring report for the inference report, the terminal device does not need to waste storage space to store the inference report. This can also be understood as the terminal device not storing the inference report. Therefore, the following activation interval rules can also be included.

[0363] Activation Interval Rule 2: The start time of the first activation interval is the time when the RRC configuration is received, sent, or activated. This RRC configuration is used to configure inference reports that are associated with at least one monitoring report. The end time of the first activation interval is the time when the RRC configuration is released.

[0364] It should be noted that the start time of the first activation interval is the time when the first RRC configuration is received, sent, or activated. The first RRC configuration is used to configure the first inference report. The first inference report is associated with at least one monitoring report.

[0365] This can also be understood as the first reasoning report being associated with at least one monitoring report.

[0366] It can also be understood that the first reasoning report corresponds to at least one monitoring report.

[0367] This can also be understood as at least one monitoring report being determined based on the first reasoning report.

[0368] It can also be understood that at least one monitoring report includes an identifier (ID) related to the first RRC configuration in its configuration information, such as the CSI Report Configuration (CSI-ReportConfig) ID sent by the network device to the terminal device.

[0369] Compared to Activation Interval Rule 1, Activation Interval Rule 2 avoids invalid storage of inference reports that are not linked to monitoring reports, thus improving storage resource utilization.

[0370] Activation interval rule three: The start time of the first activation interval is the first start time. The end time of the first activation interval is the first end time.

[0371] The first start time includes, but is not limited to, one of the following: the time when the inference report is submitted, the CSI reference resource for the inference report, or the time when the inference report is triggered (i.e., the end of the PDCCH containing the request).

[0372] The first end time includes, but is not limited to, one of the following: the end time of the effect of the inference result corresponding to the inference report, the end time of the prediction window corresponding to the inference report, the end time of the effect of one or more prediction results corresponding to the inference report, the completion time of the calculation of the performance metric of the monitoring report, the reporting time of the monitoring report, the CSI reference resource of the monitoring report, and the first interval after the first start time.

[0373] It should be noted that the end time of one or more prediction results corresponding to the inference report can be replaced by the end time of all inference results corresponding to the inference report. For example, if an inference report includes N4 inference results, the first end time can be the end time of the N5th inference result, or the first time interval after the end time of the N5th inference result, where N5 is less than or equal to N4, and can be indicated by the UE. Understandably, in the above scheme, the UE indicates the number of inference results that can be cached for a single inference report.

[0374] It should be noted that the reporting time of the inference report can be understood or replaced by at least one of the following: the time unit in which the uplink resource carrying the inference report is located, or the uplink time unit in which the inference report is located, or the first or last symbol of the physical uplink shared channel (PUSCH) or physical uplink control channel (PUCCH) used to carry the inference report, etc. This application does not impose any special restrictions on this.

[0375] It should be noted that the CSI reference resource in the inference report can be understood or replaced with the CSI reference resource corresponding to the inference report, or the reference resource corresponding to the inference report. This application does not impose any special restrictions on this.

[0376] It should be noted that the end time of the effect of the inference results corresponding to the inference report can be understood or replaced by at least one of the following: the end time of the prediction window corresponding to the inference report, or the prediction start time corresponding to the inference report + the number of prediction results * the prediction result interval. This application does not impose any special restrictions on this.

[0377] It should be noted that the completion time for calculating the performance metric in the monitoring report can be understood or replaced by at least one of the following: the completion time for calculating the performance metric in the monitoring report, or Z' symbols of the last symbol of the CSI-RS occasion for channel measurement corresponding to the monitoring report, wherein Z' can be predefined by the protocol or indicated by the UE. It is understood that Z' includes the time required to calculate the performance metric.

[0378] It should be noted that the reporting time of the monitoring report can be understood or replaced by at least one of the following: the time unit of the uplink resource carrying the monitoring report, or the uplink time unit of the monitoring report, or the first or last symbol of the PUSCH or PUCCH used to carry the monitoring report, etc. This application does not impose any special restrictions on this.

[0379] It should be noted that the CSI reference resource in the monitoring report can be understood or replaced with the CSI reference resource corresponding to the monitoring report, or the reference resource corresponding to the monitoring report. This application does not impose any special restrictions on this.

[0380] In one specific implementation, the end time of the first activation interval also includes a second interval before the aforementioned first end time, namely at least one of the following: a second interval before the end time of the effect of the inference result corresponding to the inference report, a second interval before the end time of the prediction window corresponding to the inference report, a second interval before the end time of the effect of one or more prediction results corresponding to the inference report, a second interval before the completion time of the calculation of the performance metric of the monitoring report, a second interval before the reporting time of the monitoring report, a second interval before the CSI reference resource of the monitoring report, and a second interval before the first interval after the first start time.

[0381] In one specific implementation, the end time of the first activation interval also includes a first time offset after the aforementioned first end time, namely at least one of the following: a first time offset after the end time of the effect of the inference result corresponding to the inference report, a first time offset after the end time of the prediction window corresponding to the inference report, a first time offset after the end time of the effect of one or more prediction results corresponding to the inference report, a first time offset after the completion time of the calculation of the performance metric of the monitoring report, a first time offset after the reporting time of the monitoring report, a first time offset after the CSI reference resource of the monitoring report, and a first time offset after the first interval following the first start time.

[0382] It should be noted that one or more of the first interval, second interval, and first time offset mentioned above are reported by the UE or predefined by the protocol.

[0383] Figure 12 is a schematic diagram of determining a resource activation range according to an embodiment of this application.

[0384] As an example and not a limitation, the figure illustrates a resource activation range in which an inference report is associated with a monitoring report. This application does not specifically limit the association between the inference report and the monitoring report.

[0385] As an example and not a limitation, the figure illustrates inference based on a first resource and an example of a prediction contained in the first inference report. The first resource is used for or corresponds to the measurement in the first inference report. The first resource is the resource associated with or triggered by the first DCI. The first resource includes RS resources.

[0386] The diagram also illustrates the process of monitoring predicted instances on the first resource, calculating performance metrics, and obtaining the first monitoring report.

[0387] As an example and not a limitation, if the first inference report is associated with the first monitoring report, then the first activation interval of the first inference report includes the period from the first start time to the first end time.

[0388] The first start time includes, but is not limited to, one of the following: the reporting time of the first inference report, the CSI reference resource of the first inference report, the triggering time of the first inference report, or the end of the first PDCCH containing the request.

[0389] The first end time includes, but is not limited to, one of the following: the end time of the effect of the inference result corresponding to the first inference report, the end time of the prediction window corresponding to the first inference report, the end time of the effect of one or more prediction results corresponding to the first inference report, the completion time of the calculation of the performance metric of the first monitoring report, the reporting time of the first monitoring report, the CSI reference resource of the first monitoring report, and the first interval after the first start time.

[0390] It should be noted that the end time of one or more prediction results corresponding to the first inference report can be replaced by the end time of all inference results corresponding to the first inference report. For example, if a first inference report includes N4 inference results, the first end time can be the end time of the N5th inference result, or the first time interval after the end time of the N5th inference result, where N5 is less than or equal to N4, and can be indicated by the UE. Understandably, in the above scheme, the UE indicates the number of inference results that can be cached for a single first inference report.

[0391] It should be noted that the reporting time of the first inference report can be understood or replaced by at least one of the following: the time unit in which the uplink resource carrying the first inference report is located, or the uplink time unit in which the first inference report is located, or the first or last symbol of the PUSCH or PUCCH used to carry the first inference report, etc. This application does not impose any special restrictions on this.

[0392] It should be noted that the CSI reference resource corresponding to the first inference report can be understood or replaced as the CSI reference resource corresponding to the first inference report, or the reference resource corresponding to the first inference report, and this application does not impose any special restrictions on this.

[0393] It should be noted that the end time of the effect of the inference result corresponding to the first inference report can be understood or replaced by at least one of the following: the end time of the prediction window corresponding to the first inference report, or the prediction start time corresponding to the first inference report + the number of prediction results * the prediction result interval. This application does not impose any special restrictions on this.

[0394] It should be noted that the calculation time of the performance metric in the first monitoring report can be understood or replaced as: the completion time of the performance metric calculation in the first monitoring report, or the Z' symbols of the last symbol of the CSI-RS occasion for channel measurement corresponding to the first monitoring report, wherein Z' can be predefined by the protocol or indicated by the UE. It is understood that Z' includes the time required to calculate the performance metric.

