Method and apparatus for wireless communication

Abstract

The present application discloses a method and apparatus for functionality reporting in a node for wireless communication. A first node receives first signaling and sends a first report. The first signaling comprises a first inference configuration. The first report indicates that L1 feature sets are supported, L1 being a positive integer. The L1 feature sets depend on an indication of the first signaling. Any feature set among the L1 feature sets comprises at least one of the maximum number of layers, the maximum number of downlink RS resources, the maximum number of uplink RS resources, a supported bandwidth, the maximum data rate, and the maximum modulation order. A feature corresponding to at least one index comprised in the first inference configuration does not belong to a first feature set, the first feature set being one feature set among the L1 feature sets. The method enhances a capability reporting mechanism, optimizes resource allocation, and improves resource utilization and system efficiency.

Description

A method and apparatus for wireless communication Technical Field

[0001] This application relates to transmission methods and apparatus in wireless communication systems, and more particularly to capability reporting or AI (Artificial Intelligence) / ML (Machine Learning) schemes and apparatus in wireless communication systems. Background Technology

[0002] In traditional wireless communication, the UE (User Equipment) obtains various auxiliary information by measuring downlink signals and / or channels, such as CSI (Channel State Information), beam management-related auxiliary information, and positioning-related auxiliary information. The CSI includes, but is not limited to, one or more of CRI (Channel State Information-Reference Signal Resource Indicator), RI (Rank Indicator), PMI (Precoding Matrix Indicator), or CQI (Channel Quality Indicator). The UE can use this information to select appropriate transmission parameters or report this information. The network device selects appropriate transmission parameters for the UE based on the UE's reports, such as the cell location, MCS (Modulation and Coding Scheme), TPMI (Transmitted Precoding Matrix Indicator), and TCI (Transmission Configuration Indicator) status. Furthermore, the UE's reports can be used to optimize network parameters, such as improving cell coverage and switching base stations based on the UE's location. With the adoption of new technologies, the increase in the number of antennas, the diversification of application scenarios, and the increasing demands on system performance, traditional measurement and reporting methods incur significant redundancy overhead. In NR R (release) 19, research on AI (Artificial Intelligence) / ML (Machine Learning) technologies was initiated, including AI-based channel information reporting, such as beam prediction and CSI prediction. Beyond CSI measurement, calculation, and reporting, AI / ML holds great potential in many other aspects of communication systems, including but not limited to data reception, channel estimation, modulation and demodulation, encoding and decoding, multi-antenna demodulation, and scheduling. In the future technological evolution of 5G and 6G, AI / ML-based technologies and their applications in communication systems will become research hotspots.

[0003] At the 117bis meeting of RAN (Radio Access Network) WG (Working Group) 1, Supported Functionality and Applicable Functionality were approved. Supported Functionality is indicated in UE capability reporting, while Applicable Functionality is determined by the UE based on network-side additional conditions and UE-side additional conditions, and is indicated to the network. Applicable Functionality usually means that the corresponding AI model is available, or the corresponding inference configuration can be activated / executed.

[0004] Since the specifications of AI models may extend beyond the scope of 3GPP (besides the reference model used for performance calibration), the specific implementation of AI / ML training and AI / ML inference may be determined by the hardware equipment vendors themselves. It may be based on classic models such as Transformer architecture, RNN (Recurrent Neural Network), CNN (Conventional Neural Network), or a hybrid model composed of multiple models. Summary of the Invention

[0005] The applicant's research revealed that AI / ML operations or inference require resources, such as computing resources, storage resources, and power. In the future evolution of 5G or 6G, AI / ML can be used for, but is not limited to, CSI measurement / calculation / reporting, encoding / decoding, demodulation, and channel estimation. AI / ML used for different purposes consumes resources, thus UE support for different metrics will be mutually restrictive. That is, support for some metrics will affect support for others. In this case, the existing UE capability reporting mechanism needs to be enhanced. To address the above problems, this application discloses a solution. It should be noted that although many embodiments of this application are focused on AI / ML, this application is also applicable to other solutions, such as traditional (non-AI / ML based) capability reporting solutions. Furthermore, adopting a unified solution for different scenarios (including but not limited to AI / ML-based solutions and traditional solutions) helps reduce hardware complexity and cost. Where there is no conflict, the embodiments and features in the embodiments of the first node of this application can be applied to the second node, and vice versa. Where there is no conflict, the embodiments and features in the embodiments of this application can be arbitrarily combined with each other.

[0006] The applicant's research revealed that the introduction of AI / ML functionality significantly impacts data transmission. For example, traditional receiving modules are often embedded in hardware, and their maximum data rate is independent of performance and data quality / type / business requirements. However, for AI / ML models represented by the Transformer architecture, inference performance increases with the number of parameters; a larger number of parameters leads to greater computational demands, thus limiting the data rate that inference can handle. Furthermore, AI / ML model inference generally consumes significant resources, leading to the adoption of techniques to reduce inference complexity and increase speed, such as MoE (Mixed Expert Model) and sparse activation. With these techniques, the resource consumption for inference becomes dependent on the input data; that is, different data requires different amounts of resources, thus affecting the maximum supported data rate. To address these issues, this application discloses a solution. It should be noted that while this application is motivated by the application of AI / ML models and many embodiments are specifically for AI / ML, it is also applicable to other solutions, such as traditional data transmission algorithms / solutions. While this application's specification includes descriptions of AI / ML models and algorithms, those skilled in the art will understand that these descriptions are not essential or irreplaceable for solutions related to wireless cellular communication. Furthermore, employing a unified solution across different scenarios (including but not limited to AI / ML-based solutions and traditional data transmission algorithms / solutions) helps reduce signaling overhead / complexity, hardware complexity, and cost. Where there is no conflict, the embodiments and features in the first node of this application can be applied to the second node, and vice versa. Where there is no conflict, the embodiments and features in the embodiments of this application can be arbitrarily combined with each other.

[0007] Furthermore, although some embodiments of this application are described based on the 5G protocol, this application is also applicable to communication standards after 5G without conflict.

[0008] When necessary, the interpretation of terms used in this application shall be based on the definitions in the 3GPP specification protocol TS38 series, or the definitions in the 3GPP specification protocol TS28 series.

[0009] This application discloses a method used in a first node for wireless communication, characterized by comprising:

[0010] Receive a first signaling, the first signaling including a first inference configuration;

[0011] Send a first report indicating support for L1 feature sets, where L1 is a positive integer;

[0012] Wherein, the L1 feature sets depend on the indication of the first signaling; any feature set in the L1 feature sets includes at least one of the following: maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, maximum data rate, and maximum modulation order; the feature corresponding to at least one indicator included in the first inference configuration does not belong to the first feature set, and the first feature set is a feature set in the L1 feature sets.

[0013] As an example, the problem this application aims to solve includes how to take into account the mutual constraints between different indicators in feature set reporting; in the above method, the L1 feature sets depend on the indication of the first signaling, and the feature corresponding to at least one indicator included in the first inference configuration does not belong to the first feature set, thus solving this problem.

[0014] As an example, the advantages of the above method include that it fully considers the special needs of AI / ML technology, enhances the reporting mechanism, and facilitates network-side optimization of resource allocation.

[0015] As an example, the advantages of the above method include improved resource utilization and system efficiency.

[0016] As an example, the advantages of the above method include fully leveraging the strengths of AI / ML and improving the overall system performance.

[0017] As an example, the advantages of the above method include its simplicity of implementation and minimal changes to the standard.

[0018] As an example, the advantages of the above method include good backward compatibility.

[0019] According to one aspect of this application, it is characterized by comprising:

[0020] Send a second report indicating support for L0 feature sets, where L0 is a positive integer greater than 1;

[0021] The second report is sent before the first report, and the L1 feature sets depend on the L0 feature sets.

[0022] The advantages of the above method include enabling the network side to understand the basic capabilities of the first node, which facilitates resource allocation on the network side.

[0023] The advantages of the above method include allowing the first node to flexibly adjust its reporting based on actual conditions, thereby further optimizing resource allocation.

[0024] The advantages of the above method include good backward compatibility.

[0025] According to one aspect of this application, it is characterized by comprising:

[0026] Receive a second signaling message, the second signaling message indicating the first configuration information;

[0027] The L0 feature sets depend on the indication of the second signaling.

[0028] The benefits of the above method include helping the network side to have a more comprehensive understanding of the capabilities of the first node, and further optimizing resource allocation.

[0029] The advantages of the above method include avoiding discrepancies between the network's understanding of the L0 feature sets and the first node.

[0030] The benefits of the above methods include further enhancing the performance advantages that AI / ML brings to the system.

[0031] According to one aspect of this application, the L0 feature sets are characterized by being conditional on a first hypothesis; the first hypothesis includes the association between the inference configuration and the training dataset.

[0032] The advantages of the above method include avoiding discrepancies between the network side's understanding of the L0 feature sets and the first node, further optimizing resource allocation, and further improving the performance advantages brought by AI / ML to the system.

[0033] According to one aspect of this application, the L0 feature sets are conditional upon a second assumption; the second assumption includes not employing reasoning.

[0034] The advantages of the above method include facilitating consensus between the network side and the first node on the understanding of the L0 feature sets, further optimizing resource allocation, and improving the robustness of the system.

[0035] The advantages of the above method include reduced air interface overhead.

[0036] The advantages of the above method include good backward compatibility.

[0037] According to one aspect of this application, the first report is triggered by an event in a first event set, the first event set including at least one state change of inference configuration, the state change including one or more of being activated, deactivated, configured, and reconfigured.

[0038] The advantages of the above method include avoiding unnecessary reporting of the first report, and saving air interface resources without affecting the real-time performance and accuracy of the reporting.

[0039] The advantages of the above method include a better balance between real-time reporting and overhead.

[0040] According to one aspect of this application, it is characterized by comprising:

[0041] Receive third signaling;

[0042] The third signaling triggers the first report.

[0043] The advantages of the above methods include facilitating unified scheduling and global optimization on the network side.

[0044] This application discloses a method used in a second node for wireless communication, characterized by comprising:

[0045] Send a first signaling message, the first signaling message including a first inference configuration;

[0046] Receive a first report indicating support for L1 feature sets, where L1 is a positive integer;

[0047] Wherein, the L1 feature sets depend on the indication of the first signaling; any feature set in the L1 feature sets includes at least one of the following: maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, maximum data rate, and maximum modulation order; the feature corresponding to at least one indicator included in the first inference configuration does not belong to the first feature set, and the first feature set is a feature set in the L1 feature sets.

[0048] According to one aspect of this application, it is characterized by comprising:

[0049] Receive a second report indicating support for L0 feature sets, where L0 is a positive integer greater than 1;

[0050] The second report is received before the first report, and the L1 feature sets depend on the L0 feature sets.

[0051] According to one aspect of this application, it is characterized by comprising:

[0052] Send a second signaling message, the second signaling message indicating the first configuration information;

[0053] The L0 feature sets depend on the indication of the second signaling.

[0054] According to one aspect of this application, the L0 feature sets are characterized by being conditional on a first hypothesis; the first hypothesis includes the association between the inference configuration and the training dataset.

[0055] According to one aspect of this application, the L0 feature sets are conditional upon a second assumption; the second assumption includes not employing reasoning.

[0056] According to one aspect of this application, the first report is triggered by an event in a first event set, the first event set including at least one state change of inference configuration, the state change including one or more of being activated, deactivated, configured, and reconfigured.

[0057] According to one aspect of this application, it is characterized by comprising:

[0058] Send third signaling;

[0059] The third signaling triggers the first report.

[0060] This application discloses a first node used for wireless communication, characterized in that it includes:

[0061] A first receiver receives a first signaling, the first signaling including a first inference configuration;

[0062] A first transmitter sends a first report, the first report indicating support for L1 feature sets, where L1 is a positive integer;

[0063] Wherein, the L1 feature sets depend on the indication of the first signaling; any feature set in the L1 feature sets includes at least one of the following: maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, maximum data rate, and maximum modulation order; the feature corresponding to at least one indicator included in the first inference configuration does not belong to the first feature set, and the first feature set is a feature set in the L1 feature sets.

[0064] This application discloses a second node used for wireless communication, characterized by comprising:

[0065] The second transmitter transmits a first signaling message, the first signaling message including a first inference configuration;

[0066] The second receiver receives a first report, which indicates support for L1 feature sets, where L1 is a positive integer.

[0067] Wherein, the L1 feature sets depend on the indication of the first signaling; any feature set in the L1 feature sets includes at least one of the following: maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, maximum data rate, and maximum modulation order; the feature corresponding to at least one indicator included in the first inference configuration does not belong to the first feature set, and the first feature set is a feature set in the L1 feature sets.

[0068] As an example, compared with conventional solutions, this application has the following advantages:

[0069] Taking into full account the special needs of AI / ML technologies, the capability reporting mechanism has been enhanced to facilitate network-side optimization of resource allocation;

[0070] By fully leveraging the advantages of AI / ML, the overall system performance has been improved;

[0071] It improved resource utilization and system efficiency;

[0072] Allowing UEs to flexibly adjust their capability reporting based on actual conditions helps the network side to adjust resource allocation more flexibly and dynamically.

[0073] This application discloses a method used in a first node for wireless communication, characterized by comprising:

[0074] Send a first report indicating that the idle data rate capability is a first data rate, the idle data rate capability being configured for a first inference.

[0075] As an example, the problem this application aims to solve includes how to enhance the reporting of data rate capabilities; in the above method, the first report indicates the idle data rate capability configured for the first inference, thus solving this problem.

[0076] As an example, the advantages of the above method include optimized resource utilization and improved overall system performance.

[0077] As an example, the advantages of the above method include that it helps the network side to understand the data processing capabilities of the first node more accurately and in a timely manner, thereby better optimizing scheduling and improving resource utilization and transmission efficiency.

[0078] As an example, the advantages of the above method include better meeting the needs of AI / ML functions and maximizing the improvement of system performance by AI / ML.

[0079] As an example, the advantages of the above method include improved QoS.

[0080] As an example, the advantages of the above method include more flexible design and applicability to different scenarios and terminals.

[0081] According to one aspect of this application, the first data rate is for a first inference, the first inference depending on the first inference configuration.

[0082] The essence of the above method includes reporting the idle data rate capability for a specific inference instance configured for the first inference.

[0083] The advantages of the above method include more flexible and faster updates to the network side's data processing capabilities for the first node, and optimized resource utilization and scheduling.

[0084] The advantages of the above method include ensuring that the first node and the first reporting target recipient have a consensus on the first data rate.

[0085] According to one aspect of this application, the first data rate is for a first channel, and the processing of the first channel depends on the first inference configuration.

[0086] The essence of the above method includes reporting the idle data rate capability for a specific physical layer channel used for processing by the first inference configuration.

[0087] The advantages of the above method include more flexible and faster updates to the network side's data processing capabilities for the first node, and optimized resource utilization and scheduling.

[0088] The advantages of the above method include ensuring that the first node and the first reporting target recipient have a consensus on the first data rate.

[0089] According to one aspect of this application, the first data rate is relative to a first reference time, which is associated with the first inference configuration.

[0090] The essence of the above method includes reporting the idle data rate capability for a specific time associated with the first inference configuration.

[0091] The advantages of the above method include more flexible and faster updates to the network side's data processing capabilities for the first node, and optimized resource utilization and scheduling.

[0092] The advantages of the above method include ensuring that the first node and the first reporting target recipient have a consensus on the first data rate.

[0093] According to one aspect of this application, it is characterized by comprising:

[0094] Receive a first message, which indicates multiple inference configurations;

[0095] Send a second message indicating that a reasoning configuration in a first subset of the plurality of reasoning configurations is available;

[0096] The first inference configuration subset includes the first inference configuration.

[0097] The advantages of the above method include allowing the first node to determine the available inference configuration based on its own additional conditions, providing a foundation for the application of AI / ML in communication systems.

[0098] According to one aspect of this application, it is characterized by comprising:

[0099] Receive a second signaling message, which activates the first inference configuration.

[0100] The advantages of the above methods include more flexible signaling design, making them applicable to different scenarios.

[0101] The advantages of the above methods include supporting joint optimization on the network side, which further improves the overall system performance.

[0102] According to one aspect of this application, it is characterized by comprising:

[0103] Send a second data rate, wherein the second data rate is configured for the first inference.

[0104] The advantages of the above method include that it allows the network side to fully understand the capabilities of the first node, optimizes network-side scheduling and resource allocation, and improves resource utilization.

[0105] Considering the potential mutual constraints between the parameters of AI / ML models and the inference output, the above method enables the network side to more accurately understand the capabilities connected to the first inference configuration, thereby promoting further optimization of the scheduling strategy by the network side.

[0106] According to one aspect of this application, it is characterized by comprising:

[0107] Receive a second data rate, wherein the second data rate is configured for the first inference.

[0108] The above method supports the network side in transmitting QoS-related information of the service corresponding to the first inference configuration.

[0109] The advantages of the above method include optimizing the first node's processing of different data.

[0110] The benefits of the above methods include improved QoS (Quality of Service).

[0111] According to one aspect of this application, it is characterized by comprising:

[0112] Send a second report indicating L0 Class 1 resources, where L0 is a positive integer greater than 1, and the L0 Class 1 resources are used for inference;

[0113] The first data rate depends on the amount of the first type of resources occupied by the first inference configuration.

[0114] The benefits of the above method include enabling the network side to better understand the capabilities of the first node, optimizing network-side scheduling and resource allocation, and improving transmission performance and resource utilization.

[0115] According to one aspect of this application, the first report is triggered by an event in a first event set, the first event set including at least one of the following:

[0116] The change in the number of first-type resources occupied by the first inference configuration is greater than the first threshold.

[0117] The maximum data rate variation supported by the first inference configuration is greater than the second threshold.

[0118] The change in the number of first-class resources occupied per unit data rate processed by the first inference configuration is greater than the third threshold.

[0119] The availability of at least one inference configuration has changed;

[0120] At least one inference configuration is activated or deactivated;

[0121] The first timer expired.

[0122] As an example, the advantages of the above method include good backward compatibility.

[0123] As an example, the advantages of the above method include greater flexibility to be applied to different terminals and application scenarios.

[0124] As an example, the advantages of the above method include reduced reporting overhead and latency.

[0125] According to one aspect of this application, it is characterized by comprising:

[0126] Receive the first signaling;

[0127] The first signaling triggers the first report.

[0128] The advantages of the above method include good backward compatibility.

[0129] The advantages of the above methods include facilitating joint optimization on the network side and further improving system performance.

[0130] This application discloses a method used in a second node for wireless communication, characterized by comprising:

[0131] Receive a first report indicating that the idle data rate capability is a first data rate, the idle data rate capability being configured for a first inference.

[0132] According to one aspect of this application, the first data rate is for a first inference, the first inference depending on the first inference configuration.

[0133] According to one aspect of this application, the first data rate is for a first channel, and the processing of the first channel depends on the first inference configuration.

[0134] According to one aspect of this application, the first data rate is relative to a first reference time, which is associated with the first inference configuration.

[0135] According to one aspect of this application, it is characterized by comprising:

[0136] Send a first message, which indicates multiple inference configurations;

[0137] Receive a second message indicating that a reasoning configuration in a first subset of the plurality of reasoning configurations is available;

[0138] The first inference configuration subset includes the first inference configuration.

[0139] According to one aspect of this application, it is characterized by comprising:

[0140] Send a second signaling message, which activates the first inference configuration.

[0141] According to one aspect of this application, it is characterized by comprising:

[0142] Receive a second data rate, which is configured for the first inference.

[0143] According to one aspect of this application, it is characterized by comprising:

[0144] Send a second data rate, the second data rate being configured for the first inference.

[0145] According to one aspect of this application, it is characterized by comprising:

[0146] Receive a second report indicating L0 Class 1 resources, where L0 is a positive integer greater than 1, and the L0 Class 1 resources are used for inference;

[0147] The first data rate depends on the amount of the first type of resources occupied by the first inference configuration.

[0148] According to one aspect of this application, the first report is triggered by an event in a first event set, the first event set including at least one of the following:

[0149] The change in the number of first-type resources occupied by the first inference configuration is greater than the first threshold.

[0150] The maximum data rate variation supported by the first inference configuration is greater than the second threshold.

[0151] The change in the number of first-class resources occupied per unit data rate processed by the first inference configuration is greater than the third threshold.

[0152] The availability of at least one inference configuration has changed;

[0153] At least one inference configuration is activated or deactivated;

[0154] The first timer expired.

[0155] According to one aspect of this application, it is characterized by comprising:

[0156] Send the first signaling;

[0157] The first signaling triggers the first report.

[0158] This application discloses a first node used for wireless communication, characterized in that it includes:

[0159] A first processor sends a first report indicating that the idle data rate capability is a first data rate, the idle data rate capability being configured for a first inference.

[0160] This application discloses a second node used for wireless communication, characterized by comprising:

[0161] A second processor receives a first report indicating that the idle data rate capability is a first data rate, the idle data rate capability being configured for a first inference.

[0162] As an example, compared with conventional solutions, this application has the following advantages:

[0163] Resource allocation was optimized, resource utilization and transmission efficiency were improved, and QoS of services was ensured.

[0164] The network-side scheduling was optimized, improving the overall system performance;

[0165] It maximizes the improvement of system performance by AI / ML;

[0166] It offers excellent flexibility and is suitable for different scenarios and terminals. Attached Figure Description

[0167] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0168] Figure 1 illustrates a flowchart of a first signaling and a first report according to an embodiment of this application;

[0169] Figure 2 shows a schematic diagram of a network architecture according to an embodiment of this application;

[0170] Figure 3 illustrates a schematic diagram of an embodiment of a wireless protocol architecture for the user plane and control plane according to an embodiment of this application;

[0171] Figure 4 shows a schematic diagram of a first communication device and a second communication device according to an embodiment of this application;

[0172] Figure 5 illustrates the transmission between a first node and a second node according to an embodiment of this application;

[0173] Figure 6 shows a schematic diagram of a second report according to an embodiment of this application;

[0174] Figure 7 illustrates a schematic diagram of L1 feature sets depending on L0 feature sets according to an embodiment of this application;

[0175] Figure 8 shows a schematic diagram of L1 feature sets and L0 feature sets according to an embodiment of this application;

[0176] Figure 9 shows a schematic diagram of L1 feature sets and L0 feature sets according to an embodiment of this application;

[0177] Figure 10 illustrates a schematic diagram of the second signaling, the first configuration information, and L0 feature sets according to an embodiment of this application;

[0178] Figure 11 shows a schematic diagram of L0 feature sets according to an embodiment of the present application, conditioned on a first assumption;

[0179] Figure 12 shows a schematic diagram of L0 feature sets according to an embodiment of the present application, conditioned on a second assumption;

[0180] Figure 13 illustrates a schematic diagram of a first report being triggered by an event in a first event set according to an embodiment of this application;

[0181] Figure 14 illustrates a schematic diagram of a third signaling triggering a first report according to an embodiment of this application;

[0182] Figure 15 shows a schematic diagram of a processing system based on artificial intelligence or machine learning according to an embodiment of this application;

[0183] Figure 16 illustrates a schematic diagram based on artificial intelligence or machine learning according to an embodiment of this application;

[0184] Figure 17 shows a schematic diagram of AI function deployment according to an embodiment of this application;

[0185] Figure 18 shows a schematic diagram of AI function deployment according to an embodiment of this application;

[0186] Figure 19 shows a schematic diagram of AI function deployment according to an embodiment of this application;

[0187] Figure 20 shows a schematic diagram of AI function deployment according to an embodiment of this application;

[0188] Figure 21 shows a structural block diagram of a processing apparatus for a first node according to an embodiment of the present application;

[0189] Figure 22 shows a structural block diagram of a processing apparatus for a second node according to an embodiment of the present application;

[0190] Figure 23 shows a schematic diagram of the first signaling, L1 feature sets, and the first inference configuration according to an embodiment of this application;

[0191] Figure 24 shows a flowchart of a first report according to an embodiment of this application;

[0192] Figure 25 illustrates the transmission between a first node and a second node according to an embodiment of this application;

[0193] Figure 26 shows a schematic diagram of a first inference according to an embodiment of this application;

[0194] Figure 27 shows a schematic diagram of a first data rate for a first inference according to an embodiment of this application;

[0195] Figure 28 shows a schematic diagram of a first data rate for a first inference according to an embodiment of this application;

[0196] Figure 29 shows a schematic diagram of a first channel according to an embodiment of this application;

[0197] Figure 30 shows a schematic diagram of a first data rate for a first channel according to an embodiment of this application;

[0198] Figure 31 shows a schematic diagram of a first data rate for a first channel according to an embodiment of this application;

[0199] Figure 32 shows a schematic diagram of a first reference time according to an embodiment of this application;

[0200] Figure 33 shows a schematic diagram of a first data rate relative to a first reference time according to an embodiment of this application;

[0201] Figure 34 shows a schematic diagram of a first data rate relative to a first reference time according to an embodiment of this application;

[0202] Figure 35 shows a schematic diagram of a first message and a second message according to an embodiment of this application;

[0203] Figure 36 shows a schematic diagram of a second signaling activation of a first inference configuration according to an embodiment of this application;

[0204] Figure 37 illustrates a schematic diagram of a second data rate for a first inference configuration according to an embodiment of this application;

[0205] Figure 38 shows a schematic diagram of a second report according to an embodiment of this application;

[0206] Figure 39 shows a schematic diagram of a first data rate according to an embodiment of this application;

[0207] Figure 40 shows a schematic diagram of a first data rate according to an embodiment of this application;

[0208] Figure 41 illustrates a schematic diagram of a first report being triggered by an event in a first event set according to an embodiment of this application;

[0209] Figure 42 illustrates a schematic diagram of a first signaling triggering a first report according to an embodiment of this application;

[0210] Figure 43 shows a structural block diagram of a processing apparatus for a first node according to an embodiment of the present application;

[0211] Figure 44 shows a structural block diagram of a processing apparatus for a second node according to an embodiment of this application. Detailed Implementation

[0212] The technical solutions of this application will be further described in detail below with reference to the accompanying drawings. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be arbitrarily combined with each other. Considering performance, flexibility, complexity, overhead, and compatibility, those skilled in the art are motivated to flexibly combine the embodiments in different drawings without conflict, such as, but not limited to, the embodiments in Figure 1 and the embodiments in Figures 5-23, the embodiments in Figure 24 and the embodiments in Figures 25-44, etc.

[0213] Example 1

[0214] Example 1 illustrates a flowchart of a first signaling and a first report according to an embodiment of this application, as shown in Figure 1. In Figure 1, each block represents a step. In particular, the order of the steps in the blocks does not represent a specific temporal relationship between the steps.

[0215] In Embodiment 1, the first node receives first signaling in step 101 and sends a first report in step 102. The first signaling includes a first inference configuration; the first report indicates support for L1 feature sets, where L1 is a positive integer; the L1 feature sets depend on the indication of the first signaling; any feature set in the L1 feature sets includes at least one of the following: maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, maximum data rate, and maximum modulation order; at least one indicator included in the first inference configuration corresponds to a feature that does not belong to the first feature set, and the first feature set is a feature set within the L1 feature sets.

[0216] As an example, the first signaling is carried by an RRC (Radio Resource Control) message.

[0217] As one embodiment, the first signaling is carried by higher layer signaling.

[0218] As an example, the first signaling is carried by RRC signaling.

[0219] As an example, the first signaling is carried by one or more RRC IE (Information Element).

[0220] As one embodiment, the first signaling includes some or all of the information of each of one or more RRC IEs.

[0221] As an example, the first signaling includes one or more of PDSCH-Config IE, PUSCH-Config IE, CSI-ReportConfig IE, CSI-MeasConfig, NZP-CSI-RS-ResourceSet IE, NZP-CSI-RS-Resource IE, PDSCH-ServingCellConfig IE, PUSCH-ServingCellConfig IE, SPS-Config IE, and ConfiguredGrantConfig IE.

[0222] As an example, the first signaling is carried by a MAC CE (Medium Access Control layer Control Element).

[0223] As an example, the first signaling is carried by DCI (Downlink Control Information).

[0224] As an example, the first signaling is carried by both RRC signaling and MAC CE.

[0225] As an example, the first signaling is carried by both RRC signaling and DCI.

[0226] As one embodiment, the first inference configuration includes higher-layer signaling.

[0227] As an example, the first inference configuration includes RRC signaling.

[0228] As one example, the first inference configuration includes one or more RRC IEs.

[0229] As one embodiment, the first inference configuration includes some or all of the information of each of one or more RRC IEs.

[0230] As an example, the first inference configuration includes at least one of PDSCH-Config IE, PUSCH-Config IE, CSI-ReportConfig IE, CSI-MeasConfig, NZP-CSI-RS-ResourceSet IE, NZP-CSI-RS-Resource IE, PDSCH-ServingCellConfig IE, PUSCH-ServingCellConfig IE, SPS-Config IE, and ConfiguredGrantConfig IE.

[0231] As an example, the first inference configuration includes a MAC CE.

[0232] As an example, the first inference configuration includes DCI.

[0233] As an example, the first inference configuration is carried by both RRC signaling and MAC CE.

[0234] As an example, the first inference configuration is carried by both RRC signaling and DCI.

[0235] As an example, the first inference configuration is used to configure an inference.

[0236] As an example, the first inference configuration includes inference-related parameters.

[0237] As an example, the inference refers to AI (Artificial Intelligence) inference.

[0238] As an example, the inference refers to ML (Machine Learning) inference.

[0239] As an example, the inference refers to AI inference or ML inference.

[0240] As an example, the first inference configuration is used to determine at least one of the inputs and outputs of an inference.

[0241] As an example, the first inference configuration is used to determine at least one of the three inference inputs, outputs, and models.

[0242] As an example, the first inference configuration is used to determine at least one of the inference input and the model.

[0243] As an example, the first inference configuration is used to determine at least one of the inputs and outputs of each of a plurality of inferences.

[0244] As an example, the first inference configuration is used to determine at least one of the input, output, and model for each of the plurality of inferences.

[0245] As an example, the first inference configuration is used to determine at least one of the input and model for each of the plurality of inferences.

[0246] As an example, the first inference configuration includes inference-related parameters.

[0247] As an example, the inference-related parameters include input-related parameters.

[0248] As an example, the inference-related parameters include input-related parameters and output-related parameters.

[0249] As an example, the inference-related parameters include input-related parameters, output-related parameters, and model parameters.

[0250] As an example, the inference-related parameters include input-related parameters and model parameters.

[0251] As an example, the first inference configuration includes inference-related parameters for a single inference.

[0252] As an example, the first inference configuration includes inference-related parameters for each of the plurality of inferences.

[0253] As an example, the input-related parameters are used to determine the RS (Reference Signal) resource for obtaining the inference dataset.

[0254] As an example, the RS resources include downlink RS resources.

[0255] As an example, the RS resources include at least one of CSI-RS (Channel State Information Reference Signal) resources and SS / PBCH (Synchronization Signal / Physical Broadcast Channel) block resources.

[0256] As an example, the RS resources include one or more of DMRS (Demodulation Reference Signal), PRS (Positioning Reference Signal) resources and PTRS (Phase-Tracking Reference Signal).

[0257] As one example, the RS resources include uplink RS resources.

[0258] As an example, the RS resource includes the SRS (Sounding Reference Signal) resource.

[0259] As an example, the input-related parameters are used to determine the physical layer channel for obtaining the inference dataset.

[0260] As an example, the physical layer channel includes at least one of PUSCH (Physical Uplink Shared Channel) and PDSCH (Physical Downlink Shared Channel).

[0261] As an example, the input-related parameters include configuration information for the physical layer channel used to obtain the inference dataset.

[0262] As a sub-implementation of the above embodiments, the configuration information includes one or more of the following: time-domain resources, frequency-domain resources, DMRS configuration, MCS (Modulation and Coding Scheme) table, and TCI (Transmission Configuration Indicator) status.

[0263] As an example, the input parameters are used to determine information related to the time instance being measured.

[0264] As an example, the information related to the measurement time instance includes the number of RS transmission opportunities used to obtain an input for inference.

[0265] As an example, the information related to the measurement time instance includes a lower limit on the number of RS transmission opportunities used to obtain an input for inference.

[0266] As an example, the output-related parameters are used to determine the content of the output of a reasoning.

[0267] As an example, the content includes one or more of CQI (Channel Quality Indicator), PMI (Precoding Matrix Indicator), CRI (CSI-RS Resource Indicator), LI (Layer Indicator), RI (Rank Indicator), SSBRI (SS / PBCH Block Resource Indicator), RSRP (Reference Signal received power), SINR (Signal-to-Interference and Noise Ratio), Capability Index, and TDCP (Time Domain Channel Properties).

[0268] As an example, the content includes one or more of CRI, SSBRI, RSRP, and SINR.

[0269] As one example, the content includes at least one of the following: predicted beam information, predicted CSI (Channel State Information), and compressed CSI.

[0270] As one example, the content includes location information, scheduling information, decoding or demodulation information, recovered data, or one or more of TB (Transport Block).

[0271] As an example, the output-related parameters are used to determine a set of RS resources to which the output of a reasoning is referred.

[0272] As an example, the output-related parameters are used to determine information related to the predicted time instance.

[0273] As an example, the information related to the predicted time instance includes the slot interval to which the inference output is targeted.

[0274] As an example, the information related to the prediction time instance includes at least one of the following: the number of time slot intervals targeted by the output of an inference, the length of the time slot intervals targeted by the output of an inference, and the gap between the time slot intervals targeted by the output of an inference and the time slots occupied by the reporting of the output of the inference.

[0275] As an example, the model parameters are used to determine a model for inference.

[0276] As an example, the model parameters include some or all of the parameters used to construct a model for inference.

[0277] As an example, the model parameters include a training dataset for an inference model.

[0278] As an example, the model parameters include a performance monitoring dataset of an inference model.

[0279] As an example, the model parameters include RS resources for obtaining a training dataset or performance monitoring dataset for an inference model.

[0280] As one example, the model parameters include an associated ID.

[0281] As an example, the associated ID is a non-negative integer.

[0282] As an example, the associated ID is a string.

[0283] As an example, the associated ID indicates an association between two or more RS resources, or between two or more groups of RS resources.

[0284] As a sub-implementation of the above embodiments, the association includes having the same or similar characteristics.

[0285] As a sub-implementation of the above embodiments, the association includes quasi-co-located.

[0286] As a sub-implementation of the above embodiments, the association includes quasi-co-addressing and the corresponding quasi-co-addressing type includes TypeD.

[0287] As a sub-example of the above embodiments, the association includes training datasets used to generate the same model.

[0288] As a sub-example of the above embodiments, the association includes inference datasets used to generate the same model.

[0289] As a sub-example of the above embodiments, the association includes training datasets or inference datasets used to generate the same model.

[0290] As a sub-example of the above embodiments, the association includes performance monitoring datasets or inference datasets used to generate the same model.

[0291] As an example, the features include one or more of delay spread, Doppler spread, Doppler shift, average delay, or spatial reception parameters.

[0292] As one embodiment, the feature includes a downlink transmit beam or a set of downlink transmit beams.

[0293] As an example, if two or more RS resources are associated with the same association identifier, the two or more RS resources are used to generate the same training dataset, inference dataset, or performance monitoring dataset for the same model.

[0294] As an example, if two or more RS resources are associated with the same association identifier, the two or more RS resources are used to generate the training dataset and inference dataset of the same model, respectively.

[0295] As an example, if two or more RS resources are associated with the same association identifier, the two or more RS resources are used to generate the same model's performance monitoring dataset and inference dataset, respectively.

[0296] As an example, if two RS resources are associated with the same association identifier, the two RS resources are quasi-co-located.

[0297] As an example, if two or more RS resources are associated with the same association identifier, the two or more RS resources have the same or similar characteristics.

[0298] As an example, if two sets of RS resources are associated with the same association identifier, one set of RS resources is used to generate a training dataset or inference dataset for a model, and the inference output of the model involves the other set of RS resources.

[0299] As an example, if two sets of RS resources are associated with the same association identifier, any RS resource in one set of RS resources and one RS resource in the other set of RS resources are quasi-co-located.

[0300] As an example, if two sets of RS resources are associated with the same association identifier, any RS resource in one set of RS resources and any RS resource in the other set of RS resources have the same or similar characteristics.

[0301] As an example, the association identifier indicates the association between a dataset and an inference.

[0302] As a sub-example of the above embodiments, the association includes the fact that the dataset belongs to the training dataset of the inference model.

[0303] As a sub-example of the above embodiments, the association includes the use of the dataset for training the inference model.

[0304] As an example, if a dataset and an inference are associated with the same association identifier, the dataset is either the training dataset or the inference dataset.

[0305] As an example, the association identifier indicates the association between one or a group of RS resources and an inference.

[0306] As a sub-example of the above embodiments, the association includes that the one or a group of RS resources are used to generate the training dataset, inference dataset, or performance monitoring dataset for the inference.

[0307] As a sub-example of the above embodiments, the association includes the output of the inference relating to the one or a set of RS resources.

[0308] As an example, if one or a group of RS resources are associated with the same association identifier as an inference, the one or a group of RS resources are used to generate at least one of the following three: a training dataset for the inference, an inference dataset, and a performance monitoring dataset.

[0309] As an example, if a set of RS resources and an inference are associated with the same association identifier, the output of the inference involves the set of RS resources.

[0310] As an example, the model parameters include an association identifier to which inference is associated.