[0395] It should be noted that the reporting time of the first monitoring report can be understood or replaced by at least one of the following: the time unit in which the uplink resource carrying the first monitoring report is located, or the uplink time unit in which the first monitoring report is located, or the first or last symbol of the PUSCH or PUCCH used to carry the first monitoring report, etc. This application does not impose any special restrictions on this.

[0396] It should be noted that the CSI reference resource corresponding to the first monitoring report can be understood or replaced as the CSI reference resource corresponding to the first monitoring report, or the reference resource corresponding to the first monitoring report. This application does not impose any special restrictions on this.

[0397] In one specific implementation, the end time of the first activation interval also includes a second interval before the aforementioned first end time, namely at least one of the following: a second interval before the end time of the effect of the inference result corresponding to the first inference report, a second interval before the end time of the prediction window corresponding to the first inference report, a second interval before the end time of the effect of one or more prediction results corresponding to the first inference report, a second interval before the completion time of the calculation of the performance metric of the first monitoring report, a second interval before the reporting time of the first monitoring report, a second interval before the CSI reference resource of the first monitoring report, and a second interval before the first interval after the first start time.

[0398] In one specific implementation, the end time of the first activation interval also includes a first time offset after the aforementioned first end time, namely at least one of the following: a first time offset after the end time of the effect of the inference result corresponding to the first inference report, a first time offset after the end time of the prediction window corresponding to the first inference report, a first time offset after the end time of the effect of one or more prediction results corresponding to the first inference report, a first time offset after the completion time of the calculation of the performance metric of the first monitoring report, a first time offset after the reporting time of the first monitoring report, a first time offset after the CSI reference resource of the first monitoring report, and a first time offset after the first interval following the first start time.

[0399] It should be noted that one or more of the first interval, second interval, and first time offset mentioned above are reported by the UE or predefined by the protocol.

[0400] As an example and not a limitation, the first end time is the first time offset after the end time of the prediction window corresponding to the first inference report. This first time offset is reported by the UE or predefined by the protocol. This first time offset corresponds to the processing latency of the first monitoring resource corresponding to the first monitoring report. For example, the first time offset corresponds to the processing latency of calculating the first monitoring resource and / or the performance metrics. It can also be understood as the interval duration for which the UE needs to store additional data, used to ensure the completion of the UE's performance metric calculations.

[0401] In some alternative implementations, resource activation is performed according to activation interval rule three when the inference report is associated with at least one monitoring report.

[0402] In some specific implementations, the P / SP inference report is executed according to one of the following: activation interval rule one, activation interval rule two, or activation interval rule three.

[0403] As an example rather than a limitation, the following description focuses on the execution of activation interval rule three in P / SP inference reports.

[0404] Figure 13 is a schematic diagram of determining the resource activation range of a P / SP inference report according to an embodiment of this application.

[0405] As an example and not a limitation, the diagram illustrates a resource activation range for a P / SP inference report and its associated monitoring report. When storing the inference report according to activation range rule three, the storage of the inference report is divided into multiple segments.

[0406] For the P inference report, the start time of the first activation interval is the first start time. The end time of the first activation interval is the first end time.

[0407] It should be understood that the contents of the first start time and the first end time are consistent with those in the above-mentioned activation interval rule three. For the sake of brevity, this application will not repeat them here.

[0408] It should be understood that, for the P inference report, the first activation interval can also be executed according to activation interval rule one or activation interval rule two. For the sake of brevity, this application will not elaborate further here.

[0409] For SP inference reports, the activation time rule for storage counts is one of activation interval rule one, activation interval rule two, or activation interval rule three, and the activation interval definitions of other SP resources can also be referenced. For example, activation occurs when the activation command is applied, and deactivation occurs when the deactivation command is applied. The activation command corresponds to the activation of the SP inference report, and the deactivation command corresponds to the deactivation of the SP inference report.

[0410] For AP inference reports, the activation time rule for storage counts is one of activation interval rule one, activation interval rule two, or activation interval rule three.

[0411] However, because AP inference reports can only correspond to AP monitoring reports, and the arrival time of the downlink control information (DCI) that triggers the AP monitoring report is uncertain, the following activation interval rules can also be executed for AP inference reports:

[0412] Activation Interval Rule 4: When the AP monitoring report is submitted later than the activation interval described in the above rules, the resource activation end time is the AP monitoring report submission time or the CSI reference resource.

[0413] It should be noted that if the network-side device does not trigger the AP monitoring report, the resource activation interval can be one of the activation interval rules one to three mentioned above. If the network-side device triggers the AP monitoring report, and / or the AP monitoring report is submitted after the activation interval described in the above rules, the resource activation end time is the submission time of the AP monitoring report or the CSI reference resource of the AP monitoring report.

[0414] Figure 14 is a schematic diagram of determining the resource activation range of an AP inference report according to an embodiment of this application.

[0415] As an example and not a limitation, the figure illustrates a schematic diagram of the resource activation range of an AP inference report and its associated monitoring report.

[0416] Taking the resource activation interval of the AP inference report as the first activation interval as an example, the activation interval rules of the AP inference report are described. The start time of the first activation interval is the first start time, and the end time of the first activation interval is the first end time.

[0417] The first start time includes, but is not limited to, one of the following: the reporting time of the AP inference report, the CSI reference resource of the AP inference report, the triggering time of the AP inference report, or the end of the first PDCCH containing the request.

[0418] The first end time includes, but is not limited to, one of the following: the end time of the effect of the inference result corresponding to the AP inference report, the end time of the prediction window corresponding to the AP inference report, the end time of the effect of one or more prediction results corresponding to the AP inference report, the completion time of the calculation of the performance metric of the AP monitoring report, the reporting time of the AP monitoring report, the CSI reference resource of the AP monitoring report, and the first interval after the first start time.

[0419] It should be noted that the end time of one or more prediction results corresponding to an AP inference report can be replaced by the end time of all inference results corresponding to the AP inference report. For example, if an AP inference report includes N4 inference results, the first end time can be the end time of the N5th inference result, or the first time interval after the end time of the N5th inference result, where N5 is less than or equal to N4, and can be indicated by the UE. Understandably, in the above scheme, the UE indicates the number of inference results that can be cached for one AP inference report.

[0420] As an example, and not a limitation, the first end time is the first time offset after the end time of the prediction window corresponding to the AP inference report. This first time offset is reported by the UE or predefined by the protocol. This first time offset corresponds to the processing latency of the first monitoring resource corresponding to the AP monitoring report. For example, the first time offset corresponds to the processing latency of calculating the first monitoring resource and / or the performance metrics. It can also be understood as the interval duration for which the UE needs to store additional data, and the first time offset is used to ensure that the UE's performance metrics calculation is completed.

[0421] In one specific implementation, if the AP monitoring report is submitted later than the end time of the effect of the inference result corresponding to the AP inference report, the end time of the prediction window corresponding to the AP inference report, or the calculation time of the performance metric of the AP monitoring report, the first end time is the submission time of the AP monitoring report or the CSI reference resource.

[0422] Based on the aforementioned activation interval rules, the UE does not need to continuously store AP inference reports. This also avoids the problem of the network side being unable to keep up with the monitoring of computing performance indicators due to insufficient time to store AP inference results.

[0423] Activation interval rule five: Determine the activation interval corresponding to the AP inference report through the configuration behavior on the network side.

[0424] In one specific implementation, if the downlink control information (DCI) corresponding to the performance monitoring is earlier than the reporting time of the inference report, then the UE stores the inference report; otherwise, the UE does not store the inference report.

[0425] In one specific implementation, the DCI information used for performance monitoring is submitted earlier than the inference report, and the UE's first resource count within the first active interval is incremented by A. The first active interval can be executed according to the above-mentioned active interval rule one, active interval rule two, active interval rule three, or active interval rule four.

[0426] In one specific implementation, if the DCI (Distributed Inference Code) for performance monitoring occurs earlier than the inference report submission time, the CSI-RS resource count is incremented by one (or by a value A). The activation interval of this resource is determined based on one of the activation interval rules 1, 2, 3, or 4 mentioned above.

[0427] It should be noted that the DCI corresponding to performance monitoring in this application can also be replaced or understood as at least one of the following: DCI used to trigger performance monitoring reports, or control information used to trigger performance monitoring reports, or the time unit in which the control information used to trigger performance monitoring reports is located. This application does not make any special limitations on this.