[0311] As an example, a reasoning associated with an association identifier includes the reasoning being identified by the association identifier.

[0312] As an example, an association between a reasoning and an association identifier includes the model of the reasoning being identified by the association identifier.

[0313] As an example, an association between an inference and an association identifier includes the inference dataset of the inference being identified by the association identifier.

[0314] As an example, a reasoning associated with an association identifier includes the AI ​​function or AI entity performing the reasoning being identified by the association identifier.

[0315] As an example, an association between an inference and an association identifier includes the model training of the inference being identified by the association identifier.

[0316] As an example, an association between an inference and an association identifier includes the training dataset of the inference model being identified by the association identifier.

[0317] As an example, an association between an inference and an association identifier includes the AI ​​function or AI entity that performs the inference model training being identified by the association identifier.

[0318] As an example, an inference associated with an association identifier includes performance monitoring of the inference being identified by the association identifier.

[0319] As an example, an inference associated with an association identifier includes the performance monitoring dataset of the inference being identified by the association identifier.

[0320] As an example, an inference associated with an association identifier includes the function that the inference targets being identified by the association identifier.

[0321] As an example, an inference associated with an association identifier includes one or more RS resources used to generate the training dataset or inference dataset of the inference being associated with the association identifier.

[0322] As an example, an inference associated with an association identifier includes a set of RS resources involved in the output of the inference being associated with the association identifier.

[0323] As an example, the first signaling includes only the first inference configuration.

[0324] As an example, the first signaling includes multiple inference configurations, and the first inference configuration is one of the multiple inference configurations.

[0325] As an example, any of the plurality of inference configurations includes inference-related parameters.

[0326] As an example, any two of the plurality of inference configurations include inference-related parameters for different inferences.

[0327] As an example, any of the plurality of inference configurations includes at least one of PDSCH-Config IE, PUSCH-Config IE, CSI-ReportConfig IE, CSI-MeasConfig, NZP-CSI-RS-ResourceSet IE, NZP-CSI-RS-Resource IE, PDSCH-ServingCellConfig IE, PUSCH-ServingCellConfig IE, SPS-Config IE, and ConfiguredGrantConfig IE.

[0328] As an example, one of the multiple inference configurations is carried only by RRC signaling.

[0329] As an example, one of the multiple inference configurations is carried by both RRC signaling and DCI.

[0330] As an example, one of the multiple inference configurations is carried by both RRC signaling and MAC CE.

[0331] As an example, any of the plurality of inference configurations includes one or more metrics.

[0332] As a sub-implementation of the above embodiments, any one of the one or more indicators is one of the following: number of layers, number of downlink RS resources, number of uplink RS resources, bandwidth, data rate, and modulation order.

[0333] As an example, the first report is carried by a higher-layer message.

[0334] As an example, the first report is carried by an RRC (Radio Resource Control) message.

[0335] As an example, the first report is carried by RRC signaling.

[0336] As an example, the first report is carried by MAC CE (Medium Access Control layer Control Element).

[0337] As an example, the first report includes UE capability information.

[0338] As an example, the first report includes information from all or part of the domains in a UE capability IE.

[0339] As an example, the first report includes the capability report of the first node.

[0340] As an example, the first report applies only to one carrier or one serving cell of the first node.

[0341] As an example, the first report applies to all component carriers of the first node.

[0342] As an example, the first report applies to all component carriers belonging to the same cell group of the first node.

[0343] As an example, the first report applies to all component carriers of the first node that belong to the same band or band combination.

[0344] As an example, the first report applies to all serving cells of the first node.

[0345] As an example, the first report applies to all serving cells belonging to the same cell group as the first node.

[0346] As an example, the first report applies to all serving cells of the first node that belong to the same frequency band or frequency band combination.

[0347] Typically, the same cell group is either an MCG (Master Cell Group) or an SCG (Secondary Cell Group).

[0348] As one embodiment, the first signaling is carried by one or more RRC IEs, and the first report is carried by a MAC CE.

[0349] As an example, L1 is equal to 1.

[0350] As an example, L1 is greater than 1.

[0351] As an example, any one of the L1 feature sets includes some or all of the information in a UE capability IE.

[0352] As an example, any feature set in the L1 feature sets includes some or all of the information in a FeatureSet.

[0353] As an example, any feature set in the L1 feature sets includes a FeatureSet.

[0354] As an example, any feature set in the L1 feature sets includes some or all of the information in a FeatureSetCombination.

[0355] As an example, any feature set in the L1 feature sets includes a FeatureSetCombinatio.

[0356] As an example, any feature set in the L1 feature sets includes some or all of the information of a FeatureSetsPerBand.

[0357] As an example, any feature set in the L1 feature sets includes a FeatureSetsPerBand.

[0358] As an example, any feature set in the L1 feature sets includes a combination of FeatureSets at the same position in FeatureSetsPerBand across band in FeatureSetCombination.

[0359] As an example, any feature set in the L1 feature sets includes some or all of the information in a FeatureSetDownlink or FeatureSetUplink.

[0360] As an example, any feature set in the L1 feature sets includes a FeatureSetDownlink or FeatureSetUplink.

[0361] As an example, any feature set in the L1 feature sets includes some or all of the information in a FeatureSetDownlinkPerCC or FeatureSetUplinkPerCC.

[0362] As an example, any feature set in the L1 feature sets includes a FeatureSetDownlinkPerCC or a FeatureSetUplinkPerCC.

[0363] As an example, any feature set in the L1 feature sets includes a Dummy.

[0364] As an example, the Dummy includes one or more of DummyA, DummyB, DummyC, DummyD, and DummyE.

[0365] As an example, any feature set in the L1 feature sets includes only one feature.

[0366] As an example, one feature set in the L1 feature sets is FeatureSetCombination, and the other feature set in the L1 feature sets is FeatureSet.

[0367] As an example, one feature set in the L1 feature sets is FeatureSetsPerBand, and another feature set in the L1 feature sets is FeatureSet.

[0368] As an example, one feature set in the L1 feature sets is a combination of FeatureSets at the same position in FeatureSetsPerBand across frequency bands in FeatureSetCombination, and the other feature set in the L1 feature sets is FeatureSet.

[0369] As an example, one feature set in the L1 feature set is FeatureSetDownlink, and the other feature set in the L1 feature set is FeatureSetUplink.

[0370] As an example, one feature set in the L1 feature set is FeatureSetDownlinkPerCC, and the other feature set in the L1 feature set is FeatureSetUplinkPerCC.

[0371] As an example, one feature set in the L1 feature sets is a combination of one or more FeatureSets, and the other feature set in the L1 feature sets is a single feature.

[0372] As an example, one feature set in the L1 feature set is FeatureSetDownlinkPerCC or FeatureSetUplinkPerCC, and the other feature set in the L1 feature set is a single feature.

[0373] As an example, the L1 feature set is supported in one carrier or one serving cell of the first node.

[0374] As an example, the L1 feature sets are across all component carriers of the first node.

[0375] As an example, the L1 feature set consists of all component carriers belonging to the same cell group across the first node.

[0376] As an example, the L1 feature set consists of all component carriers belonging to the same frequency band or frequency band combination across the first node.

[0377] As an example, the L1 feature set spans all serving cells of the first node.

[0378] As an example, the L1 feature set is across all serving cells belonging to the same cell group of the first node.

[0379] As an example, the L1 feature set consists of all serving cells belonging to the same frequency band or frequency band combination across the first node.

[0380] As an example, any feature set in the L1 feature sets includes one feature.

[0381] As an example, any feature set in the L1 feature sets includes only one feature.

[0382] As an example, any feature set in the L1 feature sets includes one or more features.

[0383] As an example, one of the L1 feature sets contains only one feature.

[0384] As an example, one of the L1 feature sets includes multiple features.

[0385] As an example, any feature in any feature set of the L1 feature sets indicates the capability of the first node.

[0386] As an example, any feature in any feature set of the L1 feature sets indicates the index supported by the first node.

[0387] As an example, any feature set in the L1 feature set includes at least one of the following features: maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, maximum data rate, and maximum modulation order.

[0388] As an example, at least a portion of the L1 feature set includes one or more features that are different from the maximum number of layers, the maximum number of downlink RS resources, the maximum number of uplink RS resources, the supported bandwidth, the maximum data rate, and the maximum modulation order.

[0389] As an example, any feature set in the L1 feature set includes one or more features that are different from the maximum number of layers, the maximum number of downlink RS resources, the maximum number of uplink RS resources, the supported bandwidth, the maximum data rate, and the maximum modulation order.

[0390] As an example, the other features include whether AI / ML-based encoding and decoding is supported, whether AI / ML-based beam management and CSI reporting are supported, whether AI / ML-based positioning or AI / ML-assisted positioning is supported, whether AI / ML-based joint source-channel coding is supported, the maximum number of training datasets, etc.

[0391] As an example, the other features include the maximum number of component carriers, supported frequency band combinations, whether ACK-NACK feedback for multicast is supported, the maximum number of configurations granted, whether aperiodic beam reporting is supported, beam switching time, whether BWPs for different numbers are supported, supported primary carrier spacing, supported codebook combinations, supported codebook parameters, maximum number of search spaces, maximum number of CORESETs (Control Resource Sets), whether two CORESET pools are supported, etc.

[0392] As an example, the number of layers refers to the number of MIMO (Multiple Input Multiple Output) layers.

[0393] As an example, the number of layers refers to the number of transmission layers.

[0394] As an example, the number of downlink RS resources refers to the number of CSI-RS resources.

[0395] As an example, the number of downlink RS resources refers to the number of CSI-RS resource sets.

[0396] As an example, the number of downlink RS resources refers to the number of TRS (Tracking Reference Signal) resources.

[0397] As an example, the number of downlink RS resources refers to the number of TRS resource sets.

[0398] As an example, the number of uplink RS resources refers to the number of SRS (Sounding Reference Signal) resources.

[0399] As an example, the number of uplink RS resources refers to the number of SRS resource sets.

[0400] As an example, the unit of bandwidth is Hz (Hertz), kHz (kilohertz), or MHz (megahertz).

[0401] Considering future ultra-wideband transmission, the unit of bandwidth may also be GHz.

[0402] As an example, the bandwidth is expressed as the number of RBs (Resource Blocks).

[0403] As an example, the unit of the data rate is Mbps (megabits per second).

[0404] Considering future ultra-wideband transmission, the unit of the data rate may also be Gbps or Mbpms (megabits per millisecond).

[0405] As an example, the modulation order is one of π / 2BPSK (Binary Phase Shift Keying), BPSK, QPSK (Quadrature Phase Shift Keying), 16QAM (Quadrature Amplitude Modulation), 64QAM, 256QAM or 1024QAM.

[0406] As an example, the maximum modulation order is one of 64QAM, 256QAM or 1024QAM.

[0407] As an example, the maximum number of layers, the maximum number of downlink RS resources, the maximum number of uplink RS resources, the maximum data rate, and the maximum modulation order refer to the maximum number of supported layers, the maximum number of supported downlink RS resources, the maximum number of supported uplink RS resources, the maximum data rate, and the maximum modulation order, respectively.

[0408] As an example, one feature set in the L1 feature set includes at least one of the following features: maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, and maximum data rate. The maximum number of layers, the maximum number of downlink RS resources, the maximum number of uplink RS resources, the supported bandwidth, and the maximum data rate are respectively the maximum number of layers, the maximum number of downlink RS resources, the maximum number of uplink RS resources, the supported bandwidth, and the maximum data rate supported by the first node on a carrier or serving cell.

[0409] As an example, one feature set in the L1 feature set includes at least one of the following features: maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, and maximum data rate. The maximum number of layers, the maximum number of downlink RS resources, the maximum number of uplink RS resources, the supported bandwidth, and the maximum data rate are respectively the maximum values ​​of the total number of layers, downlink RS resources, uplink RS resources, bandwidth, and data rate supported by the first node on all carriers or all serving cells.

[0410] As a sub-example of the above embodiments, all carriers or all serving cells refer to all carriers or all serving cells belonging to the same frequency band or frequency band combination.

[0411] As a sub-example of the above embodiments, all carriers or all serving cells refer to all carriers or all serving cells belonging to the same cell group.

[0412] As an example, any feature in any feature set of the L1 feature sets corresponds to an index.

[0413] As an example, the index corresponding to at least one feature in any feature set of the L1 feature sets is one of the following: number of layers, number of downlink RS resources, number of uplink RS resources, bandwidth, data rate, or modulation order.

[0414] As an example, any feature in any feature set of the L1 feature sets includes the maximum value or support value of the index corresponding to that feature.

[0415] As an example, for any given feature in any feature set of the L1 feature sets, the index corresponding to the given feature is configured to the first node, and the given feature includes the maximum value or support value of the corresponding index reported by the first node.

[0416] As an example, any feature set in the L1 feature set includes at least one of the following features: maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, maximum data rate, and maximum modulation order. The index corresponding to any of the at least one feature is one of the following: number of layers, number of downlink RS resources, number of uplink RS resources, bandwidth, data rate, or modulation order.

[0417] As an example, one feature in the L1 feature set is the maximum number of layers, and the index corresponding to the feature is the number of layers.

[0418] As an example, one feature in the L1 feature set is the maximum number of downlink RS resources, and the index corresponding to the feature is the number of downlink RS resources.

[0419] As an example, one feature in the L1 feature set is the maximum number of uplink RS resources, and the index corresponding to the feature is the number of uplink RS resources.

[0420] As an example, one feature in the L1 feature set is the supported bandwidth, and the index corresponding to the feature is bandwidth.

[0421] As an example, one of the features in the L1 feature set is the maximum data rate, and the index corresponding to the feature is the data rate.

[0422] As an example, one feature in the L1 feature set is the maximum modulation order, and the index corresponding to the feature is the modulation order.

[0423] As an example, the index corresponding to a feature in any feature set of the L1 feature sets is one of the following: the number of layers configured for the first node, the maximum number of layers, the number of downlink RS resources, the number of uplink RS resources, bandwidth, data rate, or modulation order; the feature is one of the following: the maximum number of layers reported by the first node, the maximum number of downlink RS resources, the maximum number of uplink RS resources, the supported bandwidth, the maximum data rate, or the maximum modulation order.

[0424] As an example, one feature in the L1 feature set is the maximum number of layers reported by the first node, and the index corresponding to the feature is the number of layers or the maximum number of layers configured for the first node.

[0425] As an example, one feature in the L1 feature set is the maximum number of downlink RS resources reported by the first node, and the indicator corresponding to the feature is the number of downlink RS resources configured for the first node.

[0426] As an example, one feature in the L1 feature set is the maximum number of uplink RS resources reported by the first node, and the indicator corresponding to the feature is the number of uplink RS resources configured for the first node.

[0427] As an example, one feature in the L1 feature set is the supported bandwidth reported by the first node, and the indicator corresponding to the feature is the bandwidth configured for the first node.

[0428] As an example, one feature in the L1 feature set is the maximum data rate reported by the first node, and the metric corresponding to the feature is the data rate configured for the first node.

[0429] As an example, one feature in the L1 feature set is the maximum modulation order reported by the first node, and the index corresponding to the feature is the modulation order or the maximum modulation order configured for the first node.

[0430] As an example, the first inference configuration includes multiple metrics, which include one or more of the following: number of layers, maximum number of layers, number of downlink RS resources, number of uplink RS resources, bandwidth, data rate, or modulation order.

[0431] As an example, the first inference configuration includes one or more metrics that are different from the number of layers, maximum number of layers, number of downlink RS resources, number of uplink RS resources, bandwidth, data rate, and modulation order.

[0432] In a preferred embodiment, the feature corresponding to any one of the at least one indicators included in the first inference configuration does not belong to the first feature set.

[0433] The problem that the above method needs to solve includes how to take into account the mutual constraints between different indicators in feature reporting; in the above method, the characteristics of different indicators influence / constrain each other, thus solving this problem.

[0434] The advantages of the above method include that, considering the mutual influence / constraint between different indicators, it is conducive to the global optimization of resource allocation.

[0435] As an example, the feature corresponding to an indicator refers to the maximum value or support value of the indicator.

[0436] As an example, the feature corresponding to an indicator refers to the maximum value or support value of the indicator reported by the first node.

[0437] As an example, one metric is the number of layers, and the feature corresponding to the metric is the maximum number of layers.

[0438] As a sub-implementation of the above embodiment, the metric is the number of layers configured for the first node.

[0439] As a sub-implementation of the above embodiments, the feature corresponding to the one indicator is the maximum number of layers supported by the first node.

[0440] As a sub-implementation of the above embodiment, the indicator is the number of layers or the maximum number of layers configured for the first node, and the feature corresponding to the indicator is the maximum number of layers reported by the first node.

[0441] As an example, one metric is the number of downlink RS resources, and the characteristic corresponding to the metric is the maximum number of downlink RS resources.

[0442] As a sub-example of the above embodiment, the metric is the number of downlink RS resources configured for the first node.

[0443] As a sub-example of the above embodiments, the feature corresponding to the one indicator is the maximum number of downlink RS resources supported.

[0444] As an example, one metric is the number of uplink RS resources, and the characteristic corresponding to the metric is the maximum number of uplink RS resources.

[0445] As a sub-implementation of the above embodiment, the metric is the number of uplink RS resources configured for the first node.

[0446] As a sub-implementation of the above embodiments, the feature corresponding to the one indicator is the maximum number of supported uplink RS resources.

[0447] As an example, one metric is bandwidth, and the feature corresponding to the metric is the supported bandwidth.

[0448] As a sub-implementation of the above embodiment, the metric is the bandwidth configured for the first node.

[0449] As a sub-example of the above embodiments, the feature corresponding to the one indicator is the maximum supported bandwidth.

[0450] As an example, one of the metrics is the data rate, and the characteristic corresponding to the metric is the maximum data rate.

[0451] As a sub-implementation of the above embodiment, the metric is the data rate configured for the first node.

[0452] As a sub-example of the above embodiments, the feature corresponding to the one indicator is the maximum supported data rate.

[0453] As an example, one metric is the modulation order, and the characteristic corresponding to the metric is the maximum modulation order.

[0454] As a sub-implementation of the above embodiment, the index is the modulation order configured for the first node.

[0455] As a sub-example of the above embodiments, the feature corresponding to the one index is the supported large modulation order.

[0456] As a sub-implementation of the above embodiment, the indicator is the modulation order or maximum modulation order configured for the first node, and the feature corresponding to the indicator is the maximum modulation order reported by the first node.

[0457] As an example, the feature corresponding to at least one indicator included in the first inference configuration does not belong to the first feature set, and the first feature set does not include the maximum value of any one of the at least one indicators.

[0458] As an example, the features corresponding to at least one indicator included in the first inference configuration are not included in the first feature set, and the first feature set does not include the support value of any of the at least one indicator.