[0428] It should be noted that the DCI in this application can also be replaced by the physical downlink control channel (PDCCH) or the medium access control element (MAC-CE), etc., and this application does not make any special restrictions on it.

[0429] Figure 15 is a schematic diagram of another method for determining the resource activation range of the AP inference report provided in an embodiment of this application.

[0430] As an example and not a limitation, the figure illustrates a scenario where the AP inference report is submitted earlier than the DCI for performance monitoring.

[0431] In one specific implementation, the AP's inference report is submitted earlier than the DCI's reception time, and the UE does not store the inference report.

[0432] It should be noted that the activation time of the resources described in this application can be understood as: the activation interval of the resources, or the start and end time of resource activation.

[0433] Based on the aforementioned activation interval rules, the network side can reasonably configure monitoring reports based on the terminal device's resource activation capabilities (which can also be understood as storage capabilities). Furthermore, this application can also limit the network side's monitoring behavior based on the UE's capabilities.

[0434] In one specific implementation, the terminal device does not expect the network device to perform the calculation and reporting of monitoring reports outside the resource activation range.

[0435] This can also be understood as the UE not expecting the monitoring report to be submitted after the resource activation end time, or the UE not expecting the reference resources for the monitoring report to be submitted after the resource activation end time.

[0436] This can also be understood as follows: if the network device is configured to report monitoring reports outside the resource activation range, the UE has discarded the inference reports / inference results associated with the monitoring report, so the UE cannot calculate the performance metrics.

[0437] This can also be understood as the terminal device sending instruction information 1 to the network device. Instruction information 1 instructs the network device to configure the monitoring report based on the terminal device's capabilities. The terminal device's capabilities include its resource activation capabilities or resource storage capabilities. The network device configures the monitoring report, including the inference report / inference results associated with the configuration monitoring report, and may also include the reporting and / or calculation time of the configuration monitoring report.

[0438] Figure 16 is a schematic diagram of another method for determining the resource activation range provided in an embodiment of this application.

[0439] In one specific implementation, the monitoring resources used for monitoring reports need to be within the resource activation range. This can also be understood as determining the monitoring resources used for monitoring reports based on the resource activation range.

[0440] Since the last monitored resource shown in the diagram is not within the resource activation range, the timing of its transmission is not used for performance monitoring.

[0441] In summary, the standardization of inference report storage is achieved by enhancing the activation resource counting rules. This allows terminal devices to pre-store the corresponding inference results for monitoring reports. Network-side devices and terminal devices can align the activation resource counts of terminal devices, improving the utilization rate of activation resources on the terminal device side.

[0442] In some specific implementations, the storage of inference results can be bound to / associated with the active resource count of the inference resource. This can be understood as storing the inference results based on the RS buffer of the inference resource.

[0443] This can be understood as the first resource being the first reasoning resource, and the first reasoning resource being the measurement resource corresponding to the first reasoning report.

[0444] In one specific implementation, the first inference resource includes the measurement resource corresponding to the inference report. The UE makes a prediction based on the first inference resource and then generates an inference result.

[0445] If the time-domain type of the first inference resource is P / SP, then the count of the first inference resource is incremented by 1 or by a value A (i.e., the count is Kp+1 or Kp+A). Kp can be understood as the number of observation resources corresponding to the inference report.

[0446] In one specific implementation, if an inference report is associated with one or more monitoring reports, the measurement resource count corresponding to the inference report is incremented by 1 or by a value A (i.e., the count is Kp+1 or Kp+A).

[0447] It should be noted that in some specific implementations, the measurement resources corresponding to the inference report are also referred to as measurement resources, and this application does not make any special limitation on this.

[0448] This can be understood as the terminal device needing to store measurement information for Kp resources for observation in order to generate an inference report.

[0449] This can be understood as the terminal device needing to consume Kp primary resources to obtain an inference report in order to acquire the measurement information associated with the inference report.

[0450] This can be understood as the additional 1 or A first resources added to the measurement resource count corresponding to the inference report are used to store one inference report.

[0451] This can be understood as follows: an inference report is used for subsequent performance monitoring, and storing an inference report requires A first resources, where A is a natural number greater than or equal to 1.

[0452] When the time-domain type of the measurement resource corresponding to the inference report is P / SP, the resource activation interval of the first inference resource can be consistent with the activation time of P / SP in the legacy, or it can be consistent with one or more of the activation times in the activation interval rules one to five mentioned above. For the sake of brevity, this application will not elaborate further here.

[0453] If the temporal type of the first inference resource is AP, then the count of the first inference resource is incremented by 1 or by the value A.

[0454] In one specific implementation, if an inference report is associated with one or more monitoring reports, the measurement resource count corresponding to the inference report is incremented by 1 or by a value A.

[0455] When the time-domain type of the measurement resource corresponding to the inference report is AP, the resource activation interval of the first inference resource can be consistent with the activation time of AP in the legacy, or it can be consistent with one or more activation times of the activation interval rules one to five mentioned above. For the sake of brevity, this application will not elaborate further here.

[0456] In one specific implementation, the activation start time of the first inference resource includes the last time-domain unit of the PDCCH that triggers the first inference resource.

[0457] In one specific implementation, the activation end time of the first inference resource includes a first time offset after the end time of the prediction window corresponding to the inference report. This first time offset can be reported by the UE or predefined by the protocol.

[0458] In summary, by binding the activation resource count of inference resources to achieve the storage of inference reports, terminal devices can pre-store the corresponding inference results for monitoring reports. Furthermore, network-side devices and terminal devices can align the activation resource counts of terminal devices, improving the utilization rate of activation resources on the terminal device side.

[0459] In some specific implementations, the storage of inference results can be bound to / associated with the activation resource count of the monitored resource. This can be understood as storing the inference results based on the RS buffer of the monitored resource.

[0460] This can be understood as the first resource being the first monitoring resource, and the first monitoring resource being the measurement resource corresponding to the first monitoring report.

[0461] In one specific implementation, the first monitoring resource includes the measurement resource corresponding to the monitoring report, and the UE calculates the monitoring performance index based on the measurement information and inference results of the first monitoring resource.

[0462] If the time domain type of the first monitored resource is P / SP, then the count of the first monitored resource is incremented by 1 or by a value A.

[0463] In one specific implementation, if an inference report is associated with one or more monitoring reports, the measurement resource count corresponding to the monitoring report is incremented by 1 or by a value A.

[0464] This can be understood as follows: in addition to monitoring the measurement information of the resources, the terminal device also needs to store the inference results for subsequent performance monitoring in order to generate a monitoring report.

[0465] When the time-domain type of the measurement resource corresponding to the monitoring report is P / SP, the resource activation interval of the first monitoring resource can be consistent with the activation time of P / SP in the legacy, or it can be consistent with one or more of the activation times in the activation interval rules one to five mentioned above. For the sake of brevity, this application will not elaborate further here.

[0466] If the time domain type of the first monitored resource is AP, then the count of the first monitored resource is incremented by 1 or by the value A.

[0467] In one specific implementation, if an inference report is associated with one or more monitoring reports, the measurement resource count corresponding to the monitoring report is incremented by 1 or by a value A.

[0468] In one specific implementation, if the terminal device receives the DCI of the monitoring report before the inference report is submitted, the activation start time of the first monitoring resource can refer to activation interval rules one to five in the above implementation, and be consistent with one or more of the activation start times in activation interval rules one to five in the above implementation. Otherwise, the activation start time is the trigger time of the monitoring report.

[0469] In one specific implementation, if the terminal device receives the DCI of the monitoring report before the inference report is reported, then the activation start time of the first monitoring resource is the inference report reporting time. Otherwise, the activation start time of the first monitoring resource is the end of the PDCCH containing the request.

[0470] In one specific implementation, the resource activation start time of the first monitored resource can be consistent with the activation start time of the AP in the legacy system. For the sake of brevity, this application will not elaborate further.

[0471] In one specific implementation, it is also necessary to limit the monitoring configuration information on the network side. For example, the trigger time for AP monitoring reports should be configured to be no later than the reporting time of the associated inference reports.

[0472] In one specific implementation, the resource activation start time of the first monitored resource is the trigger time of the monitoring report, which can also be understood as the end of the PDCCH containing the request.

[0473] When the time domain type of the measurement resource corresponding to the monitoring report is AP, the resource activation end time of the first monitoring resource can be consistent with the activation end time of P / SP in the legacy, or it can be consistent with one or more of the activation end times of the activation interval rules one to five mentioned above. For the sake of brevity, this application will not elaborate further here.

[0474] In one specific implementation, the resource activation end time of the first monitored resource includes the reporting time of the monitoring report.

[0475] In summary, this application achieves the storage of inference reports by binding the activation resource count of monitoring resources, enabling terminal devices to store the corresponding inference results in advance for monitoring reports. Furthermore, network devices and terminal devices can align the activation resource counts of terminal devices, improving the utilization rate of activation resources on the terminal device side.

[0476] In the legacy protocol, the storage size and duration of CSI-RS measurement results are reflected through resource counts and activation times. These CSI-RS measurement results include, but are not limited to, codebooks, precoding matrices, and raw channel matrices.

[0477] In one specific implementation, the UE needs to store inference reports / results in a separate storage space. This can be understood as the UE's storage space / hardware unit for inference reports / results being inconsistent with the storage space / hardware unit for CSI-RS measurement results. In this implementation, the storage size and time cannot be represented by the active resource count in the legacy protocol. Furthermore, for AP-type CSI reports, the UE will not store CSI-RS measurement results based on legacy protocol rules. Therefore, this application introduces a new UE storage capability to reflect the storage size and time of inference reports / results.

[0478] In one specific implementation, the terminal device sends a first indication message to the network device. This can also be understood as the terminal device reporting its capabilities using the first indication message.

[0479] In one specific implementation, the first indication information is used to indicate the maximum number of storage units that the UE can support. This can also be understood as the UE reporting the maximum number of storage units it can support.

[0480] It should be noted that the storage unit corresponds to the storage of inference results or the storage of inference reports.

[0481] It should be noted that the storage unit can also store other AI data. For example, it can store performance metrics and training data.

[0482] In one specific implementation, for AP inference reports, the UE can also report the maximum number of storage units corresponding to the largest AP inference report that it can support, or the UE can also report the maximum number of storage units corresponding to the inference results of the largest AP inference report that it can support.

[0483] In one specific implementation, for P / SP inference reports, the UE can also report the maximum number of storage units corresponding to the largest P / SP inference report that it can support, or the UE can also report the maximum number of storage units corresponding to the inference results of the largest P / SP inference report that it can support.

[0484] It should be noted that the number of storage units mentioned in the implementation of this application can be replaced by at least one of the following: number of storage units, number of storage units, storage unit time, storage time, number of resources, and number of activated resources.

[0485] It should be noted that if the UE reports the maximum storage time, then the activation interval rule one mentioned above does not need to be considered. This can be understood as the UE reporting a static capability, and the network side can configure the monitoring report appropriately based on the UE's capability.

[0486] In one specific implementation, the UE reports or the protocol normalizes the number of storage units corresponding to each inference report / inference result.

[0487] In another specific implementation, if the UE does not report or the protocol does not standardize the number of storage units corresponding to each inference report / inference result, the default storage unit count value is 1, or the storage unit time is the storage activation time described below.

[0488] In one specific implementation, the UE and NW determine whether the UE's maximum storage capacity has been exceeded based on the maximum number of storage units, the number of storage units corresponding to each inference report / result, and the storage activation interval. If the UE's maximum storage capacity has been exceeded, the UE is not required to update the monitoring report corresponding to the inference report / result.

[0489] In one specific implementation, the rules for the activation interval of the storage count include the following:

[0490] Storage Activation Interval Rule 1: The start time of the storage activation interval is the time when the radio resource control (RRC) configuration is received. This RRC configuration is used to configure inference reports. The end time of the storage activation interval is the time when this RRC configuration is released.

[0491] Storage Activation Interval Rule 2: The start time of the storage activation interval is the time when the RRC configuration is received. This RRC configuration is used to configure inference reports that are associated with at least one monitoring report. The end time of the storage activation interval is the time when the RRC configuration is released.

[0492] It should be noted that the start time of the storage activation interval is the time when the first RRC configuration is received. The first RRC configuration is used to configure the first inference report. The first inference report is associated with at least one monitoring report.

[0493] Storage activation interval rule three: The start time of the storage activation interval is the first start time. The end time of the storage activation interval is the first end time.

[0494] The first start time includes, but is not limited to, one of the following: the time when the inference report is submitted, the CSI reference resource for the inference report, or the time when the inference report is triggered (i.e., the end of the PDCCH containing the request).

[0495] The first end time includes, but is not limited to, one of the following: the end time of the effect of the inference result corresponding to the inference report, the end time of the prediction window corresponding to the inference report, the end time of the effect of one or more prediction results corresponding to the inference report, the completion time of the calculation of the performance metric of the monitoring report, the reporting time of the monitoring report, the CSI reference resource of the monitoring report, and the first interval after the first start time.

[0496] It should be noted that the end time of one or more prediction results corresponding to the inference report can be replaced by the end time of all inference results corresponding to the inference report. For example, if an inference report includes N4 inference results, the first end time can be the end time of the N5th inference result, or the first time interval after the end time of the N5th inference result, where N5 is less than or equal to N4, and can be indicated by the UE. Understandably, in the above scheme, the UE indicates the number of inference results that can be cached for a single inference report.

[0497] Storage Activation Interval Rule 4: When the AP monitoring report is submitted later than the storage activation interval described in the above rules, the end time of the storage activation interval shall be the submission time of the AP monitoring report or the CSI reference resource.

[0498] It should be noted that if the network-side device does not trigger an AP monitoring report, the storage activation interval can be one of the storage activation interval rules one to three mentioned above. If the network-side device triggers an AP monitoring report, and / or the AP monitoring report is submitted after the storage activation interval described in the above rules, the storage activation end time is the submission time of the AP monitoring report or the CSI reference resource of the AP monitoring report.

[0499] Storage activation interval rule five: Determine the storage activation interval corresponding to the AP inference report through network-side configuration behavior.

[0500] It should be noted that the further explanations and extensions of the rules in Rules 1 to 5 of the aforementioned storage activation interval can be understood as or referred to in the descriptions in Rules 1 to 5 of the aforementioned activation interval. For the sake of brevity, this application will not elaborate further here.

[0501] It should be noted that, in the implementation of this application, the activation time of the storage count can be understood as: the storage activation interval, the time during which the storage unit counts, or the start and end time of the storage unit counting.

[0502] In summary, based on the technical solution of this application, the network-side equipment can reasonably configure monitoring reports based on the UE's storage capacity.

[0503] For CSI prediction performance monitoring, UE-assisted performance monitoring with additional UE capabilities is supported, and the UE-assisted performance monitoring is based on Type 3 performance monitoring.

[0504] The steps for performance monitoring are described below with reference to the accompanying diagram.

[0505] Figure 17 is a schematic diagram of the performance monitoring steps provided in an embodiment of this application.

[0506] The main steps of Type 3 performance monitoring include: UE calculating performance metrics; UE reporting performance metrics to NW; and NW making decisions such as function rollback operations.

[0507] For performance monitoring Type 3, square generalized cosine similarity (SGCS) is supported as a performance metric. SGCS is defined as follows: For a given layer l∈{1,…,v}, subband n3∈{1,…,N3}, and prediction instance n4∈{1,…,N4}, SGCS is defined as:

[0508] in, It is a prediction precoder represented by PMI, which is used for inference of the i-th layer, the n3-th subband, and the n4-th prediction instance; The precoder is represented by a PMI, which is used for the ground-truth label of the predicted instance in the i-th layer, the n3-th subband, and the n4-th subband. How to handle layer mapping mismatches (if any) depends on the UE implementation.

[0509] For the report content of UE assisted performance monitoring, please select from the following options (down-select):

[0510] Alt1: An SGCS, which is calculated based on the predicted CSI and ground-truth CSI of an inference report.

[0511] Alt2: One SGCS is calculated based on the predicted CSI and ground-truth CSI from an inference report; the other SGCS is calculated based on the ground-truth CSI and the CSI corresponding to the most recent CSI-RS transmission occasion from a CSI reference resource no later than the inference report instance.

[0512] Alt3: Within the configured monitoring window, statistics of the inference report instances (e.g., mean, x%CDF of SGCS values). The monitoring window can be configured by NW, and the signaling details are FFS.

[0513] Does FFS report each predicted instance, the selected predicted instances(s), and the average of the predicted instances?

[0514] FFS reports frequency granularity (e.g., per broadband or per subband, or selected subbands or average values ​​over selected subbands);

[0515] Does FFS report the average values ​​for each layer, the first layer, and all layers?