[0459] As an example, one metric included in the first inference configuration is the number of layers, and the first feature set does not include the maximum number of layers.

[0460] As an example, the first inference configuration includes an indicator of the maximum number of layers configured for the first node, and the first feature set does not include the maximum number of layers reported by the first node.

[0461] As an example, one metric included in the first inference configuration is the number of downlink RS resources, and the first feature set does not include the maximum number of downlink RS resources.

[0462] As an example, one metric included in the first inference configuration is the number of uplink RS resources, and the first feature set does not include the maximum number of uplink RS resources.

[0463] As an example, one metric included in the first inference configuration is bandwidth, and the first feature set does not include supported bandwidth.

[0464] As an example, one metric included in the first inference configuration is the data rate, and the first feature set does not include the maximum data rate.

[0465] As an example, one metric included in the first inference configuration is the modulation order, and the first feature set does not include the maximum modulation order.

[0466] As an example, the first inference configuration includes an indicator of the maximum modulation order configured for the first node, and the first feature set does not include the maximum modulation order reported by the first node.

[0467] As an example, the first inference configuration includes a metric that depends on the first inference configuration.

[0468] As one embodiment, the first inference configuration includes an indicator, which is used by the first node to calculate the indicator.

[0469] As an example, the first inference configuration includes an indicator that indicates the first inference configuration.

[0470] As one example, the first inference configuration includes multiple metrics.

[0471] As a sub-implementation of the above embodiments, the first inference configuration explicitly indicates some or all of the plurality of indicators.

[0472] As a sub-implementation of the above embodiments, the first inference configuration directly indicates some or all of the plurality of indicators.

[0473] As a sub-implementation of the above embodiments, the first inference configuration implicitly indicates some or all of the plurality of indicators.

[0474] As a sub-example of the above embodiments, the first inference configuration indicates some or all of the plurality of indicators by indicating other information.

[0475] As a sub-implementation of the above embodiments, the first inference configuration directly indicates a portion of the plurality of indicators, and indicates another portion of the plurality of indicators by indicating other information.

[0476] As an example, one metric included in the first inference configuration is the number of layers, which is indicated by indicating the DMRS port.

[0477] As an example, one metric included in the first inference configuration is the maximum number of layers, and the first inference configuration directly indicates the maximum number of layers.

[0478] As an example, one metric included in the first inference configuration is the data rate, which is indicated by indicating the modulation order or MCS table.

[0479] As an example, one metric included in the first inference configuration is the data rate, which is indicated by scheduling information of physical layer channels, including at least one of PDSCH and PUSCH.

[0480] As a sub-example of the above embodiments, the scheduling information of the physical layer channel includes one or more of time-domain resources, frequency-domain resources, period, time slot offset, MCS and DMRS ports.

[0481] As an example, one metric included in the first inference configuration is a data rate, which depends on the modulation order, MCS table, or physical layer channel scheduling information indicated by the first inference configuration, wherein the physical layer channel includes at least one of PDSCH and PUSCH.

[0482] As a sub-example of the above embodiments, the scheduling information of the physical layer channel includes one or more of time-domain resources, frequency-domain resources, period, time slot offset, MCS and DMRS ports.

[0483] As an example, one metric included in the first inference configuration is bandwidth, which is indicated by configuring one or more of the serving cell, carrier, and BWP (Bandwidth Part).

[0484] As an example, one metric included in the first inference configuration is bandwidth, which depends on at least one of the serving cell, carrier, and BWP configured in the first inference configuration.

[0485] As an example, one metric included in the first inference configuration is bandwidth, which is indicated by indicating scheduling information of physical layer channels, including one or more of PDCCH, PUCCH, PDSCH, and PUSCH.

[0486] As a sub-example of the above embodiments, the scheduling information includes one or more of the following: time domain resources, frequency domain resources, period, time slot offset, MCS, DMRS port, and CCE (Control Channel Element) to REG (Resource-Element Group) mapping.

[0487] As an example, one metric included in the first inference configuration is bandwidth, which depends on the scheduling information of the physical layer channels indicated by the first inference configuration, the physical layer channels including one or more of PDCCH, PUCCH, PDSCH and PUSCH.

[0488] As a sub-example of the above embodiments, the scheduling information includes one or more of the following: time domain resources, frequency domain resources, period, time slot offset, MCS, DMRS port, and CCE (Control Channel Element) to REG (Resource-Element Group) mapping.

[0489] As an example, one of the indicators included in the first inference configuration is the modulation order or the maximum modulation order, which is indicated by indicating the MCS table.

[0490] As an example, a feature set includes a feature that depends on the feature set.

[0491] As an example, a feature set includes a feature included, and the feature set is used to calculate the feature.

[0492] As an example, a feature set includes a feature that indicates the feature.

[0493] As a sub-implementation of the above embodiments, the feature set explicitly indicates the feature.

[0494] As a sub-implementation of the above embodiments, the feature set directly indicates the feature.

[0495] As a sub-implementation of the above embodiments, the feature set implicitly indicates the feature.

[0496] As a sub-implementation of the above embodiments, the feature set indicates the feature by indicating other information.

[0497] As an example, a feature set may include a maximum number of layers, and the feature set may directly indicate the maximum number of layers.

[0498] As an example, a feature set includes a feature of maximum data rate, which is indicated by indicating the maximum modulation order or MCS table.

[0499] As an example, a feature set may include a feature that is supported bandwidth, the feature set directly indicating the supported bandwidth.

[0500] As an example, a feature set includes a feature of supported bandwidth, which indicates the supported bandwidth by indicating at least one of supported band combination and bandwidth class.

[0501] As an example, one feature of a feature set is the maximum modulation order, and the feature set directly indicates the maximum modulation order.

[0502] As an example, the first feature set is any feature set in the L1 feature sets.

[0503] In a preferred embodiment, the feature corresponding to any one of the at least one indicators included in the first inference configuration does not belong to any feature set in the L1 feature sets.

[0504] As an example, the L1 feature sets depending on the indication of the first signaling include the L1 feature sets being conditional on the indication of the first signaling.

[0505] As an example, the L1 feature sets depending on the indication of the first signaling include the fact that the L1 feature sets are obtained under the conditions indicated by the first signaling.

[0506] As an example, the L1 feature sets depending on the indication of the first signaling include the first node being able to support the L1 feature sets under the indication of the first signaling.

[0507] As an example, the indication that the L1 feature sets depend on the first signaling includes that the L1 feature sets are conditional on the first inference configuration.

[0508] As an example, the L1 feature sets depend on the indication of the first signaling, including that the L1 feature sets are obtained under the conditions of the first inference configuration.

[0509] As an example, the indication that the L1 feature sets depend on the first signaling includes that the first node can support the L1 feature sets under the conditions of the first inference configuration.

[0510] As an example, the first node supports the L1 feature sets, provided that it supports all the metrics included in the first inference configuration.

[0511] As an example, the first signaling includes multiple inference configurations, each of which includes one or more metrics, and the first node supports the L1 feature set provided that it supports all the metrics included in the multiple inference configurations.

[0512] The advantages of the two embodiments described above include allowing the first node to adjust the feature set reporting according to the instructions of the first signaling, which provides better flexibility and accuracy, and further optimizes resource allocation.

[0513] As an example, when the indicators configured for the first node change compared to the indications of the first signaling, it cannot be assumed that the L1 feature sets can still be supported.

[0514] Example 2

[0515] Example 2 illustrates a schematic diagram of a network architecture according to an embodiment of this application, as shown in Figure 2.

[0516] Figure 2 illustrates network architecture 200. Network architecture 200 is a 5G NR (New Radio) / LTE (Long-Term Evolution) / LTE-A (Long-Term Evolution Advanced) system, or a 5G+ network architecture, or a 6G network architecture, or a network architecture adopted in future evolutions by 3GPP; network architecture 200 may be referred to as 5GS (5G System) / EPS (Evolved Packet System), or 6GS (6G System); network architecture 200 includes at least one of UE (User Equipment) 201, RAN (Radio Access Network) 202, core network 210, HSS (Home Subscriber Server) / UDM (Unified Data Management) 220, and Internet service 230. The network architecture 200 can interconnect with other access networks, but these entities / interfaces are not shown for simplicity. As shown, the network architecture 200 provides packet-switched services; however, those skilled in the art will readily understand that the various concepts presented throughout this application can be extended to networks providing circuit-switched services or other cellular networks. The RAN includes node 203. The RAN may also include other nodes 204. Node 203 provides user and control plane protocol termination toward UE 201. Node 203 may be connected to other nodes 204 via an Xn interface (e.g., backhaul) / X2 interface. Node 203 may also be referred to as a base station, base transceiver station, radio base station, radio transceiver, transceiver function, basic service set (BSS), extended service set (ESS), TRP (transmitter-receiver node), or some other suitable term. The core network 210 is a 5GC (5G Core Network) / EPC (Evolved Packet Core), or the core network 210 is a 6GC; node 203 provides UE 201 with an access point to the core network 210.Examples of UE201 include cellular phones, smartphones, Session Initiation Protocol (SIP) phones, laptops, personal digital assistants (PDAs), satellite radios, non-terrestrial base station communications, satellite mobile communications, global positioning systems, multimedia devices, video devices, digital audio players (e.g., MP3 players), cameras, game consoles, drones, aircraft, narrowband IoT devices, machine-type communication devices, land vehicles, automobiles, wearable devices, or any other similar functional devices. Those skilled in the art may also refer to UE201 as a mobile station, subscriber station, mobile unit, subscriber unit, radio unit, remote unit, mobile device, radio device, wireless communication device, remote device, mobile subscriber station, access terminal, mobile terminal, wireless terminal, remote terminal, handheld device, user agent, mobile client, client, or any other suitable term. Node 203 is connected to the core network 210 via an S1 / NG interface. The core network 210 includes an MME (Mobility Management Entity) / AMF (Authentication Management Field) / SMF (Session Management Function) 211, other MMEs / AMFs / SMFs 214, an S-GW (Service Gateway) / UPF (User Plane Function) 212, and a P-GW (Packet Data Network Gateway) / UPF 213. The MME / AMF / SMF 211 is the control node that handles signaling between the UE 201 and the core network 210. Generally, the MME / AMF / SMF 211 provides bearer and connection management. All user IP (Internet Protocol) packets are transmitted through the S-GW / UPF 212, which is itself connected to the P-GW / UPF 213. The P-GW provides UE IP address allocation and other functions. The P-GW / UPF 213 is connected to the Internet service 230. Internet services 230 include operator-compliant Internet protocol services, which may specifically include Internet, intranet, IMS (IP Multimedia Subsystem), and packet switching services.

[0517] As an example, the first node includes the UE201.

[0518] As one embodiment, the second node includes the node 203.

[0519] As an example, the wireless link between the UE201 and the node203 includes a cellular link.

[0520] As an example, the sender of the first signaling includes the node 203.

[0521] As an example, the recipient of the first signaling includes the UE201.

[0522] As an example, the sender of the first report includes the UE201.

[0523] As an example, the recipient of the first report includes node 203.

[0524] As one embodiment, the sender of the second signaling includes the node 203.

[0525] As an example, the recipient of the second signaling includes the UE201.

[0526] As an example, the sender of the second report includes the UE201.

[0527] As an example, the recipient of the second report includes the node 203.

[0528] As an example, the sender of the third signaling includes the node 203.

[0529] As an example, the recipient of the third signaling includes the UE201.

[0530] As an example, the sender of the first message includes the node 203.

[0531] As an example, the recipient of the first message includes the UE201.

[0532] As an example, the sender of the second message includes the UE201.

[0533] As an example, the recipient of the second message includes the node 203.

[0534] As an example, the UE201 supports AI- or ML-based operations.

[0535] As an example, node 203 supports AI- or ML-based operations.

[0536] Example 3

[0537] Example 3 illustrates a schematic diagram of an embodiment of a wireless protocol architecture for the user plane and control plane according to an embodiment of this application, as shown in Figure 3.

[0538] Example 3 illustrates a schematic diagram of an embodiment of a wireless protocol architecture for a user plane and control plane according to this application, as shown in Figure 3. Figure 3 is a schematic diagram illustrating an embodiment of a radio protocol architecture for a user plane 350 and a control plane 300. Figure 3 shows the radio protocol architecture for the control plane 300 between a first communication node device (UE, gNB, or RSU in V2X) and a second communication node device (gNB, UE, or RSU in V2X), or between two UEs, using three layers: Layer 1, Layer 2, and Layer 3. Layer 1 (L1 layer) is the lowest layer and implements various PHY (physical layer) signal processing functions. Layer 1 will be referred to herein as PHY 301. Layer 2 (L2 layer) 305 is above PHY 301 and is responsible for the link between the first communication node device and the second communication node device, or between two UEs. Layer L2 305 includes a MAC (Medium Access Control) sublayer 302, an RLC (Radio Link Control) sublayer 303, and a PDCP (Packet Data Convergence Protocol) sublayer 304, which terminate at the second communication node device. The PDCP sublayer 304 provides multiplexing between different radio bearers and logical channels. It also provides security through encrypted data packets and supports cross-cell mobility between the second communication node devices and the first communication node device. The RLC sublayer 303 provides upper-layer packet segmentation and reassembly, retransmission of lost packets, and packet reordering to compensate for out-of-order reception due to HARQ. The MAC sublayer 302 provides multiplexing between logical and transport channels. It is also responsible for allocating various radio resources (e.g., resource blocks) within a cell among the first communication node devices. Furthermore, the MAC sublayer 302 handles HARQ operations. In the control plane 300, the Radio Resource Control (RRC) sublayer 306 of Layer 3 (L3) is responsible for acquiring radio resources (i.e., radio bearers) and configuring the lower layers using RRC signaling between the second and first communication node devices. The user plane 350's radio protocol architecture includes Layer 1 (L1) and Layer 2 (L2). The radio protocol architecture for the first and second communication node devices in the user plane 350 is largely the same as the corresponding layers and sublayers in the control plane 300 for Physical Layer 351, PDCP sublayer 354 in L2 Layer 355, RLC sublayer 353 in L2 Layer 355, and MAC sublayer 352 in L2 Layer 355. However, PDCP sublayer 354 also provides header compression for upper layer data packets to reduce radio transmission overhead.The L2 layer 355 in the user plane 350 also includes an SDAP (Service Data Adaptation Protocol) sublayer 356, which is responsible for mapping between QoS streams and data radio bearers (DRBs) to support service diversity. Although not illustrated, the first communication node device may have several upper layers above the L2 layer 355, including a network layer (e.g., IP layer) terminating at the P-GW on the network side and an application layer terminating at the other end of the connection (e.g., a remote UE, server, etc.).

[0539] As an example, the wireless protocol architecture in Figure 3 is applicable to the first node.

[0540] As an example, the wireless protocol architecture in Figure 3 is applicable to the second node.

[0541] As an example, the higher layer mentioned in this application refers to the layer above the physical layer.

[0542] As an example, the first signaling is generated in the RRC sublayer 306.

[0543] As an example, the first signaling is generated in the MAC sublayer 302 or the MAC sublayer 352.

[0544] As an example, the first report is generated in the RLC sublayer 353.

[0545] As an example, the first report is generated in a layer above the MAC sublayer.

[0546] As an example, the first report is generated in the MAC sublayer 302 or the MAC sublayer 352.

[0547] As an example, the first report is generated in the PHY301 or the PHY351.

[0548] As an example, the second signaling is generated in the RRC sublayer 306.

[0549] As an example, the second report is generated in the RLC sublayer 353.

[0550] As an example, the second report is generated in a layer above the MAC sublayer.

[0551] As an example, the third signaling is generated in the MAC sublayer 302 or the MAC sublayer 352.

[0552] As an example, the third signaling is generated in the PHY301 or the PHY351.

[0553] As an example, the first message is generated in the RRC sublayer 306.

[0554] As an example, the second message is generated in the RRC sublayer 306.

[0555] As an example, the second message is generated in the MAC sublayer 302 or the MAC sublayer 352.

[0556] As one embodiment, the second signaling is generated in the MAC sublayer 302 or the MAC sublayer 352.

[0557] As an example, the second report is generated in the RRC sublayer 306.

[0558] As an example, the second report is generated in a layer above the RRC sublayer 306.

[0559] Example 4

[0560] Example 4 illustrates a schematic diagram of a first communication device and a second communication device according to an embodiment of this application, as shown in Figure 4. Figure 4 is a block diagram of a first communication device 410 and a second communication device 450 communicating with each other in an access network.

[0561] The first communication device 410 includes a controller / processor 475, a memory 476, a receiver processor 470, a transmitter processor 416, a multi-antenna receiver processor 472, a multi-antenna transmitter processor 471, a transmitter / receiver 418, and an antenna 420.

[0562] The second communication device 450 includes a controller / processor 459, a memory 460, a data source 467, a transmitting processor 468, a receiving processor 456, a multi-antenna transmitting processor 457, a multi-antenna receiving processor 458, a transmitter / receiver 454, and an antenna 452.

[0563] In the transmission from the first communication device 410 to the second communication device 450, at the first communication device 410, upper-layer data packets from the core network are provided to the controller / processor 475. The controller / processor 475 implements L2 layer functionality. In DL (Downlink), the controller / processor 475 provides header compression, encryption, packet segmentation and reordering, multiplexing between logical and transport channels, and radio resource allocation to the second communication device 450 based on various priority metrics. The controller / processor 475 is also responsible for HARQ operation, retransmission of lost packets, and signaling to the second communication device 450. The transmit processor 416 and the multi-antenna transmit processor 471 implement various signal processing functions for L1 layer (i.e., physical layer). Transmit processor 416 performs encoding and interleaving to facilitate forward error correction (FEC) at the second communication device 450, and constellation mapping based on various modulation schemes (e.g., binary phase shift keying (BPSK), quadrature phase shift keying (QPSK), M-phase shift keying (M-PSK), and M-quadrature amplitude modulation (M-QAM). Multi-antenna transmit processor 471 performs digital spatial precoding on the encoded and modulated symbols, including codebook-based precoding and non-codebook-based precoding, and beamforming processing, generating one or more parallel... The transmit processor 416 then maps each parallel stream to a subcarrier, multiplexes the modulated symbols with a reference signal (e.g., a pilot) in the time and / or frequency domains, and then uses an inverse fast Fourier transform (IFFT) to generate a physical channel carrying the time-domain multicarrier symbol stream. The multi-antenna transmit processor 471 then performs transmit analog precoding / beamforming operations on the time-domain multicarrier symbol stream. Each transmitter 418 converts the baseband multicarrier symbol stream provided by the multi-antenna transmit processor 471 into an RF stream, which is then provided to a different antenna 420.