[0516] How FFS quantifies and reports.

[0517] Understandably, for SGCS calculation, the UE needs to convert the Doppler codebook of the inference report into a precoding matrix W (corresponding to inverse calculation A), convert the measurement results of the monitoring CSI-RS into a codebook, and then convert the measurement codebook into a precoding matrix W' (corresponding to inverse calculation B). Then, SGCS is calculated using W and W'. In other words, unlike the legacy CSI-RS measurement process, for performance monitoring, the UE needs to undergo multiple inverse calculation processes to obtain the final metric.

[0518] Based on this, the technical solution disclosed in this application further includes the CPU count value of the monitoring report.

[0519] In one specific implementation, the count value of the third processing resource corresponding to the second monitoring report is determined based on the second monitoring report. The second monitoring report is associated with one or more inference reports.

[0520] In one specific implementation, the third processing resource includes, but is not limited to, processing units such as CPU and APU.

[0521] In one specific implementation, the count value of the third processing resource corresponding to the second monitoring report is related to the count value of the third processing resource corresponding to the first measurement resource and / or the first monitoring factor; the first measurement resource is used for channel measurement corresponding to the second monitoring report.

[0522] In one specific implementation, the CPU count value corresponding to the second monitoring report is: O_cpu = O_cpu_measurement * factor.

[0523] In one specific implementation, the count value of the third processing resource corresponding to the second monitoring report is also associated with the first count value. The first count value is indicated by the first device, or the first count value includes a predefined count value, or the first count value is associated with at least one of the following: inference result inverse calculation, monitoring inverse calculation, and squared generalized cosine similarity (SGCS) calculation.

[0524] In another specific implementation, the CPU count value corresponding to the second monitoring report is: O_cpu = O_cpu_measurement + O_cpu_m * factor.

[0525] Wherein, O_cpu is the count value of CPU resources corresponding to the second monitoring report. The value of O_cpu_m is indicated by the terminal device, or it can be understood that the value of O_cpu_m is related to one of the following: inference result inverse calculation, monitoring inverse calculation, and SGCS calculation.

[0526] In this context, the value of O_cpu_measurement is Ks, meaning O_cpu_measurement = Ks. Ks represents the number of resources within a resource set. Specifically, in CSI prediction, O_cpu_measurement = 1.

[0527] It should be noted that the value of O_cpu_m and the value of O_cpu_measured can be the same or different, and this application does not impose any special restrictions on this.

[0528] In one specific implementation, the first monitoring factor is indicated by a first device, or the first monitoring factor includes predefined monitoring factors, or the first monitoring factor has a corresponding relationship with a first count value.

[0529] It should be noted that "factor" is a specific implementation of the first monitoring factor in this application. The factor is reported by the UE or predefined by the protocol, and can be understood as the inverse calculation process of the codebook corresponding to the factor.

[0530] In one specific implementation, the factors corresponding to the inference result back calculation, and / or monitoring back calculation, and / or SGCS calculation are inconsistent. That is, O_cpu = O_cpu_measurement + O_cpu_m1*factor1 + O_cpu_m2*factor2 + O_cpu_m3*factor3.

[0531] In summary, this application takes into account the multiple codebook transformations in the SGCS calculation process, enabling the NW and UE to accurately align the use of computing resources and improve the utilization rate of computing resources.

[0532] In summary, this paper introduces new UE storage capabilities for inference reports, enabling UEs to store corresponding inference results in advance for monitoring reports; by defining storage rules for inference reports, the NW and UE can align UE storage resources, thereby improving the utilization rate of storage resources on the UE side.

[0533] In communication systems, the network side schedules downlink data transmission at the terminal side based on channel state information (CSI). CSI is information reflecting channel characteristics and quality, which can be obtained by the terminal side by measuring reference signals transmitted by the network side. The terminal side generates CSI reports according to protocol predefined methods or network side configurations and feeds them back to the network side for downlink data transmission scheduling. Processing CSI reports requires resources, such as a CSI processing unit (CPU). Terminal devices with different capabilities can support a different number of CSI reports that can be processed simultaneously; therefore, each terminal device uses its CPU for CSI report processing according to its own capabilities.

[0534] With the increasing application of artificial intelligence (AI) in air interface tasks, AI-based CSI reports, such as those containing CSI data obtained through inference based on AI models, can also be used for CSI reporting. Processing these CSI reports also requires resources. However, existing protocols only consider the CPU usage rules for traditional CSI reports. The resource consumption during the processing of both traditional and AI-based CSI reports is a problem that urgently needs to be addressed.

[0535] Based on this, this application provides a communication method for processing CSI reports, offering a solution to the resource consumption of different types of CSI reports and helping to improve the utilization of resources used for CSI report processing.

[0536] The CSI reports involved in this application embodiment include Type #2 CSI reports and Type #1 CSI reports. Type #1 CSI reports refer to CSI reports whose processing necessarily involves AI model inference or other AI model-related processes, i.e., they necessarily involve the use of second-type resources. In contrast, Type #2 CSI reports refer to CSI reports that do not involve AI model inference or other AI model-related processes; therefore, Type #2 CSI reports do not involve the use of second-type resources, but only the use of first-type resources.

[0537] For example, even if a CSI report is AI-related (e.g., a report for AI model performance monitoring that does not involve the AI ​​inference process but only the measurement process of reference signals used for model performance monitoring), but does not involve AI inference (i.e., does not involve the use of second-type resources), the CSI report is still classified as type #2.

[0538] In this application, a first processing resource and a second processing resource are defined. The first processing resource can refer to an AI / ML-based CSI processing unit, and the second processing resource can be a non-AI / ML-based CSI processing unit. As an example, the first processing resource can be an AI / MI-based CSI processing unit (hereinafter referred to as APU), which can be replaced by at least one of the following: ACPU, AI CPU, or enhanced CPU, etc. The second processing resource can be a CSI processing unit (CPU). In another example, the second processing resource can be an AI / MI-based CSI processing unit (hereinafter referred to as APU), which can be replaced by at least one of the following: ACPU, AI CPU, or enhanced CPU, etc.; the first processing resource can be a CSI processing unit (CPU).

[0539] In one specific implementation, when a CSI report has insufficient APU resources, the lower-priority CSI report will not be updated even if the APU resources of a lower-priority CSI report are sufficient.

[0540] This can also be understood as follows: based on the priority of all CSI reports of type #1 from high to low, APU resources are consumed. When the APU resource demand of a certain type #1 CSI report is insufficient, all subsequent low-priority type #1 CSI reports will not be updated.

[0541] The following examples, with reference to the accompanying diagrams, illustrate this point.

[0542] Figure 18 is a schematic diagram of the performance monitoring steps provided in an embodiment of this application.

[0543] Assume the priority of CSI reports is report 1 > report 2 > report 3 > report 4 > report 5 > report 6 > report 7 > report 8. Among them, report 1, report 3, and report 5 are type #1 CSI reports, and report 2, report 4, report 6, report 7, and report 8 are type #2 CSI reports.

[0544] Here, we assume that the APU or CPU usage values ​​for each report are as follows, where O_APU represents the APU usage value and O_CPU represents the CPU usage value.

[0545] Reports for type #1: report1(O_APU=3, O_CPU=1), report3(O_APU=3, O_CPU=1), report5.(O_APU=2, O_CPU=1).

[0546] For reports of type #2: report2, report4, report6, report7, report8, O_CPU = 1.

[0547] Suppose that on a symbol, the available CPU resource utilization is 7 and the available APU resource utilization is 5.

[0548] In one specific implementation, CPU and APU are allocated on the current symbol according to the priority of all CSI reports from highest to lowest. When the CPU or APU (if any) for a certain CSI report is insufficient, CSI reports with lower priority than the CSI report will not be updated.

[0549] Specifically, the CPU and APU usage process can be summarized as follows:

[0550] - For report 1, the CPU resources and APU resources can meet the needs of report 1. Therefore, report 1 occupies the corresponding CPU and APU resources, that is, O_cpu1 = 1 and O_apu1 = 3.

[0551] - For report 2, the CPU resources can meet the needs of report 2; therefore, report 2 occupies the corresponding CPU usage value, that is, O_cpu2 = 1.