[0564] In the transmission from the first communication device 410 to the second communication device 450, at the second communication device 450, each receiver 454 receives a signal through its corresponding antenna 452. Each receiver 454 recovers the information modulated onto the radio frequency carrier and converts the radio frequency stream into a baseband multicarrier symbol stream, which is then provided to the receiver processor 456. The receiver processor 456 and the multi-antenna receiver processor 458 implement various signal processing functions of the L1 layer. The multi-antenna receiver processor 458 performs receive analog precoding / beamforming operations on the baseband multicarrier symbol stream from the receiver 454. The receiver processor 456 uses a Fast Fourier Transform (FFT) to convert the baseband multicarrier symbol stream after the receive analog precoding / beamforming operations from the time domain to the frequency domain. In the frequency domain, the physical layer data signal and the reference signal are demultiplexed by the receiver processor 456, where the reference signal is used for channel estimation, and the data signal is recovered in the multi-antenna receiver processor 458 after multi-antenna detection to recover any parallel stream destined for the second communication device 450. Symbols on each parallel stream are demodulated and recovered in the receive processor 456, generating soft decisions. The receive processor 456 then decodes and deinterleaves the soft decisions to recover the upper-layer data and control signals transmitted over the physical channel by the first communication device 410. The upper-layer data and control signals are then provided to the controller / processor 459. The controller / processor 459 implements the functions of Layer 2 (L2). The controller / processor 459 may be associated with a memory 460 storing program code and data. The memory 460 may be referred to as computer-readable media. In the DL (Layered Logic), the controller / processor 459 provides multiplexing, packet reassembly, decryption, header decompression, and control signal processing between the transmission and logical channels to recover upper-layer packets from the core network. The upper-layer packets are then provided to all protocol layers above Layer 2. Various control signals may also be provided to Layer 3 (L3) for L3 processing. The controller / processor 459 is also responsible for error detection using ACK and / or NACK protocols to support HARQ operation.

[0565] In the transmission from the second communication device 450 to the first communication device 410, at the second communication device 450, a data source 467 is used to provide upper-layer data packets to the controller / processor 459. The data source 467 represents all protocol layers above the L2 layer. Similar to the transmission functions at the first communication device 410 described in the DL, the controller / processor 459 implements header compression, encryption, packet segmentation and reordering, and multiplexing between logical and transport channels based on the radio resource allocation of the first communication device 410, implementing L2 layer functions for the user plane and control plane. The controller / processor 459 is also responsible for HARQ operations, retransmission of lost packets, and signaling to the first communication device 410. Transmit processor 468 performs modulation mapping and channel coding processing, while multi-antenna transmit processor 457 performs digital multi-antenna spatial precoding, including codebook-based and non-codebook-based precoding, and beamforming processing. Subsequently, transmit processor 468 modulates the generated parallel stream into a multi-carrier / single-carrier symbol stream. After analog precoding / beamforming operations in multi-antenna transmit processor 457, the stream is provided to different antennas 452 via transmitter 454. Each transmitter 454 first converts the baseband symbol stream provided by multi-antenna transmit processor 457 into a radio frequency symbol stream before providing it to antenna 452.

[0566] In the transmission from the second communication device 450 to the first communication device 410, the function at the first communication device 410 is similar to the receiving function at the second communication device 450 described in the transmission from the first communication device 410 to the second communication device 450. Each receiver 418 receives radio frequency signals through its corresponding antenna 420, converts the received radio frequency signals into baseband signals, and provides the baseband signals to the multi-antenna receiving processor 472 and the receiving processor 470. The receiving processor 470 and the multi-antenna receiving processor 472 jointly implement the L1 layer functions. The controller / processor 475 implements the L2 layer functions. The controller / processor 475 may be associated with a memory 476 that stores program code and data. The memory 476 may be referred to as computer-readable media. The controller / processor 475 provides multiplexing, packet reassembly, decryption, header decompression, and control signal processing between the transmission and logical channels to recover upper-layer data packets from the second communication device 450. The upper-layer data packets from the controller / processor 475 may be provided to the core network. The controller / processor 475 is also responsible for error detection using ACK and / or NACK protocols to support HARQ operation.

[0567] As one embodiment, the second communication device 450 includes: at least one processor and at least one memory, the at least one memory including computer program code; the at least one memory and the computer program code are configured to be used with the at least one processor. The second communication device 450 means at least: receiving the first signaling; and sending the first report. The first signaling includes a first inference configuration; the first report indicates support for L1 feature sets, where L1 is a positive integer; the L1 feature sets depend on the indication of the first signaling; any feature set in the L1 feature sets includes at least one of a maximum number of layers, a maximum number of downlink RS resources, a maximum number of uplink RS resources, supported bandwidth, a maximum data rate, and a maximum modulation order; the feature corresponding to at least one indicator included in the first inference configuration does not belong to the first feature set, and the first feature set is a feature set within the L1 feature sets.

[0568] As one embodiment, the second communication device 450 includes: a memory storing a computer-readable instruction program that produces actions when executed by at least one processor, the actions including: receiving the first signaling; and sending the first report.

[0569] As one embodiment, the first communication device 410 includes: at least one processor and at least one memory, the at least one memory including computer program code; the at least one memory and the computer program code are configured to be used with the at least one processor. The first communication device 410 means at least: transmitting the first signaling; and receiving the first report. The first signaling includes a first inference configuration; the first report indicates support for L1 feature sets, where L1 is a positive integer; the L1 feature sets depend on the indication of the first signaling; any feature set in the L1 feature sets includes at least one of a maximum number of layers, a maximum number of downlink RS resources, a maximum number of uplink RS resources, supported bandwidth, a maximum data rate, and a maximum modulation order; the feature corresponding to at least one indicator included in the first inference configuration does not belong to the first feature set, and the first feature set is a feature set within the L1 feature sets.

[0570] As one embodiment, the first communication device 410 includes: a memory storing a computer-readable instruction program that produces actions when executed by at least one processor, the actions including: sending the first signaling; and receiving the first report.

[0571] As an example, the first node in this application includes the second communication device 450.

[0572] As an example, the second node in this application includes the first communication device 410.

[0573] As an example, at least one of {the antenna 452, the receiver 454, the receiving processor 456, the multi-antenna receiving processor 458, the controller / processor 459, the memory 460, and the data source 467} is used to receive the first signaling; at least one of {the antenna 420, the transmitter 418, the transmitting processor 416, the multi-antenna transmitting processor 471, the controller / processor 475, and the memory 476} is used to transmit the first signaling.

[0574] As an example, at least one of {the antenna 420, the receiver 418, the receiving processor 470, the multi-antenna receiving processor 472, the controller / processor 475, and the memory 476} is used to receive the first report; at least one of {the antenna 452, the transmitter 454, the transmitting processor 468, the multi-antenna transmitting processor 457, the controller / processor 459, the memory 460, and the data source 467} is used to transmit the first report.

[0575] As an example, at least one of {the antenna 452, the receiver 454, the receiving processor 456, the multi-antenna receiving processor 458, the controller / processor 459, the memory 460, and the data source 467} is used to receive the second signaling; at least one of {the antenna 420, the transmitter 418, the transmitting processor 416, the multi-antenna transmitting processor 471, the controller / processor 475, and the memory 476} is used to transmit the second signaling.

[0576] As an example, at least one of {the antenna 420, the receiver 418, the receiving processor 470, the multi-antenna receiving processor 472, the controller / processor 475, and the memory 476} is used to receive the second report; at least one of {the antenna 452, the transmitter 454, the transmitting processor 468, the multi-antenna transmitting processor 457, the controller / processor 459, the memory 460, and the data source 467} is used to transmit the second report.

[0577] As an example, at least one of {the antenna 452, the receiver 454, the receiving processor 456, the multi-antenna receiving processor 458, the controller / processor 459, the memory 460, and the data source 467} is used to receive the third signaling; at least one of {the antenna 420, the transmitter 418, the transmitting processor 416, the multi-antenna transmitting processor 471, the controller / processor 475, and the memory 476} is used to transmit the third signaling.

[0578] As one embodiment, the second communication device 450 includes: at least one processor and at least one memory, the at least one memory including computer program code; the at least one memory and the computer program code are configured to be used with the at least one processor. The second communication device 450 is equipped to at least: transmit the first report. The first report indicates that the idle data rate capability is a first data rate, the idle data rate capability being configured for a first inference.

[0579] As one embodiment, the second communication device 450 includes: a memory storing a computer-readable instruction program that produces an action when executed by at least one processor, the action including: sending the first report.

[0580] As one embodiment, the first communication device 410 includes: at least one processor and at least one memory, the at least one memory including computer program code; the at least one memory and the computer program code are configured to be used with the at least one processor. The first communication device 410 means at least: receiving the first report. The first report indicates that the idle data rate capability is a first data rate, the idle data rate capability being configured for a first inference.

[0581] As one embodiment, the first communication device 410 includes: a memory storing a computer-readable instruction program that produces an action when executed by at least one processor, the action including: receiving the first report.

[0582] As an example, the first node in this application includes the second communication device 450.

[0583] As an example, the second node in this application includes the first communication device 410.

[0584] As an example, at least one of {the antenna 420, the receiver 418, the receiving processor 470, the multi-antenna receiving processor 472, the controller / processor 475, and the memory 476} is used to receive the first report; at least one of {the antenna 452, the transmitter 454, the transmitting processor 468, the multi-antenna transmitting processor 457, the controller / processor 459, the memory 460, and the data source 467} is used to transmit the first report.

[0585] As an example, at least one of {the antenna 452, the receiver 454, the receiving processor 456, the multi-antenna receiving processor 458, the controller / processor 459, the memory 460, and the data source 467} is used to receive the first message; at least one of {the antenna 420, the transmitter 418, the transmitting processor 416, the multi-antenna transmitting processor 471, the controller / processor 475, and the memory 476} is used to transmit the first message.

[0586] As an example, at least one of {the antenna 420, the receiver 418, the receiving processor 470, the multi-antenna receiving processor 472, the controller / processor 475, and the memory 476} is used to receive the second message; at least one of {the antenna 452, the transmitter 454, the transmitting processor 468, the multi-antenna transmitting processor 457, the controller / processor 459, the memory 460, and the data source 467} is used to transmit the second message.

[0587] As an example, at least one of {the antenna 452, the receiver 454, the receiving processor 456, the multi-antenna receiving processor 458, the controller / processor 459, the memory 460, and the data source 467} is used to receive the first signaling; at least one of {the antenna 420, the transmitter 418, the transmitting processor 416, the multi-antenna transmitting processor 471, the controller / processor 475, and the memory 476} is used to transmit the first signaling.

[0588] As an example, at least one of {the antenna 452, the receiver 454, the receiving processor 456, the multi-antenna receiving processor 458, the controller / processor 459, the memory 460, and the data source 467} is used to receive the second signaling; at least one of {the antenna 420, the transmitter 418, the transmitting processor 416, the multi-antenna transmitting processor 471, the controller / processor 475, and the memory 476} is used to transmit the second signaling.

[0589] As an example, at least one of {the antenna 420, the receiver 418, the receiving processor 470, the multi-antenna receiving processor 472, the controller / processor 475, and the memory 476} is used to receive the second report; at least one of {the antenna 452, the transmitter 454, the transmitting processor 468, the multi-antenna transmitting processor 457, the controller / processor 459, the memory 460, and the data source 467} is used to transmit the second report.

[0590] Example 5

[0591] Example 5 illustrates a transmission flowchart according to an embodiment of this application; as shown in Figure 5. In Figure 5, the second node U1 and the first node U2 are communication nodes transmitted via an air interface. In Figure 5, the steps in blocks F51 to F53 are optional.

[0592] For the second node U1, a second signaling is sent in step S5101; a second report is received in step S5102; a first signaling is sent in step S511; a third signaling is sent in step S5103; and a first report is received in step S512.

[0593] For the first node U2, in step S5201, a second signaling is received; in step S5202, a first report is sent; in step S521, a first signaling is received; in step S5203, a third signaling is received; and in step S522, a first report is sent.

[0594] In Embodiment 5, the first signaling includes a first inference configuration; the first report indicates support for L1 feature sets, where L1 is a positive integer; the L1 feature sets depend on the indication of the first signaling; any feature set in the L1 feature sets includes at least one of the following: maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, maximum data rate, and maximum modulation order; the feature corresponding to at least one indicator included in the first inference configuration does not belong to the first feature set, and the first feature set is a feature set in the L1 feature sets.

[0595] As an example, the first node U2 is the first node in this application.

[0596] As an example, the second node U1 is the second node in this application.

[0597] As one embodiment, the air interface between the second node U1 and the first node U2 includes a wireless interface between the base station equipment and the user equipment.

[0598] As one embodiment, the air interface between the second node U1 and the first node U2 includes a wireless interface between the relay node device and the user equipment.

[0599] As one embodiment, the air interface between the second node U1 and the first node U2 includes the interface between the core network equipment and the user equipment.

[0600] As one embodiment, the air interface between the second node U1 and the first node U2 includes the interface between the OTT server (Over-The-Top server) and the user equipment.

[0601] As one embodiment, the air interface between the second node U1 and the first node U2 includes the interface between the NAS (Network Access Server) device and the user equipment.

[0602] As one embodiment, the air interface between the second node U1 and the first node U2 includes a wireless interface between user equipment and user equipment.

[0603] As one embodiment, the first node U2 includes a terminal.

[0604] As one embodiment, the first node U2 includes a user equipment.

[0605] As one embodiment, the second node U1 includes the serving cell sustaining base station of the first node U2.

[0606] As one embodiment, the second node U1 includes an OTT server (Over-The-Top server).

[0607] As an example, the second node U1 includes OAM (Operation Administration and Maintenance).

[0608] As one embodiment, the second node U1 includes a NAS device.

[0609] As one embodiment, the second node U1 includes core network equipment.

[0610] As an example, the first signaling is transmitted on PDSCH (Physical Downlink Shared Channel).

[0611] As an example, the first signaling is transmitted on PDSCH and PDCCH (Physical Downlink Control Channel).

[0612] As an example, the physical layer channel occupied by the first signaling includes PDSCH.

[0613] As an example, the physical layer channels occupied by the first signaling include PDSCH and PDCCH.

[0614] As an example, the first report is transmitted on PUSCH (Physical Uplink Shared Channel).

[0615] As an example, the first report is transmitted on PUCCH (Physical Uplink Control Channel).

[0616] As an example, the physical layer channel occupied by the first report includes PUSCH.

[0617] As an example, the step in block F52 of Figure 5 includes a second report indicating support for L0 feature sets, where L0 is a positive integer greater than 1; wherein the second report is sent before the first report, and the L1 feature sets depend on the L0 feature sets.

[0618] As an example, the second report is transmitted on PUSCH.

[0619] As an example, the step in block F51 of Figure 5 includes the second signaling indicating the first configuration information; wherein the L0 feature sets depend on the indication of the second signaling.

[0620] As an example, the second signaling is transmitted on the PBCH (Physical Broadcast Channel).

[0621] As an example, the second signaling is transmitted on the PDSCH.

[0622] As an example, the L0 feature sets are conditional on a first hypothesis; the first hypothesis includes the association between the inference configuration and the training dataset.

[0623] As an example, the L0 feature sets are conditional on a second hypothesis; the second hypothesis includes not using reasoning.

[0624] As an example, the first report is triggered by an event in a first event set, the first event set including at least one state change of inference configuration, the state change including one or more of being activated, deactivated, configured and reconfigured.

[0625] As an example, the step in block F53 of Figure 5 exists, whereby the third signaling triggers the first report.

[0626] As an example, the third signaling is transmitted on the PDCCH.

[0627] As an example, the third signaling is transmitted on the PDSCH.

[0628] Example 6

[0629] Example 6 illustrates a schematic diagram of a second report according to one embodiment of this application; as shown in Figure 6. In Example 6, the second report indicates support for L0 feature sets.

[0630] As an example, the second report is sent before the first signaling is received.

[0631] The advantages of the above method include allowing the first node to adjust or update the feature set reported according to the instructions of the first signaling, improving the flexibility and accuracy of reporting, and further improving system efficiency.

[0632] As an example, the second report is carried by a higher-level message.

[0633] As an example, the second report is carried by an RRC message.

[0634] As an example, the second report is carried by RRC signaling.

[0635] As an example, the second report is carried by MAC CE.

[0636] As one example, the second report includes UE capability information.

[0637] As an example, the second report is carried by the UE capability IE.

[0638] As one example, the second report includes all or part of the information in a UE capability IE.

[0639] As an example, the second report includes information from one or more UE capability IEs.

[0640] As an example, the second report includes all or part of the information in the FeatureSetCombination IE.

[0641] As one example, the second report includes all or part of the information in the BandCombinationList IE.

[0642] As an example, the second report includes all or part of the information in the FeatureSets IE.

[0643] As one example, the second report includes the capability report of the first node.

[0644] As an example, the second report includes the UE processing capability of the first node.

[0645] As one example, the second report includes the UE capability indication of the first node.

[0646] As one example, the second report applies only to one carrier or one serving cell of the first node.

[0647] As an example, the second report applies to all component carriers of the first node.

[0648] As one example, the second report applies to all component carriers belonging to the same cell group of the first node.

[0649] As one embodiment, the second report applies to all component carriers of the first node that belong to the same band or band combination.

[0650] As one example, the second report applies to all serving cells of the first node.

[0651] As one example, the second report applies to all serving cells belonging to the same cell group as the first node.

[0652] As one example, the second report applies to all serving cells of the first node that belong to the same frequency band or frequency band combination.

[0653] As an example, any one of the L0 feature sets includes some or all of the information in a UE capability IE.

[0654] As an example, any feature set in the L0 feature sets includes a FeatureSet.

[0655] As an example, any feature set in the L0 feature sets includes a FeatureSetCombinatio.

[0656] As an example, any feature set in the L0 feature sets includes a FeatureSetsPerBand.

[0657] As an example, any feature set in the L0 feature sets includes a combination of FeatureSets at the same position in FeatureSetsPerBand across band in FeatureSetCombination.

[0658] As an example, any feature set in the L0 feature sets includes a FeatureSetDownlink or FeatureSetUplink.

[0659] As an example, any feature set in the L0 feature sets includes a FeatureSetDownlinkPerCC or a FeatureSetUplinkPerCC.

[0660] As an example, any feature set in the L0 feature sets includes a Dummy.

[0661] As an example, one feature set in the L0 feature set includes FeatureSetDownlink, and the other feature set in the L0 feature set includes FeatureSetUplink.

[0662] As an example, one feature set in the L0 feature set includes FeatureSetDownlinkPerCC, and the other feature set in the L0 feature set includes FeatureSetUplinkPerCC.

[0663] As one example, the second report includes one or more UE capability IEs, and the first report includes a MAC CE.

[0664] As one example, the second report is carried by RRC signaling, and the first report is carried by MAC CE.

[0665] As an example, any feature set in the L0 feature sets includes one or more features.

[0666] As an example, each feature set in one or more feature sets of the L0 feature sets includes at least one of the following features: maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, maximum data rate, and maximum modulation order.