[0552] - For report 3, CPU resources are sufficient to meet its requirements, but APU resources are insufficient. Therefore, according to the current implementation's occupancy rules, report 3's CPU occupancy is O_cpu3 = 1, and O_apu3 = 0 (the O_apu3 = 3 shown in the diagram indicates that report 3 requires 3 APUs, but the actual occupancy is 0).

[0553] Therefore, in the current example, report 3 does not utilize the APU and is thus not updated. If the APU requirement for report 3 is not met, the lower-priority reports 4, 5, 6, 7, and 8 will not be updated.

[0554] Figure 19 is a schematic diagram of another performance monitoring step provided in an embodiment of this application.

[0555] In one specific implementation, on the current symbol, CPU is used according to the priority of all CSI reports. When the CPU for a certain CSI report is insufficient, CSI reports with lower priority than the CSI report will not be updated. APU is used according to the priority of all CSI reports of type #1. When the APU for a certain CSI report is insufficient, CSI reports of type #1 with lower priority than the CSI report will not be updated.

[0556] Specifically, the CPU and APU usage process can be summarized as follows:

[0557] For CPU usage, in order of priority: report 1>report 2>report 3>report 4>report 5>report 6>report 7>report 8.

[0558] For the APU, the priority order for usage is: report 1 > report 3 > report 5.

[0559] - For report 1, the CPU resources and APU resources can meet the needs of report 1. Therefore, report 1 occupies the corresponding CPU and APU resources, that is, O_cpu1 = 1 and O_apu1 = 3.

[0560] - For report 2, the CPU resources can meet the needs of report 2; therefore, report 2 occupies the corresponding CPU usage value, that is, O_cpu2 = 1.

[0561] - For report 3, CPU resources are sufficient to meet its requirements, but APU resources are insufficient. Therefore, according to the current implementation's occupancy rules, report 3's CPU and APU occupancy values ​​are O_cpu3 = 1 and O_apu3 = 0 (the O_apu3 = 3 shown in the diagram indicates that report 3 requires 3 APUs, but the actual occupancy value is 0). Since report 3 does not occupy any APUs, it is not updated.

[0562] - For report 4, the CPU resources can meet the needs of report 4; therefore, report 4 occupies the corresponding CPU usage value, that is, O_cpu4 = 1.

[0563] - For report 5, since report 3 is not updated, report 5 will also not be updated according to the current occupancy rules. Therefore, the CPU and APU occupancy of report 5 is O_cpu5 = 1 and O_apu5 = 0 (the O_apu5 = 2 shown in the figure means that report 5 requires 2 APUs, but the actual occupancy value is 0).

[0564] - For report 6, the CPU resources can meet the needs of report 6; therefore, report 6 occupies the corresponding CPU usage value, that is, O_cpu6 = 1.

[0565] - For report 7, the CPU resources can meet the needs of report 7; therefore, report 7 occupies the corresponding CPU usage value, that is, O_cpu7 = 1.

[0566] - For report 8, CPU resources are insufficient to meet its requirements; therefore, according to the current implementation's CPU usage rules, the CPU usage value for report 8 is O_cpu8 = 0 (the O_cpu8 = 1 shown in the diagram indicates that report 8 requires 1 CPU, but the actual usage value is 0). Since report 8 is not using CPU, it will not be updated.

[0567] Therefore, in this example, report 3 does not utilize the APU and is therefore not updated. Since the APU requirement for report 3 is not met, the lower-priority CSI report of type #1 is not updated, meaning report 5 is not updated. In this example, report 8 does not utilize the CPU and is therefore not updated.

[0568] It's important to note that CSI report updates are determined independently based on APU and CPU resources. If a CSI report's CPU resources are insufficient, lower-priority CSI reports will not be updated. Similarly, if a CSI report of type #1 has insufficient APU resources, lower-priority CSI reports of type #1 will not be updated.

[0569] In one specific implementation, CPU and APU are occupied in order of priority on the current symbol. When the APU (if any) of a CSI report of type #1 is insufficient, the CSI report is not updated and the report does not occupy CPU and APU. If the CPU resources are not yet fully occupied, CSI reports with lower priority than the CSI report continue to occupy CPU.

[0570] In one specific implementation, on the current symbol, CPU and APU are occupied according to the priority of all CSI reports from high to low. When the APU demand of a certain type #1 CSI report is insufficient, the CPU and APU corresponding to the CSI report are not occupied, and CSI reports with lower priority than the CSI report (including CSI reports of type #1 and reports of type #2) continue to occupy CPU and APU.

[0571] Specifically, regarding the example report above, if the APU requirement for report 3 is not met, report 3 will not be updated and will not consume CPU and APU resources. That is, the CPU and APU usage of report 3 is O_cpu3 = 0 and O_apu3 = 0. Subsequent reports 4, 5, 6, 7, and 8 will be updated. The specific process is similar to the implementation described above, and for the sake of brevity, it will not be elaborated further here.

[0572] In one specific implementation, on the current symbol, CPU and APU are occupied according to the priority of all CSI reports. When the APU demand of a certain type #1 CSI report is insufficient, the CPU and APU corresponding to the CSI report are not occupied. CSI reports of type #1 with lower priority than the CSI report are not updated, and CSI reports of type #2 with lower priority than the CSI report continue to occupy CPU.

[0573] Specifically, regarding the example report above, since the APU requirement for report 3 is not met, report 3 is not updated and does not consume CPU or APU resources. That is, the CPU and APU usage of report 3 is O_cpu3 = 0 and O_apu3 = 0. Because report 3 is not updated, the lower-priority AI report (type #1) is also not updated; that is, report 5 is not updated and does not consume CPU resources, i.e., O_cpu5 = 0 and O_apu5 = 0. Subsequent reports of type #2 continue to be updated, i.e., reports 4, 6, 7, and 8 are updated. The specific process is similar to the implementation described above, and for the sake of brevity, it will not be elaborated further here.

[0574] In the current implementation, the following code may be included, but is not limited to:

[0575] In summary, in the implementation of this application, if a CSI report lacks CPU resources, even if a lower-priority CSI report has sufficient CPU resources, the lower-priority CSI report will not be updated. Insufficient APU resources for a higher-priority CSI report do not affect the updating of lower-priority CSI reports, further improving resource utilization.

[0576] It should be understood that the sequence number of each process does not imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of this application.

[0577] It should also be understood that, in the various embodiments of this application, unless otherwise specified or in case of logical conflict, the terms and / or descriptions between different embodiments are consistent and can be referenced by each other, and the technical features in different embodiments can be combined to form new embodiments according to their inherent logical relationships.

[0578] It should also be understood that in the above embodiments, the examples are mainly based on devices in existing network architectures, and it should be understood that the specific form of the device is not limited in the embodiments of this application. For example, any device that can achieve the same function in the future is applicable to the embodiments of this application.

[0579] It is understood that, in the above-described method embodiments, the methods and operations implemented by a device (such as a terminal device or a network device) can also be implemented by components of the device (such as chips or circuits).

[0580] Those skilled in the art will recognize that, based on the units and algorithm steps described in conjunction with the embodiments disclosed herein, this application can be implemented in hardware or a combination of hardware and computer software. Whether a function is implemented in hardware or by computer software driving hardware depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0581] The methods provided in the embodiments of this application have been described in detail above with reference to several accompanying drawings. The apparatus provided in the embodiments of this application will now be described with reference to the accompanying drawings.

[0582] Figures 20 and 21 are schematic block diagrams of possible apparatuses provided in embodiments of this application. These apparatuses can be used to implement the terminal-side or network-side functions in the above method embodiments, and thus can also achieve the beneficial effects of the above method embodiments.

[0583] Figure 20 is a schematic block diagram of the device provided in an embodiment of this application.

[0584] The device 1900 shown in Figure 20 may include a processing module 1910 and a communication module 1920.

[0585] In one possible design, device 1900 can be used to implement the communication method implemented by the terminal device in any of the embodiments shown in Figures 10 to 19. For example, processing module 1910 is used to implement processing-related steps performed by the terminal device in each method embodiment, such as determining the count of the first resource, determining the first activation interval corresponding to the first resource, determining the start time of the first activation interval, determining the end time of the first activation interval, determining the first time offset, etc.; communication module 1920 is used to implement sending and / or receiving steps performed by the terminal device in each method embodiment, such as sending first indication information, etc.

[0586] For example, the processing module 1920 is configured to: determine the count of the first resource and / or the first activation interval corresponding to the first resource.