[0667] As an example, one or more feature sets in the L0 feature set include features other than the maximum number of layers, the maximum number of downlink RS resources, the maximum number of uplink RS resources, the supported bandwidth, the maximum data rate, and the maximum modulation order.

[0668] Examples of the other features are described in Example 1.

[0669] As an example, at least two feature sets in the L0 feature sets include different features.

[0670] As an example, one feature set in the L0 feature sets does not include one or more features from another feature set in the L0 feature sets.

[0671] As an example, at least two feature sets in the L0 feature sets include features corresponding to different indicators.

[0672] As an example, one feature set in the L0 feature set includes the maximum number of layers, while the other feature set in the L0 feature set does not include the maximum number of layers.

[0673] As an example, one feature set in the L0 feature set includes the maximum number of downlink RS resources, while the other feature set in the L0 feature set does not include the maximum number of downlink RS resources.

[0674] As an example, one feature set in the L0 feature set includes the maximum number of uplink RS resources, while the other feature set in the L0 feature set does not include the maximum number of uplink RS resources.

[0675] As an example, one feature set in the L0 feature set includes the supported bandwidth, while the other feature set in the L0 feature set does not include the supported bandwidth.

[0676] As an example, one feature set in the L0 feature set includes the maximum data rate, while the other feature set in the L0 feature set does not include the maximum data rate.

[0677] As an example, one feature set in the L0 feature set includes the maximum modulation order, while the other feature set in the L0 feature set does not include the maximum modulation order.

[0678] As an example, at least two feature sets in the L0 feature sets include features corresponding to the same index.

[0679] As a sub-example of the above embodiments, the same index corresponds to different values ​​of the features in the at least two feature sets.

[0680] As an example, both feature sets in the L0 feature sets include a maximum number of layers, but the maximum number of layers in one feature set is not equal to the maximum number of layers in the other feature set.

[0681] As an example, both feature sets in the L0 feature sets include the maximum number of downlink RS resources, but the maximum number of downlink RS resources included in one feature set of the two feature sets is not equal to the maximum number of downlink RS resources included in the other feature set of the two feature sets.

[0682] As an example, both feature sets in the L0 feature sets include the maximum number of uplink RS resources, but the maximum number of uplink RS resources included in one feature set of the two feature sets is not equal to the maximum number of uplink RS resources included in the other feature set of the two feature sets.

[0683] As an example, both feature sets in the L0 feature sets include supported bandwidth, but the supported bandwidth included in one feature set of the two feature sets is not equal to the supported bandwidth included in the other feature set of the two feature sets.

[0684] As an example, both feature sets in the L0 feature sets include a maximum data rate, but the maximum data rate included in one feature set of the two feature sets is not equal to the maximum data rate included in the other feature set of the two feature sets.

[0685] As an example, both feature sets in the L0 feature sets include the maximum modulation order, but the maximum modulation order included in one feature set of the two feature sets is different from the maximum modulation order included in the other feature set of the two feature sets.

[0686] Example 7

[0687] Example 7 illustrates a schematic diagram of L1 feature sets depending on L0 feature sets according to an embodiment of the present application; as shown in Figure 7.

[0688] As an example, the L1 feature sets depend on the L0 feature sets, wherein the first node determines the L1 feature sets based on the L0 feature sets.

[0689] As an example, the L1 feature sets depending on the L0 feature sets include the indication of the L1 feature sets depending on the L0 feature sets.

[0690] As an example, the L1 feature sets depending on the L0 feature sets include the target recipient of the first report interpreting the L1 feature sets based on the L0 feature sets.

[0691] As an example, the L1 feature sets depend on the L0 feature sets, and the L1 feature sets are subsets of the L0 feature sets.

[0692] As an example, the L1 feature sets depend on the L0 feature sets, wherein any feature set in the L1 feature sets is a feature set in the L0 feature sets.

[0693] As a sub-implementation of the above embodiment, the L1 feature sets are feature sets that are still supported in the L0 feature sets.

[0694] As a sub-implementation of the above embodiments, any feature set other than the L1 feature set in the L0 feature set is no longer supported.

[0695] As a sub-example of the above embodiments, the first report indicates the L1 feature sets from the L0 feature sets.

[0696] As a sub-example of the above embodiment, the first report indicates a feature set that is no longer supported from the L0 feature set, and the L1 feature set is the feature set in the L0 feature set other than the feature set that is no longer supported.

[0697] As an example, any feature set in the L1 feature sets is a subset of a feature set in the L0 feature sets.

[0698] As an example, the L1 feature sets depend on the L0 feature sets, wherein the index corresponding to any feature in any feature set of the L1 feature sets is the index corresponding to a feature in a feature set of the L0 feature sets.

[0699] As an example, the L1 feature sets depend on the L0 feature sets, wherein any feature set in the L1 feature sets indicates that at least a portion of the features in one of the feature sets in the L0 feature sets are updated.

[0700] As an example, the second feature set in the L1 feature set indicates that some features in the third feature set in the L0 feature set have been updated, and the first report indicates the third feature set and indicates the updated partial features.

[0701] As a sub-implementation of the above embodiments, the second feature set indicates the updated partial features.

[0702] As a sub-implementation of the above embodiments, the partial features include a maximum number of layers, the third feature set indicates that the maximum number of layers is L1, and the second feature set indicates that the maximum number of layers is updated to Q1.

[0703] As a reference embodiment of the above sub-example, the second feature set directly indicates Q1.

[0704] As a reference embodiment of the above sub-example, the second feature set indicates the difference between Q1 and L1.

[0705] As a reference embodiment of the above sub-example, Q1 is greater than L1, or Q1 is less than L1.

[0706] As a sub-implementation of the above embodiments, the partial features include the maximum number of downlink RS resources or the maximum number of uplink RS resources, the third feature set indicates that the maximum number of downlink RS resources or the maximum number of uplink RS resources is L2, and the second feature set indicates that the maximum number of downlink RS resources or the maximum number of uplink RS resources is updated to Q2.

[0707] As a reference embodiment of the above sub-example, the second feature set directly indicates Q2.

[0708] As a reference embodiment of the above sub-example, the second feature set indicates the difference between Q2 and L2.

[0709] As a reference embodiment of the above sub-example, Q2 is greater than L2, or Q2 is less than L2.

[0710] As a sub-implementation of the above embodiments, the partial features include supported bandwidth, the third feature set indicates that the supported bandwidth is L3 MHz, and the second feature set indicates that the supported bandwidth has been updated to Q3 MHz.

[0711] As a reference embodiment of the above sub-example, the second feature set directly indicates Q3.

[0712] As a reference embodiment of the above sub-example, the second feature set indicates the difference between Q3 and L3.

[0713] As a reference embodiment of the above sub-example, Q3 is greater than L3, or Q3 is less than L3.

[0714] As a sub-implementation of the above embodiments, the partial features include a maximum data rate, the third feature set indicates that the maximum data rate is L4 Mbps, and the second feature set indicates that the supported bandwidth is updated to Q4 Mbps.

[0715] As a reference embodiment of the above sub-example, the second feature set directly indicates Q4.

[0716] As a reference embodiment of the above sub-example, the second feature set indicates the difference between Q4 and L4.

[0717] As a reference embodiment of the above sub-example, Q4 is greater than L4, or Q4 is less than L4.

[0718] As a sub-implementation of the above embodiments, the partial features include a maximum modulation order, the third feature set indicates that the maximum modulation order is a first modulation order, and the second feature set indicates that the maximum modulation order is updated to a second modulation order; the first modulation order and the second modulation order are one of 64QAM, 256QAM or 1024QAM respectively.

[0719] As a reference embodiment of the above sub-example, the second feature set directly indicates the second modulation order.

[0720] As an example, the L1 feature sets depend on the L0 feature sets, wherein any feature set in the L1 feature sets indicates that at least a portion of the features in one of the feature sets in the L0 feature sets are adjusted.

[0721] As an example, adjusting a feature includes increasing the maximum value or support value of the indicator corresponding to the feature.

[0722] As an example, adjusting a feature includes reducing the maximum value or support value of the indicator corresponding to the feature.

[0723] As an example, the second feature set in the L1 feature set indicates that some features in the third feature set in the L0 feature set are adjusted.

[0724] As a sub-implementation of the above embodiment, the partial features include a maximum number of layers, the third feature set indicates that the maximum number of layers is L1, the second feature set indicates that the maximum number of layers is adjusted by K1, and the adjusted maximum number of layers is (L1+K1).

[0725] As a reference embodiment of the above sub-example, K1 is greater than 0, or K1 is less than 0.

[0726] As a sub-implementation of the above embodiments, the partial features include the maximum number of downlink or uplink RS resources, the third feature set indicates that the maximum number of downlink or uplink RS resources is L2, the second feature set indicates that the maximum number of downlink or uplink RS resources is adjusted by K2, and the adjusted maximum number of downlink or uplink RS resources is (L2+K2).

[0727] As a reference embodiment of the above sub-example, K2 is greater than 0, or K2 is less than 0.

[0728] As a sub-example of the above embodiment, the partial features include a supported bandwidth, the third feature set indicates that the supported bandwidth is L3 MHz, the second feature set indicates that the supported bandwidth is adjusted to K3 MHz, and the adjusted supported bandwidth is (L3+K3)MHz.

[0729] As a reference embodiment of the above sub-example, K3 is greater than 0, or K3 is less than 0.

[0730] As a sub-implementation of the above embodiment, the partial features include a maximum data rate, the third feature set indicates that the maximum data rate is L4 Mbps, the second feature set indicates that the maximum data rate is adjusted to K4 Mbps, and the adjusted maximum data rate is (L4+K4) Mbps.

[0731] As a reference embodiment of the above sub-example, K4 is greater than 0, or K4 is less than 0.

[0732] As an example, the second feature set in the L1 feature set includes M1 features, the third feature set in the L0 feature set includes M2 features, the index corresponding to any feature in the M1 features is the index corresponding to one feature in the M2 features, the second feature set is conditional on M3 features, the M3 features are a subset of the M2 features, and the index corresponding to any feature in the M3 features is different from the index corresponding to any feature in the second feature set.

[0733] As a sub-implementation of the above embodiment, M2 is greater than M1.

[0734] As a sub-example of the above embodiment, M2 is equal to the sum of M1 and M3.

[0735] As a sub-example of the above embodiment, the M1 features are conditional upon the M3 features.

[0736] As a sub-example of the above embodiment, the M1 features are supported only if the M3 features are supported.

[0737] As a sub-implementation of the above embodiment, the first node supports the combination of the M3 features and the M1 features.

[0738] As a sub-implementation of the above embodiments, the M3 features and the M1 features are supported as a whole.

[0739] As a sub-implementation of the above embodiment, the second feature set is any feature set in the L1 feature sets.

[0740] As an example, the second feature set in the L1 feature sets indicates that some features in the third feature set in the L0 feature sets are updated or adjusted. The updated or adjusted features are conditional on another set of features in the third feature set, which are other features in the third feature set besides the aforementioned features.

[0741] As a sub-implementation of the above embodiments, the updated or adjusted partial features are supported only if the other partial features are supported.

[0742] As a sub-example of the above embodiments, the first node supports a combination of the other partial features and the updated or adjusted partial features.

[0743] As a sub-implementation of the above embodiments, the other part of the features and the updated or adjusted part of the features are supported as a whole.

[0744] As a sub-implementation of the above embodiment, the second feature set is any feature set in the L1 feature sets.

[0745] The advantages of the above method include that, considering the mutual constraints between different indicators, it facilitates global optimization of resource allocation on the network side.

[0746] As an example, the L1 feature sets depend on the L0 feature sets in that the maximum value or support value of any indicator indicated by any feature set in the L1 feature sets does not exceed the maximum value or support value of the same indicator indicated by the L0 feature sets.

[0747] As an example, for each feature set in the L1 feature sets, there exists a corresponding feature set in the L0 feature sets; the maximum value or support value of each indicator indicated by any feature set in the L1 feature sets does not exceed the maximum value or support value of the same indicator indicated by the corresponding feature set.

[0748] As an example, the second report includes the number of training datasets.

[0749] The benefits of the above methods include that more information reported helps the network side to better optimize resource allocation, especially inference-related resource allocation.

[0750] The advantages of the above method include ensuring that there is no ambiguity in the understanding of the L0 feature sets by the network side and the first node.

[0751] As an example, the second report includes the maximum number of training datasets.

[0752] As an example, the second report includes a maximum number of training datasets, with the L0 feature sets conditioned on the maximum number of training datasets.

[0753] As an example, the first node supports the L0 feature sets provided that the number of training datasets does not exceed the maximum number of training datasets.

[0754] As an example, when the number of training datasets exceeds the maximum number of training datasets, it cannot be assumed that the L0 feature sets can still be supported.

[0755] As an example, the second report includes the minimum number of training datasets.

[0756] As an example, the second report includes a minimum number of training datasets, wherein the L0 feature sets are conditioned on the minimum number of training datasets.

[0757] As an example, the first node supports the L0 feature sets provided that the number of training datasets is not less than the minimum number of training datasets.

[0758] As an example, when the number of training datasets is less than the minimum number of training datasets, the maximum value or support value of the indicator corresponding to a feature in the L0 feature sets exceeds the maximum value or support value of the corresponding indicator indicated by the L0 feature sets.

[0759] Example 8

[0760] Example 8 illustrates a schematic diagram of L1 feature sets and L0 feature sets according to an embodiment of this application; as shown in Figure 8. In Example 8, any feature set in the L1 feature sets is a feature set in the L0 feature sets. In Figure 8, the L0 feature sets are represented as feature set #0, ..., feature set #(L0-1); the diagonally filled boxes represent feature sets in the L1 feature sets. In Figure 8, i and j are positive integers less than L0-1, and i is not equal to j.

[0761] As an example, the first report indicates the L1 feature set from the L0 feature set.

[0762] As an example, the first report indicates a set of features that are no longer supported from the L0 feature set, and the L1 feature set is the feature set in the L0 feature set excluding the no longer supported feature set.

[0763] In Figure 8, the no longer supported feature sets include feature set #0 and feature set #j.

[0764] As an example, in the L0 feature set, only the L1 feature set is still supported.

[0765] As an example, any feature set other than the L1 feature set in the L0 feature set is no longer supported.

[0766] Example 9

[0767] Example 9 illustrates a schematic diagram of L1 feature sets and L0 feature sets according to an embodiment of this application; as shown in Figure 9. In Example 9, any feature set in the L1 feature sets indicates that at least some features in one feature set of the L0 feature sets are updated or adjusted. In Figure 9, the L0 feature sets are represented as feature set #0, ..., feature set #(L0-1); the diagonally filled boxes represent feature sets updated by the L1 feature sets; i and j are positive integers less than L0-1 and i is not equal to j.

[0768] In Figure 9(a), at least some features in feature set #i of the L0 feature sets are updated by a first given feature set of the L1 feature sets, and at least some features in feature set #(L0-1) of the L0 feature sets are updated by a second given feature set of the L1 feature sets.

[0769] For example, the feature set #i indicates that the maximum number of layers is L1, and the first given feature set indicates that the maximum number of layers is updated to Q1.

[0770] For example, the feature set #(L0-1) indicates that the maximum modulation order is the first modulation order, and the second given feature set indicates that the maximum modulation order is updated to the second modulation order; the first modulation order and the second modulation order are one of 64QAM, 256QAM or 1024QAM.

[0771] As an example, any feature set in the L1 feature sets indicates the updated features.

[0772] In Figure 9(b), at least some features in feature set #i of the L0 feature sets are adjusted by a first given feature set of the L1 feature sets, and at least some features in feature set #(L0-1) of the L0 feature sets are adjusted by a second given feature set of the L1 feature sets.

[0773] For example, the feature set #i indicates that the maximum number of layers is L1, and the first given feature set indicates that the maximum number of layers is adjusted by K1, and the adjusted maximum number of layers is (L1+K1).

[0774] For example, the feature set #(L0-1) indicates a maximum data rate of L4 Mbps, and the second given feature set indicates that the maximum data rate is adjusted to K4 Mbps, and the adjusted maximum data rate is (L4+K4) Mbps.

[0775] As an example, any feature set in the L1 feature sets indicates the adjustment amount.

[0776] As an example, the index corresponding to any feature in any feature set of the L1 feature sets is the index corresponding to a feature in one feature set of the L0 feature sets.

[0777] As an example, any feature set in the L1 feature sets corresponds to a feature set in the L0 feature sets, and the index corresponding to any feature in any feature set in the L1 feature sets is the index corresponding to a feature in the corresponding feature set.

[0778] For example, in Figure 9, the index corresponding to feature a1 in feature set #i is the number of layers, and the index corresponding to features (features c1 and c2) in the first given feature set is also the number of layers; in Figure 9(a), the index corresponding to feature b2 in feature set #(L0-1) is the modulation order, and the index corresponding to feature d1 in the second given feature set is also the modulation order; in Figure 9(b), the index corresponding to feature b1 in feature set #(L0-1) is the data rate, and the index corresponding to feature d2 in the second given feature set is also the data rate.

[0779] As an example, any feature set in the L1 feature sets corresponds to a feature set in the L0 feature sets, and any feature set in the L1 feature sets indicates that some features in the corresponding feature sets are updated or adjusted; the updated or adjusted features indicated by any feature set in the L1 feature sets are conditional on features in the corresponding feature sets that have not been updated or adjusted.

[0780] For example, in Figure 9(a), the first node supports updating the maximum number of layers to Q1 under the condition that the maximum number of downlink RS resources is L2; ​​or, the first node supports a combination of (the maximum number of downlink RS resources is L2 and the maximum number of layers is Q1).

[0781] For example, in Figure 9(a), the first node supports updating the maximum modulation order to the second modulation order under the condition that the maximum data rate is L4 Mpbs; or, the first node supports a combination of (the maximum data rate is L4 Mpbs and the maximum modulation order is the second modulation order).

[0782] For example, in Figure 9(b), the first node supports adjusting the maximum number of layers to K1 under the condition that the maximum number of downlink RS resources is L2; ​​or, the first node supports a combination of (the maximum number of downlink RS resources is L2, and the maximum number of layers is (L1+K1)).

[0783] For example, in Figure 9(b), the first node supports the maximum data rate being adjusted to K4 Mbps under the condition that the maximum modulation order is the first modulation order; or, the first node supports a combination of (the maximum modulation order is the first modulation order and the maximum data rate is (L4+K4) Mbps).

[0784] As an example, for each feature set in the L1 feature sets, there exists a corresponding feature set in the L0 feature sets; the maximum value or support value of each indicator indicated by any feature set in the L1 feature sets does not exceed the maximum value or support value of the same indicator indicated by the corresponding feature set.

[0785] For example, in Figure 9(a), Q1 is not greater than L1.

[0786] For example, in Figure 9(a), Q1 is smaller than L1.