[0587] For example, the processing module 1920 is used to: determine whether the first CSI report and / or the second CSI report has been updated.

[0588] For example, the processing module 1920 is used to: determine the count value of the third processing resource corresponding to the second monitoring report.

[0589] A more detailed description of the processing module 1910 and the communication module 1920 can be obtained directly from the relevant descriptions in the embodiments shown in Figures 10 to 19, and will not be repeated here.

[0590] In another possible design, device 1900 can be used to implement the communication method implemented by a network device in any of the embodiments shown in Figures 10 to 19. For example, processing module 1910 is used to implement the processing-related steps performed by the network device in each method embodiment, and communication module 1920 is used to implement the sending and / or receiving steps performed by the network device in each method embodiment, such as receiving first indication information.

[0591] A more detailed description of the processing module 1910 and the communication module 1920 can be obtained directly from the relevant descriptions in the embodiments shown in Figures 10 to 19, and will not be repeated here.

[0592] It should be noted that the communication module can also be called a transceiver module, transceiver unit, transceiver, transceiver device, or transceiver apparatus, etc. The processing module can also be called a processor, processing board, processing unit, or processing apparatus, etc. Optionally, the communication module is used to perform the sending and receiving operations of the terminal device or network device in the above method. The device in the communication module that implements the receiving function can be considered as the receiving module, and the device in the communication module that implements the sending function can be considered as the sending module; that is, the communication module can include both a receiving module and a sending module.

[0593] It should also be noted that, in one possible design, the aforementioned processing module and / or communication module can be implemented through virtual modules. For example, the processing module can be implemented through software functional units or virtual devices, and the communication module can be implemented through software functions or virtual devices. In another possible design, the processing module or communication module can also be implemented through physical devices. For example, if the device is implemented using a chip / chip circuit, the communication module can be an input / output circuit and / or a communication interface, performing input operations (corresponding to the aforementioned receiving operation) and output operations (corresponding to the aforementioned sending operation); the processing module can be an integrated processor, a microprocessor, or an integrated circuit.

[0594] The module division in this embodiment is illustrative and represents only one logical functional division; in actual implementation, other division methods may be used. Furthermore, the functional modules in the various examples of this embodiment can be integrated into a single processor, exist as separate physical entities, or be integrated into a single module. The integrated modules described above can be implemented in hardware or as software functional modules.

[0595] Figure 21 is a schematic diagram of the structure of a communication device provided in another embodiment of this application.

[0596] The device 2000 includes a processing circuit 2010 and a communication circuit 2020. The processing circuit 2010 and the communication circuit 2020 are coupled to each other.

[0597] It can be understood that the processing circuit 2010 can be one or more processors, or it can be all or part of the circuits with processing functions in one or more processors.

[0598] It is understandable that the communication circuit 2020 can be a transceiver or an input / output interface.

[0599] Optionally, the device 2000 may further include a memory 2030 for storing instructions executed by the processing circuit 2010, or storing input data required for the running instructions of the processing circuit 2010, or storing data generated after the running instructions of the processing circuit 2010.

[0600] It is understood that the memory 2030 may be located outside the processing circuit 2010 or inside the processing circuit 2010.

[0601] As an example, the processing circuit 2010 is used to implement the functions of the processing module 1910, and the communication circuit 2020 is used to implement the functions of the communication module 1920.

[0602] As an example, device 2000 can be a communication device (such as a terminal device or network device) or a chip used in a communication device.

[0603] When device 2000 is a communication device, the communication circuit can be a transceiver; when device 2000 is a chip, the communication circuit can be an input / output circuit, a bus, pins, or other types of communication interfaces. The input circuit in the input / output circuit can be used for receiving, and the output interface can be used for transmitting.

[0604] In one possible implementation, the apparatus 2000 is used to implement the various processes and steps corresponding to the terminal device in the above method embodiments. In another possible implementation, the apparatus 2000 is used to implement the various processes and steps corresponding to the network device in the above method embodiments.

[0605] It is understood that the device 2000 can specifically be the terminal device or network device in the above embodiments, or it can be a chip or chip system. Correspondingly, the communication circuit can be the interface circuit of the chip, or an input / output circuit, which is not limited here. Specifically, the device 2000 can be used to execute the various steps and / or processes corresponding to the terminal device or network device in the above method embodiments.

[0606] When the aforementioned communication device is a chip or OTT device applied to a terminal device, the chip or OTT device of the terminal device implements the functions of the terminal device in the above method embodiments, for example, implementing the processing functions of the terminal device. The terminal device chip or OTT device receiving information from the network device can be understood as the information being first received by other modules (such as radio frequency modules or antennas) in the terminal device, and then sent by these modules to the terminal device chip or OTT device. The terminal device chip or OTT device sending information to the network device can be understood as the information being first sent by the terminal device chip or OTT device to other modules (such as radio frequency modules or antennas) in the network device, and then sent by these modules to the network device.

[0607] When the aforementioned communication device is a chip or OTT device applied to a network device, the chip or OTT device of the network device implements the functions of the network device in the above method embodiments, for example, implementing the processing functions of the network device. The chip or OTT device of the network device receiving information from the terminal device can be understood as the information being first received by other modules (such as radio frequency modules or antennas) in the network device, and then sent by these modules to the chip or OTT device. The chip or OTT device of the network device sending information to the terminal device can be understood as the information being first sent by the chip or OTT device to other modules (such as radio frequency modules or antennas) in the network device, and then sent by these modules to the terminal device.

[0608] Optionally, as shown in Figure 22, the device 2000 may also include a reasoning result storage and determination module.

[0609] Figure 22 is a schematic diagram of the reasoning result storage and determination module provided in an embodiment of this application.

[0610] The inference result storage and determination module is used to determine the first resource count and the first activation interval based on the first inference report configuration and the first monitoring report configuration.

[0611] Alternatively, the inference result storage and determination module can be understood as being used to determine the active resource (first resource) count and resource activation interval (first activation interval) based on the inference report configuration and monitoring report configuration signaling, thereby controlling the transmit / receive chain through communication and processing circuits. The functionality of the inference result storage and determination module can also be processed on a computer-readable medium.

[0612] This application also provides a computer program product that, when run on a processor, can implement the communication method executed by the terminal device or the communication method executed by the network device in the above method embodiments.

[0613] This application also provides a computer-readable storage medium containing computer instructions that, when executed on a processor, can implement the communication method executed by a terminal device or a network device in the above method embodiments.

[0614] This application also provides a communication system, including the aforementioned terminal equipment and network equipment.

[0615] It is understood that the processor in the embodiments of this application may be any of the following devices or all or part of the circuitry used for processing functions: a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSPs), processors for AI, field-programmable gate arrays (FPGAs), or other programmable logic devices, transistor logic devices, hardware components, or any combination thereof. A general-purpose processor may be a microprocessor or any conventional processor.

[0616] For example, the processor used for AI can be one or more of the following: graphics processing unit (GPU), neural processing unit (NPU), tensor processing unit (TPU), and data processing unit (DPU).

[0617] For example, one possible implementation of a processor for AI could be the AI ​​processor 1500 shown in Figure 23.

[0618] Figure 23 is a schematic block diagram of the AI ​​processor provided in an embodiment of this application.

[0619] The AI ​​processor 2200 may include one or more of the following: AI core, digital vision pre-processing (DVPP) module, task scheduler (TS), L3 cache, AI CPU (central processing unit), control CPU (central processing unit), L2 cache, universal serial bus (USB) interface, network interface card, peripheral component interconnect express (PCIe) interface (PCIe is a high-speed serial computer expansion bus standard), double data rate (DDR) / high bandwidth memory (HBM) interface, generation purpose input / output (GPIO) / inter-integrated circuit (I2C) bus, etc. It is understood that the specific meanings of these terms are well known to those skilled in the art and will not be elaborated upon here.

[0620] The terms “unit”, “module”, etc., used in this specification may be used to refer to computer-related entities, hardware, firmware, combinations of hardware and software, software, or software in execution.

[0621] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0622] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0623] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0624] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0625] In addition, the functional units in the various embodiments of this application can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit.

[0626] In the above embodiments, the functions of each functional unit can be implemented entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. This computer program product includes one or more computer instructions (programs). When the computer program instructions (programs) are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another via wired (e.g., coaxial cable, fiber optic, digital subscriber line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium accessible to a computer or a data storage device such as a server or data center that integrates one or more available media. The available media can be magnetic media (e.g., floppy disks, hard disks, magnetic tapes), optical media (e.g., digital video discs (DVDs)), or semiconductor media (e.g., solid-state disks (SSDs)).