[0787] For example, in Figure 9(a), the first modulation order is 256QAM and the second modulation order is 64QAM, or the first modulation order is 1024QAM and the second modulation order is 64QAM or 256QAM.

[0788] For example, in Figure 9(b), K1 is not greater than 0.

[0789] For example, in Figure 9(b), K1 is less than 0.

[0790] For example, in Figure 9(a), K4 is not greater than 0.

[0791] For example, in Figure 9(a), K4 is less than 0.

[0792] Example 10

[0793] Example 10 illustrates a schematic diagram of a second signaling, a first configuration information, and L0 feature sets according to an embodiment of this application, as shown in Figure 10. In Example 10, the second signaling indicates the first configuration information; the L0 feature sets depend on the indication of the second signaling.

[0794] In a preferred embodiment, the second signaling is received before the first signaling.

[0795] In a preferred embodiment, the second signaling is broadcast or multicast.

[0796] As one example, the second signaling is carried by SS / PBCH.

[0797] As one embodiment, the second signaling includes system information.

[0798] As one embodiment, the second signaling is carried by the MIB (Master Information Block).

[0799] As one example, the second signaling is carried by the SIB (System Information Block).

[0800] As one embodiment, the second signaling is carried by the MIB or SIB.

[0801] As one example, the second signaling is jointly carried by the MIB and SIB.

[0802] As one embodiment, the second signaling includes the first configuration information.

[0803] As an example, the second signaling explicitly indicates the first configuration information.

[0804] As one example, the second signaling implicitly indicates the first configuration information.

[0805] As an example, the second signaling directly indicates the first configuration information.

[0806] As one embodiment, the second signaling indicates the first configuration information by indicating other information.

[0807] As one embodiment, the second signaling directly indicates a portion of the first configuration information, and indicates another portion of the first configuration information by indicating other information.

[0808] As one embodiment, the first configuration information includes higher layer signaling.

[0809] As one example, the first configuration information includes RRC signaling.

[0810] As one embodiment, the first configuration information includes some or all of the information of each of one or more RRC IEs.

[0811] As one example, the first configuration information includes some or all of the information in the ServingCellConfigCommonSIB IE.

[0812] As an example, the first configuration information includes at least one of DownlinkConfigCommonSIB IE and UplinkConfigCommonSIB.

[0813] As an example, the first configuration information is used to configure a reasoning.

[0814] As an example, the first configuration information includes inference-related parameters.

[0815] As an example, the first configuration information is used to determine at least one of the three factors: the input, the output, and the model for inference.

[0816] As an example, the first configuration information is used to determine at least one of the input, output, and model for each of the multiple inferences.

[0817] As an example, the first configuration information includes inference-related parameters.

[0818] As an example, the first configuration information includes inference-related parameters for an inference.

[0819] As an example, the first configuration information includes inference-related parameters for each of the multiple inferences.

[0820] As an example, the first configuration information is not used for configuration inference.

[0821] As an example, the first configuration information includes multiple metrics, which include at least one of the following: number of layers, maximum number of layers, number of downlink RS resources, number of uplink RS resources, bandwidth, data rate, and modulation order.

[0822] As one example, the second signaling indicates the association between the feature set and the training dataset.

[0823] As one example, the second signaling indicates whether the feature set depends on the training dataset.

[0824] As an example, the second signaling indicates whether the feature set is conditioned on a certain training dataset size.

[0825] As one embodiment, the second signaling indicates the association between the feature set indicated by the second report and the training dataset.

[0826] As one embodiment, the second signaling indicates whether the feature set indicated by the second report depends on the training dataset.

[0827] As an example, the second signaling indicates whether the feature set indicated by the second report is conditioned on a certain training dataset size.

[0828] As one example, the second signaling indicates the number of training datasets.

[0829] As one embodiment, the second signaling indicates a training dataset size, and the feature set indicated by the second report is conditioned on the training dataset size.

[0830] As an example, the L0 feature sets depending on the indication of the second signaling include the L0 feature sets being conditional on the indication of the second signaling.

[0831] As an example, the L0 feature sets depending on the indication of the second signaling include the fact that the L0 feature sets are obtained under the conditions indicated by the second signaling.

[0832] As an example, the L0 feature sets depending on the indication of the second signaling include the first node supporting the L0 feature sets under the indication of the second signaling.

[0833] As an example, the indication that the L0 feature sets depend on the second signaling includes that the L0 feature sets are conditional on the first configuration information.

[0834] As an example, the indication that the L0 feature sets depend on the second signaling includes that the L0 feature sets are obtained under the conditions of the first configuration information.

[0835] As an example, the indication that the L0 feature sets depend on the second signaling includes that, under the conditions of the first configuration information, the first node supports the L0 feature sets.

[0836] As an example, the first configuration information includes multiple indicators, and the first node supports the L0 feature sets if it supports the multiple indicators indicated by the first configuration information.

[0837] As an example, when the indicators configured for the first node change compared to the indications of the second signaling, it cannot be assumed that the L0 feature sets will still be supported.

[0838] As one embodiment, the second signaling indicates the number of the first training dataset; the L0 feature sets depend on the indication of the second signaling including supporting the L0 feature sets provided that the number of training datasets does not exceed the number of the first training datasets.

[0839] As an example, when the number of training datasets exceeds the number of the first training datasets, it cannot be assumed that the L0 feature sets can still be supported.

[0840] Example 11

[0841] Example 11 illustrates a schematic diagram of L0 feature sets according to an embodiment of the present application, conditioned on a first assumption; as shown in Figure 11.

[0842] In Figure 11(a), the first assumption includes that any two inference configurations cannot be associated with the same training dataset.

[0843] As an example, the first assumption includes that any two inference configurations configured for the first node cannot be associated with the same training dataset.

[0844] In Figure 11(b), the first assumption includes that at most W1 inference configurations are associated with the same training dataset, wherein W1 is predefined or configurable.

[0845] As an example, the first assumption includes that the number of inference configurations associated with the same training dataset configured for the first node does not exceed W1, where W1 is predefined or configurable.

[0846] In Figure 11(c), the first assumption includes that up to W2 inference configurations are associated with different training datasets, wherein W2 is predefined or configurable.

[0847] As an example, the first assumption includes that the number of inference configurations associated with different training datasets configured for the first node does not exceed W2, where W2 is predefined or configurable.

[0848] As an example, the first node supports the L0 feature sets only if the first assumption is true.

[0849] As an example, when the first assumption is not true, it cannot be assumed that the L0 feature sets are still supported.

[0850] As an example, an inference configuration associated with a training dataset means that the training dataset is used for training the model of the inference.

[0851] As an example, an inference configuration associated with a training dataset means that the training dataset of the inference model includes the training dataset.

[0852] As an example, an inference configuration associated with a training dataset means that the inference and the training dataset are associated with the same association identifier.

[0853] As an example, an inference configuration associated with a training dataset means that the inference model and the training dataset are associated with the same association identifier.

[0854] For an example of the associated identifier, please refer to Example 1.

[0855] Example 12

[0856] Example 12 illustrates a schematic diagram of L0 feature sets conditioned on a second assumption according to an embodiment of this application; as shown in Figure 12. In Example 12, the second assumption includes not employing reasoning.

[0857] As an example, the first node supports the L0 feature sets only if the second assumption is true.

[0858] As an example, when the second assumption is not true, it cannot be assumed that the L0 feature sets are still supported.

[0859] As an example, without employing reasoning, the first node supports the L0 feature sets.

[0860] As an example, when the first node uses inference, it cannot be assumed that the L0 feature sets are still supported.

[0861] As an example, the second assumption includes that any of the metrics included in the first configuration information are independent of reasoning.

[0862] As an example, the second assumption includes that any metric configured for the first node is independent of inference.

[0863] As an example, an indicator that does not depend on reasoning means that the completion or attainment of the indicator does not depend on reasoning.

[0864] As an example, one metric is the number of layers, and the inference-independent metric includes MIMO transmission and reception being inference-independent.

[0865] As an example, one metric is the number of downlink or uplink RS resources, and the inference-independent metric includes CSI calculation, generation, and reporting that are inference-independent.

[0866] As an example, one metric is bandwidth, and the inference-independent metric includes that receiving and transmitting are inference-independent.

[0867] As an example, one metric is the data rate, and the inference-independent metric includes data reception and data transmission being inference-independent.

[0868] As an example, one metric is the modulation order, and the inference-independent metric includes modulation and demodulation that are inference-independent.

[0869] Example 13

[0870] Example 13 illustrates a schematic diagram of a first report being triggered by an event in a first event set according to an embodiment of this application; as shown in Figure 13.

[0871] As an example, the first event set includes at least one event.

[0872] As an example, the first report is triggered by any event in the first event set.

[0873] As an example, the first report is triggered when any event in the first event set occurs.

[0874] As an example, the first report is triggered in response to any event occurring in the first event set.

[0875] As an example, the first report is triggered when any event in the first event set occurs.

[0876] As an example, the first report is triggered if any event in the first event set occurs.

[0877] As one embodiment, the first event set includes at least one state change of the inference configuration configured for the first node.

[0878] As an example, the first event set includes at least one inference configuration being activated.

[0879] As an example, the first event set includes at least one inference configuration being deactivated.

[0880] As one embodiment, the first event set includes at least one inference configuration being configured.

[0881] As one embodiment, the first event set includes at least one inference configuration being reconfigured.

[0882] As one embodiment, the first event set includes at least one inference configuration being activated, deactivated, configured, or reconfigured.

[0883] As one example, the state change includes changing from unavailable to available and changing from available to unavailable.

[0884] As one embodiment, the first event set includes at least one inference configuration changing from unavailable to available or from available to unavailable.

[0885] As one embodiment, the first event set includes at least one inference configuration being activated, deactivated, configured, reconfigured, changed from unavailable to available, or changed from available to unavailable.

[0886] As an example, the availability of an inference configuration means that the inference configured by the inference configuration is available.

[0887] As an example, the availability of an inference configuration means that the first node is ready to apply the inference configured by the inference configuration.

[0888] As an example, the availability of an inference configuration means that the training of the inference model configured by the inference configuration has been completed.

[0889] As an example, the availability of an inference configuration means that the inference configuration can be activated or executed.

[0890] As an example, the availability of an inference configuration means that the inference model configured by the inference configuration is available.

[0891] As an example, the availability of an inference configuration means that the inference model configured by the inference configuration can be activated.

[0892] As an example, "inference configuration unavailable" means that the inference configured by the inference configuration is unavailable.

[0893] As an example, an inference configuration being unavailable means that the first node is not ready to use the inference configured by the inference configuration.

[0894] As an example, an unavailable inference configuration means that the training of the inference model configured by the inference configuration has not yet been completed.

[0895] As an example, an unavailable inference configuration means that the inference configuration is not yet ready to be activated or executed.

[0896] As an example, "inference configuration unavailable" means that the inference model configured by the inference configuration is unavailable.

[0897] As an example, an unavailable inference configuration means that the inference model configured by the inference configuration is not yet ready to be activated.

[0898] As an example, the first event set also includes at least one model being deployed or redeployed.

[0899] As a sub-example of the above embodiments, the model refers to an AI model or an ML model.

[0900] As one embodiment, the first event set also includes the expiration of a first timer.

[0901] As an example, the first timer is configurable.

[0902] As an example, the first timer depends on the configuration of higher-level parameters.

[0903] As an example, the first timer is configured by RRC signaling.

[0904] As one example, the first timer depends on the capabilities of the first node.

[0905] As an example, the first timer is reported by the first node through the UE capability IE.

[0906] As an example, the first timer is started or restarted after the first node reports the supported feature set.

[0907] As an example, the first event set includes at least one of a first event, a second event, and a third event. The first event includes at least one inference configuration being activated, deactivated, configured, or reconfigured. The second event includes at least one model being deployed or redeployed. The third event includes a first timer expiring.

[0908] As one embodiment, the first event includes at least one inference configuration being activated, deactivated, configured, reconfigured, changed from unavailable to available, or changed from available to unavailable.

[0909] As an example, the first event set includes the first event, the second event, and the third event.

[0910] As one embodiment, the first event set includes the first event and the third event.

[0911] As one embodiment, the first event set includes the first event and the second event.

[0912] As an example, any one of the at least one inference configurations includes higher-layer signaling.

[0913] As an example, any one of the at least one inference configurations includes RRC signaling.

[0914] As an example, any one of the at least one inference configurations includes one or more RRC IEs.

[0915] As an example, any one of the at least one inference configurations includes some or all of the information of each of one or more RRC IEs.

[0916] As an example, any one of the at least one inference configurations is used to configure an inference.

[0917] As a preferred embodiment, any one of the at least one inference configurations includes inference-related parameters.

[0918] As an example, any one of the at least one inference configurations includes inference-related parameters for the inference configured for that inference configuration.

[0919] As an example, the inference-related parameters include at least one of the following: input-related parameters, output-related parameters, and model parameters.

[0920] Example 14

[0921] Example 14 illustrates a schematic diagram of a third signaling triggering a first report according to an embodiment of this application; as shown in Figure 14.

[0922] As one example, the third signaling includes RRC signaling.

[0923] As an example, the third signaling includes MAC CE.

[0924] As one example, the third signaling includes DCI (Downlink Control Information).

[0925] As an example, the third signaling is DCI.

[0926] As an example, the first report is triggered along with the receipt of the third signaling.

[0927] As an example, the first report is triggered in response to the receipt of the third signaling.

[0928] As an example, the first report is triggered when the third signaling is received.

[0929] As an example, the first report is triggered if the third signaling is received.

[0930] Example 15

[0931] Example 15 illustrates a schematic diagram of an artificial intelligence or machine learning-based processing system according to an embodiment of this application, as shown in Figure 15. In Example 15, the third processor sends a first dataset to the fourth processor and a second dataset to the fifth processor; the fourth processor generates a target first-class parameter set based on the first dataset, and sends the generated target first-class parameter set to the fifth processor; the fifth processor processes the second dataset using the target first-class parameter set to obtain a first-class output, and sends the first-class output to the sixth processor. In Figure 15, the first-class feedback and the second-class feedback are optional; the fourth processor includes ML training functionality; the fifth processor includes ML inference functionality.

[0932] As one embodiment, the sixth processor includes ML inference capabilities.

[0933] As one embodiment, the sixth processor includes ML testing functionality.

[0934] As one example, the sixth processor includes performance monitoring / evaluation of the ML model.

[0935] As one embodiment, the sixth processor includes the inverse operation of the fifth processor.

[0936] As one embodiment, the sixth processor includes ML testing functionality.

[0937] As one example, the sixth processor includes performance monitoring / evaluation of the ML model.

[0938] As an example, the fifth processor sends a first type of feedback to the fourth processor. The first type of feedback is used to trigger the recalculation or update of the target first type of parameter set, that is, to trigger ML initial training or ML retraining.

[0939] As one embodiment, the sixth processor sends a second type of feedback to the third processor, the second type of feedback being used to generate the first dataset or the second dataset, or the second type of feedback being used to trigger the sending of the first dataset or the second dataset.

[0940] As one embodiment, the third processor generates at least one of the first dataset and the second dataset based on the measurement of the reference signal.

[0941] As an example, the third processor generates at least one of the first dataset and the second dataset based on measurements of the physical layer channel.

[0942] As one embodiment, the third processor generates at least one of the first dataset and the second dataset based on data from the MAC layer or a layer higher than the MAC layer.

[0943] As one embodiment, the fifth processor is located at the first node, and the sixth processor is located at the second node.

[0944] As an example, the second dataset includes inference data.

[0945] As an example, the second dataset is an inference dataset.

[0946] As an example, the input to an inference belongs to an inference dataset.

[0947] As an example, the second dataset includes measurements for RS.

[0948] As an example, the second dataset includes the reception of PDSCH.

[0949] As an example, the second dataset includes TB or CB.

[0950] As an example, the first dataset includes training data.

[0951] As an example, the first dataset is a training dataset.

[0952] As an example, the fourth processor is used to train an ML model, and the trained model is described by the target first class of parameter sets.

[0953] As one embodiment, the fourth processor is located at the first node.

[0954] The above embodiments save on air interface overhead.

[0955] As one embodiment, the fourth processor is located at the second node.

[0956] The above embodiments support joint training and optimize system performance.

[0957] As one embodiment, the fourth processor is located in the core network.

[0958] The above embodiments support network-wide joint training, further optimizing system performance.

[0959] As an example, the fifth processor constructs a model based on the target first type of parameter group, and then inputs the second dataset into the constructed model to obtain the first type of output.

[0960] As an example, the fifth processor compares the real data with the first type of output, and the resulting error is used to generate the first type of feedback.

[0961] As an example, the fifth processor generates the first type of feedback through performance monitoring.

[0962] As an example, the first type of feedback is used to reflect the performance of the trained model; when the performance of the trained model fails to meet the requirements, the fourth processing opportunity recalculates the target first type of parameter set.

[0963] As an example, the sixth processor compares the real data with the first type of output, and the resulting error is used to generate the second type of feedback.

[0964] As an example, the sixth processor generates the second type of feedback through performance monitoring.

[0965] As an example, the second type of feedback is used to reflect the performance of the trained model; when the performance of the trained model fails to meet the requirements, the third processor sends the first dataset to trigger or assist the fourth processor in recalculating the target first type of parameter set.

[0966] As an example, when the error is too large or the update has not been performed for too long, the performance of the trained model is considered to be unsatisfactory.

[0967] As an example, the target first type of parameter group includes one or more of the following: convolution kernel size, number of convolution layers, convolution stride, pooling kernel size, pooling kernel stride, pooling function, activation function, or number of feature maps.

[0968] As an example, the target first type of parameter group includes one or more of the following: convolution kernel, pooling kernel, pooling function, activation function, parameters of pooling function, or parameters of activation function.

[0969] As one example, the ML includes AI.

[0970] As an example, the ML includes ML and AI.

[0971] Example 16

[0972] Example 16 illustrates a schematic diagram based on artificial intelligence or machine learning according to an embodiment of this application; as shown in Figure 16. Figure 16 includes a first operation, a second operation, a third operation, a fourth operation, and a fifth operation. In Example 16, the first and second operations belong to a first stage, the third operation belongs to a second stage, the fourth operation belongs to a third stage, and the fifth operation belongs to a fourth stage. In Figure 16, lines with arrows indicate the sequence of processes.

[0973] As an example, the first operation includes ML training, the second operation includes ML testing, the third operation includes ML emulation, the fourth operation includes ML entity loading, and the fifth operation includes AI inference.

[0974] As one embodiment, the first stage includes a training phase, the second stage includes an emulation phase, the third stage includes a deployment phase, and the fourth stage includes an emulation phase.

[0975] As an example, the first stage includes ML model training.

[0976] As an example, the first stage includes ML model training and ML testing.