[0627] If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0628] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

Claims

A communication method, characterized in that, A chip used in a first device or used in the first device, comprising: Determine the count of the first resource and / or the first activation interval corresponding to the first resource; wherein, The first resource corresponds to the first inference report, or the first resource corresponds to the configuration of the first inference report, or the first resource corresponds to a inference result in the first inference report, or the first resource corresponds to the first monitoring report associated with the first inference report, or the first resource corresponds to the configuration of the first monitoring report associated with the first inference report; The first resource is counted A times, where A is a natural number greater than or equal to 1. The method according to claim 1, characterized in that, The first inference report corresponds to one or more monitoring reports. The method according to claim 1 or 2, characterized in that, The first resource is a channel state information reference signal resource or a corresponding channel state information reference signal resource, and the first activation interval corresponds to the activation interval of the channel state information reference signal resource; or The first resource is a storage unit or a corresponding storage unit, and the first activation interval corresponds to the activation interval of the storage unit. The method according to any one of claims 1 to 3 is characterized in that, The start time of the first activation interval includes the first start time. The first start time includes the following: the time unit corresponding to the first Radio Resource Control (RRC) configuration, the reporting time of the first inference report, the time unit where the Channel State Information (CSI) reference resource of the first inference report is located, the trigger time of the first inference report, and the trigger time of the first monitoring report. The first RRC configuration is used to configure the first inference report, or to configure the first inference report to be associated with one or more monitoring reports. The method according to any one of claims 1 to 4, characterized in that, The end time of the first activation interval is the first end time. The first end time includes the following: the time of releasing the first RRC configuration, the end time of the effect of the inference result corresponding to the first inference report, the end time of the prediction window corresponding to the first inference report, the end time of the effect of one or more prediction results corresponding to the first inference report, the time of completion of the performance index calculation of the first monitoring report, the reporting time of the first monitoring report, and the time unit where the CSI reference resource of the first monitoring report is located. The first RRC configuration is used to configure the first inference report, or to configure the first inference report to be associated with one or more monitoring reports. The method according to claim 4, characterized in that, The end time of the first activation interval is the first time offset after the first start time. The first time offset is indicated by the first device, or the first time offset includes a predefined time offset. The method according to any one of claims 1 to 4, characterized in that, The end time of the first activation interval is the first end time, and there is a first time offset between the first end time and the first time; wherein, The first time offset is indicated by the first device, or the first time offset includes a predefined time offset; The first moment includes the following: the time of releasing the first RRC configuration, the end time of the effect of the inference result corresponding to the first inference report, the end time of the prediction window corresponding to the first inference report, the end time of the effect of one or more prediction results corresponding to the first inference report, the time of completion of the performance index calculation of the first monitoring report, the reporting time of the first monitoring report, and the time unit where the CSI reference resource of the first monitoring report is located. The first RRC configuration is used to configure the first inference report, or to configure the first inference report to be associated with one or more monitoring reports. The method according to any one of claims 1 to 4, characterized in that, The first monitoring report is submitted later than the first moment, and the end time of the first activation interval is the submission time of the first monitoring report; The first moment includes the following: the time of releasing the first RRC configuration, the end time of the effect of the inference result corresponding to the first inference report, the end time of the prediction window corresponding to the first inference report, the end time of the effect of one or more prediction results corresponding to the first inference report, the time of completion of the calculation of the monitoring performance index of the first monitoring report, and the time unit in which the CSI reference resource of the first monitoring report is located. The method according to any one of claims 1 to 8, characterized in that, The method further includes: Send a first indication message, which is used to indicate A. The method according to any one of claims 1 to 9, characterized in that, The first activation intervals of the first resource corresponding to at least two inference reports or at least two inference results correspond to the same time unit within the first interval, and the first resource within the first interval is not counted repeatedly, or is counted as 1. The method according to any one of claims 1 to 9, characterized in that, The first activation intervals of the first resource corresponding to at least two inference reports or at least two inference results correspond to the same time unit within the first interval. The first resource is counted repeatedly within the first interval. The count of the first resource is related to the number of times the inference reports or inference results overlap within the first interval. Wherein, the at least two inference reports correspond to the same inference report configuration, or the inference reports corresponding to the at least two inference results correspond to the same inference report configuration. The method according to any one of claims 1 to 11, characterized in that, Determining the count of the first resource includes determining the count of the first resource within the first time unit. The method according to any one of claims 1 to 12, characterized in that, Determining the count of the first resource includes, The timing of the first downlink control information is earlier than the reporting time of the first inference report, and the first resource is counted A times within the first activation interval; and / or, The time of the first downlink control information is later than the reporting time of the first inference report. The first resource is counted 0 times in the first activation interval, or is not counted, or the first resource is not activated or is in a deactivated state in the first activation interval. The first downlink control information is used to trigger the first monitoring report. The method according to any one of claims 1 to 13, characterized in that, The first resource is a first inference resource and / or a first monitoring resource; Wherein, the first inference resource is the measurement resource corresponding to the first inference report, and the first monitoring resource is the measurement resource corresponding to the first monitoring report. The method according to claim 14, characterized in that, The first resource is the first inference resource, and the first resource is additionally counted Kp times; The Kp is related to the configuration of the first inference report, and / or the Kp is indicated by the first device. The method according to any one of claims 1 to 15, characterized in that, The first inference report is associated with one or more monitoring reports, and the first resource is not counted repeatedly. The method according to any one of claims 1 to 16, characterized in that, The time-domain type of the first inference report is one or more of periodic, semi-periodic, or non-periodic. The method according to any one of claims 3 to 15, characterized in that, The first indication information is also used to indicate one or more of the following parameters: the maximum number of supported storage units, the maximum supported storage time; wherein, the storage unit corresponds to the storage of inference reports, or the storage unit corresponds to the storage of inference results. A communication method, characterized in that, A chip used in a second device or used in the second device, comprising: The first processing resource is not used for the first Channel State Information (CSI) report, and the second processing resource is used for the second CSI report; wherein, the processing of the first CSI report corresponds to the use of the first processing resource, the processing of the second CSI report corresponds to the use of the second processing resource, and the first processing resource is not used for the processing of the second CSI report. The method according to claim 19, characterized in that, The processing of the first CSI report also corresponds to the occupancy of the second processing resources. The method according to claim 19 or 20 is characterized in that, At least one second CSI report has a lower priority than at least one first CSI report. The method according to any one of claims 19 to 21 is characterized in that, The first processing resources include resources corresponding to the channel state information processing unit or resources corresponding to the channel state information reporting processing unit based on artificial intelligence / machine learning; The second processing resources include resources corresponding to the channel state information reporting processing unit based on artificial intelligence / machine learning, or resources corresponding to the channel state information processing unit. A communication method, characterized in that, A chip used in a third device or used in said third device, comprising: Based on the second monitoring report, determine the count value of the third processing resource corresponding to the second monitoring report, wherein the second monitoring report is associated with one or more inference reports. The method according to claim 23, characterized in that, The count value of the third processing resource corresponding to the second monitoring report is related to the count value of the third processing resource corresponding to the first measurement resource and / or the first monitoring factor, wherein the first measurement resource is used for channel measurement corresponding to the second monitoring report. The method according to claim 23 or 24 is characterized in that, The count value of the third processing resource corresponding to the second monitoring report is also related to the first count value; The first count value is indicated by the first device, or the first count value includes a predefined count value, or the first count value is associated with at least one of the following: inference result inverse calculation, monitoring inverse calculation, or squared generalized cosine similarity (SGCS) calculation. The method according to any one of claims 23 to 25, characterized in that, The first monitoring factor is indicated by the first device, or the first monitoring factor includes predefined monitoring factors, or the first monitoring factor has a corresponding relationship with the first count value. A communication device, characterized in that, It includes functional modules for implementing the method as described in any one of claims 1 to 26. A communication device, characterized in that, It includes one or more processors and communication circuitry, the communication circuitry being used by the communication device to perform at least one of signal input or output; the one or more processors being used to implement the method as described in any one of claims 1 to 26. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program or instructions that, when executed by a processor, cause the method as described in any one of claims 1 to 26 to be performed. A computer program product, characterized in that, Includes a computer program that, when run, causes the method as described in any one of claims 1 to 26 to be performed.