[0977] As an example, the ML model training includes initial training and re-training of one or a group of ML models.

[0978] As an example, the training of the ML model depends on training data.

[0979] As an example, the ML model training includes ML entity validation.

[0980] As an example, the ML entity verification is used to evaluate the performance of the ML entity.

[0981] As an example, the ML entity verification depends on verification data.

[0982] As an example, if the results of ML entity verification do not meet expectations, the ML model will be retrained.

[0983] As an example, the ML testing includes testing the validated ML entities to estimate the performance of the trained ML model.

[0984] As an example, if the ML test results meet expectations, the ML entity proceeds to the next stage; otherwise, the ML model will be retrained.

[0985] As an example, the ML test relies on test data.

[0986] As one embodiment, the second stage includes ML simulation, which performs inference of ML entities in a simulation environment.

[0987] As an example, the ML simulation estimates the performance of ML entity reasoning in a simulation environment before using ML entities.

[0988] As one embodiment, the second stage is optional.

[0989] As an example, the third stage includes ML entity loading, which is to obtain trained ML entities to obtain the desired AI inference capabilities.

[0990] As an example, the third stage is optional.

[0991] As an example, the third stage is no longer needed when the training and inference functions are co-located.

[0992] As an example, the fourth stage includes AI inference.

[0993] As an example, the AI ​​inference relies on inference data.

[0994] As an example, the input to an AI inference belongs to the inference dataset of the AI ​​inference model.

[0995] As one example, the ML includes AI.

[0996] As one example, the AI ​​includes ML.

[0997] Example 17

[0998] Example 17 illustrates a schematic diagram of AI function deployment according to one embodiment of this application; as shown in Figure 17.

[0999] In Example 17, the AI ​​training function of the RAN (Radio Access Network) domain is located in the RAN domain-specific management function, while the AI ​​inference function is located in the UE.

[1000] In Example 17, RAN domain-specific management functions provide AI training function management capabilities and AI inference function management capabilities.

[1001] Example 18

[1002] Example 18 illustrates a schematic diagram of AI function deployment according to one embodiment of this application; as shown in Figure 18.

[1003] In Example 18, the AI ​​training function is a management function specific to the RAN domain, while the AI ​​inference function is located locally on the UE.

[1004] In Example 18, the management capability of the AI ​​training function is provided by the RAN domain-specific management function, while the management capability of the AI ​​inference function is provided locally by the UE.

[1005] In Figure 18, MnF refers to Management Function.

[1006] Example 19

[1007] Example 19 illustrates a schematic diagram of AI function deployment according to one embodiment of this application; as shown in Figure 19.

[1008] In Example 19, both the AI ​​training function and the AI ​​inference function are located in the UE, wherein the UE provides the ability to train and infer.

[1009] In Example 19, RAN domain-specific management functions provide management capabilities for both AI training and AI inference functions.

[1010] Example 20

[1011] Example 20 illustrates a schematic diagram of AI function deployment according to one embodiment of this application; as shown in Figure 20.

[1012] In Example 20, both the AI ​​training function and the AI ​​inference function are located in the UE.

[1013] In Example 20, the management capabilities of both the AI ​​training function and the AI ​​inference function are provided locally by the UE.

[1014] In Figure 20, MnF refers to Management Function.

[1015] Example 21

[1016] Example 21 illustrates a structural block diagram of a processing apparatus for a first node according to an embodiment of this application; as shown in Figure 21. In Figure 21, the processing apparatus 2100 in the first node includes a first receiver 2101 and a first transmitter 2102.

[1017] In embodiment 21, the first receiver 2101 receives the first signaling and the first transmitter 2102 sends the first report.

[1018] In embodiment 21, the first signaling includes a first inference configuration; the first report indicates support for L1 feature sets, where L1 is a positive integer; the L1 feature sets depend on the indication of the first signaling; any feature set in the L1 feature sets includes at least one of the following: maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, maximum data rate, and maximum modulation order; at least one indicator included in the first inference configuration corresponds to a feature that does not belong to the first feature set, and the first feature set is a feature set in the L1 feature sets.

[1019] As an example, the first report is carried by MAC CE.

[1020] As one embodiment, the first transmitter 2102 sends a second report indicating support for L0 feature sets, where L0 is a positive integer greater than 1; wherein the second report is sent before the first report, and the L1 feature sets depend on the L0 feature sets.

[1021] As a sub-implementation of the above embodiments, the second report is sent earlier than the first signaling is received.

[1022] As one embodiment, the first receiver 2101 receives a second signaling, the second signaling indicating first configuration information; wherein the L0 feature sets depend on the indication of the second signaling.

[1023] As an example, the L0 feature sets are conditional on a first hypothesis; the first hypothesis includes the association between the inference configuration and the training dataset.

[1024] As an example, the L0 feature sets are conditional on a second hypothesis; the second hypothesis includes not using reasoning.

[1025] As an example, the first report is triggered by an event in a first event set, the first event set including at least one state change of inference configuration, the state change including one or more of being activated, deactivated, configured and reconfigured.

[1026] As an example, the first receiver 2101 receives a third signaling; wherein the third signaling triggers the first report.

[1027] As one embodiment, the first node includes a terminal.

[1028] As one embodiment, the first node includes a user equipment.

[1029] As one embodiment, the first node includes a relay node device.

[1030] As an example, the first receiver 2101 includes at least one of the following in embodiment 4: {antenna 452, receiver 454, receiver processor 456, multi-antenna receiver processor 458, controller / processor 459, memory 460, data source 467}.

[1031] As an example, the first transmitter 2102 includes at least one of the following in embodiment 4: {antenna 452, transmitter 454, transmission processor 468, multi-antenna transmission processor 457, controller / processor 459, memory 460, data source 467}.

[1032] Example 22

[1033] Example 22 illustrates a structural block diagram of a processing apparatus for a second node according to an embodiment of the present application; as shown in Figure 22. In Figure 22, the processing apparatus 2200 in the second node includes a second transmitter 2201 and a second receiver 2202.

[1034] In embodiment 22, the second transmitter 2201 sends a first signaling; the second receiver 2202 receives a first report.

[1035] In embodiment 22, the first signaling includes a first inference configuration; the first report indicates support for L1 feature sets, where L1 is a positive integer; the L1 feature sets depend on the indication of the first signaling; any feature set in the L1 feature sets includes at least one of the following: maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, maximum data rate, and maximum modulation order; at least one indicator included in the first inference configuration corresponds to a feature that does not belong to the first feature set, and the first feature set is a feature set in the L1 feature sets.

[1036] As an example, the first report is carried by MAC CE.

[1037] As one embodiment, the second receiver 2202 receives a second report indicating support for L0 feature sets, where L0 is a positive integer greater than 1; wherein the second report is received before the first report, and the L1 feature sets depend on the L0 feature sets.

[1038] As a sub-implementation of the above embodiments, the second report is sent earlier than the first signaling is received.

[1039] As one embodiment, the second transmitter 2201 sends a second signaling, the second signaling indicating first configuration information; wherein the L0 feature sets depend on the indication of the second signaling.

[1040] As an example, the L0 feature sets are conditional on a first hypothesis; the first hypothesis includes the association between the inference configuration and the training dataset.

[1041] As an example, the L0 feature sets are conditional on a second hypothesis; the second hypothesis includes not using reasoning.

[1042] As an example, the first report is triggered by an event in a first event set, the first event set including at least one state change of inference configuration, the state change including one or more of being activated, deactivated, configured and reconfigured.

[1043] As one embodiment, the second transmitter 2201 sends a third signaling; wherein the third signaling triggers the first report.

[1044] As one embodiment, the second node includes a base station.

[1045] As one embodiment, the second node includes a base station device.

[1046] As one embodiment, the second node includes a relay node device.

[1047] As one embodiment, the second node includes the sustaining base station of the serving cell of the first node.

[1048] As one embodiment, the second node includes an OTT (Over-The-Top) server.

[1049] As an example, the second node provides OAM (Operation Administration and Maintenance).

[1050] As one embodiment, the second node includes a NAS (Network Access Server).

[1051] As one embodiment, the second node includes a NAS device.

[1052] As one example, the second node provides network access services.

[1053] As one embodiment, the second node includes core network equipment.

[1054] As one embodiment, the second node includes base station equipment and core network equipment.

[1055] As one embodiment, the second node includes a base station device and a NAS device.

[1056] As one embodiment, the second transmitter 2201 includes at least one of the following in embodiment 4: {antenna 420, transmitter 418, transmission processor 416, multi-antenna transmission processor 471, controller / processor 475, memory 476}.

[1057] As one embodiment, the second receiver 2202 includes at least one of the following in embodiment 4: {antenna 420, transmitter / receiver 418, transmitter processor 416, receiver processor 470, multi-antenna transmitter processor 471, multi-antenna receiver processor 472, controller / processor 475, memory 476}.

[1058] Example 23

[1059] Example 23 illustrates a schematic diagram of a first signaling, L1 feature sets, and a first inference configuration according to an embodiment of this application, as shown in Figure 23. In Example 23, any feature set in the L1 feature sets includes at least one of the following: maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, maximum data rate, and maximum modulation order; the L1 feature sets depend on the indication of the first signaling, which includes the first inference configuration, and the feature corresponding to at least one indicator included in the first inference configuration does not belong to a feature set in the L1 feature sets. In Figure 23, the L1 feature sets are respectively represented as feature set #0, ..., feature set #(L1-1).

[1060] In Figure 23, feature set #0 includes a feature of maximum number of layers, and the indicator corresponding to the feature of feature set #0 is the number of layers; feature set #1 includes a feature of maximum data rate, and the indicator corresponding to the feature of feature set #1 is the data rate; feature set #(L1-1) includes a feature of maximum number of downlink RS resources, and the indicator corresponding to the feature of feature set #(L1-1) is the number of downlink RS resources; the two indicators included in the first inference configuration are modulation order and bandwidth.

[1061] Example 24

[1062] Example 1 illustrates a flowchart of a first report according to an embodiment of this application, as shown in Figure 24. In Figure 24, each block represents a step.

[1063] In embodiment 24, the first node sends a first report in step 2401. The first report indicates that the idle data rate capability is a first data rate, and the idle data rate capability is for a first inference configuration.

[1064] As an example, the first report is carried by higher layer signaling.

[1065] As an example, the first report is carried by RRC (Radio Resource Control) signaling.

[1066] As an example, the first report is carried by an RRC IE (Information Element).

[1067] As an example, the first report is carried by the UE capability IE.

[1068] As an example, the first report is carried by MAC CE (Medium Access Control layer Control Element).

[1069] As an example, the first report is carried by DCI (Downlink Control Information).

[1070] As an example, the first report indicates that the idle data rate capability configured for the first inference is the first data rate.

[1071] As an example, the first report indicates that the idle data rate capability is configured for the first inference.

[1072] As one embodiment, the first inference configuration includes higher layer signaling.

[1073] As an example, the first inference configuration includes RRC signaling.

[1074] As one example, the first inference configuration includes one or more RRC IEs.

[1075] As one embodiment, the first inference configuration includes some or all of the information of each of one or more RRC IEs.

[1076] As an example, the first inference configuration includes at least one of PDSCH-Config IE, PUSCH-Config IE, SPS-Config IE, PDSCH-ServingCellConfig IE, PUSCH-ServingCellConfig IE, and ConfiguredGrantConfig IE.

[1077] As an example, the first inference configuration includes a MAC CE.

[1078] As an example, the first inference configuration includes DCI.

[1079] As an example, the first inference configuration is carried by both RRC signaling and MAC CE.

[1080] As an example, the first inference configuration is carried by both RRC signaling and DCI.

[1081] In a preferred embodiment, the first inference configuration is used to configure inference.

[1082] As an example, the first inference configuration includes inference-related parameters.

[1083] As an example, the inference refers to AI (Artificial Intelligence) inference.

[1084] As an example, the inference refers to ML (Machine Learning) inference.

[1085] As an example, the inference refers to AI inference or ML inference.

[1086] As an example, the first inference configuration is used to determine at least one of the inference input, output, and model.

[1087] As an example, the first inference configuration is used to determine at least one of the inference input and the model.

[1088] As an example, the first inference configuration is used to determine at least one of the inference output and the model.

[1089] As an example, the first inference configuration is used to determine at least one of the three: the input, the output, and the model of the inference it is configured to use.

[1090] As an example, the first inference configuration includes inference-related parameters.

[1091] As an example, the first inference configuration includes inference-related parameters configured thereon.

[1092] As an example, the inference-related parameters include at least one of the following: input-related parameters, output-related parameters, and model parameters.

[1093] As an example, the inference-related parameters included in the first inference configuration include at least one of the following: input-related parameters, output-related parameters, and model parameters.

[1094] As an example, the inference-related parameters included in the first inference configuration include at least one of input-related parameters and model parameters.

[1095] As an example, the inference-related parameters included in the first inference configuration include input-related parameters and model parameters.

[1096] As an example, the inference-related parameters included in the first inference configuration include at least one of output-related parameters and model parameters.

[1097] As an example, the inference-related parameters included in the first inference configuration include output-related parameters and model parameters.

[1098] In a preferred embodiment, the first inference configuration is used for data reception or data transmission.

[1099] As an example, the inference configured in the first inference configuration is used for data reception or data transmission.

[1100] As an example, the first inference configuration is used for data reception, and at least one of PDSCH and DMRS is used as the input for the inference configured by the first inference configuration, which is used to recover TB or CB.

[1101] As a sub-implementation of the above embodiments, at least one of PDSCH and DMRS is used to generate the input for the inference configured by the first inference configuration.

[1102] As a sub-implementation of the above embodiments, the inference input configured by the first inference configuration includes at least one of PDSCH and DMRS.

[1103] As a reference embodiment of the above sub-example, input including PDSCH means that the input includes symbols received on PDSCH.

[1104] As a sub-example of the above embodiments, the inference input configured by the first inference configuration includes at least one of preprocessed DMRS and preprocessed PDSCH.

[1105] As a sub-implementation of the above embodiments, the output of the inference configured by the first inference configuration includes at least one of the following: channel estimation result, demodulated symbol, and decoded bit.

[1106] As a sub-example of the above embodiments, the output of the inference configured by the first inference configuration includes the recovered TB or CB.

[1107] As a sub-example of the above embodiments, the output of the inference configured by the first inference configuration is used to recover TB or CB.

[1108] As an example, the preprocessing includes one or more of quantization, shortening, puncture, padding, DFT (Discrete Fourier Transform), IDFT (Inverse Discrete Fourier Transform), domain transformation, and multi-antenna reception.

[1109] As an example, the domain transformation includes one or more of the following: angular domain to spatial domain transformation, spatial domain to angular domain transformation, time domain to frequency domain transformation, frequency domain to time domain transformation, delay domain to frequency domain transformation, frequency domain to delay domain transformation, Doppler domain to time domain transformation, and time domain to Doppler domain transformation.

[1110] As an example, the first inference configuration is used for data transmission, TB or CB is used as input to the inference configured by the first inference configuration, and the inference configured by the first inference configuration is used to generate PUSCH.

[1111] As a sub-implementation of the above embodiments, the input of the inference configured by the first inference configuration includes TB or CB.

[1112] As a sub-example of the above embodiments, the input of the inference configured by the first inference configuration includes bits obtained by channel coding of TB or CB.

[1113] As a sub-example of the above embodiments, the input of the inference configured by the first inference configuration includes symbols obtained by channel coding and modulation of TB or CB.

[1114] As a sub-example of the above embodiments, the input of the inference configured by the first inference configuration includes symbols obtained by channel coding, modulation and precoding of TB or CB.

[1115] As a sub-example of the above embodiments, the output of the inference configured by the first inference configuration includes PUSCH.

[1116] As a sub-example of the above embodiments, the output of the inference configured by the first inference configuration includes symbols transmitted on the PUSCH.

[1117] As a sub-example of the above embodiments, the output of the inference configured by the first inference configuration is used to generate PUSCH.

[1118] As a sub-implementation of the above embodiments, the output of the inference configured by the first inference configuration includes at least one of DMRS, PTRS, OFDM symbols, precoded symbols, modulated symbols, and encoded bits.

[1119] As an example, the first inference configuration is used for data reception.

[1120] As an example, the first inference configuration is used for data transmission.

[1121] As one example, the data reception includes the reception of TB or CB.

[1122] As one embodiment, the data reception includes the reception of PDSCH.

[1123] As an example, the data reception refers to the reception of TB or CB.

[1124] As an example, the data reception refers to the reception of the PDSCH.

[1125] As one embodiment, the reception includes one or more of the following: TBS determination, channel estimation, demodulation, multi-antenna reception, channel decoding, and CRC (Cyclic Redundancy Check).

[1126] As one example, the data transmission includes the transmission of TB or CB.

[1127] As an example, the data transmission includes the transmission of PUSCH.

[1128] As an example, the data transmission refers to the transmission of TB or CB.

[1129] As an example, the data transmission refers to the transmission of PUSCH.

[1130] As an example, the transmission includes one or more of the following: TBS determination, DMRS generation, DMRS insertion, modulation, precoding, channel coding, and CRC bit generation.

[1131] As an example, the first inference configuration indicates one or more of the following: resources occupied by the DMRS, an MCS (Modulation and Coding Scheme) table, a maximum modulation order, and a maximum number of layers.

[1132] As an example, the input-related parameters included in the first inference configuration are used to determine the physical layer channel for obtaining the inference dataset.

[1133] As a sub-example of the above embodiments, the physical layer channel includes PDSCH.

[1134] As an example, the first inference configuration includes input-related parameters indicating configuration information for obtaining the physical layer channel of the inference dataset.

[1135] As an example, the configuration information of the physical layer channel includes one or more of the following: time-domain resources, frequency-domain resources, DMRS configuration, MCS table, maximum modulation order, maximum number of layers, and TCI state.

[1136] As an example, the configuration information of the physical layer channel includes a TBS scaling factor, which is used to determine the TBS (Transport Block Size).

[1137] As an example, the TBS scaling factor is used to adjust the TBS obtained based on the method in R18 or earlier versions.

[1138] As an example, the input-related parameters included in the first inference configuration are used to determine at least one of the number of quantized bits and the quantization criterion of the input.

[1139] As an example, the input-related parameters included in the first inference configuration are used to determine the physical layer signals for obtaining the inference dataset.

[1140] As one embodiment, the physical layer signal includes at least one of downlink RS (Reference Signal) and uplink RS.

[1141] As an example, the physical layer signal includes at least one of DMRS (Demodulation Reference Signal) and PTRS (Phase-Tracking Reference Signal).

[1142] As an example, the first inference configuration includes input-related parameters indicating configuration information for obtaining physical layer signals for the inference dataset.

[1143] As an example, the configuration information of the physical layer signal includes one or more of the following: DMRS type, occupied resources, RS sequence, time domain density, frequency domain density, number of ports, CDM type, and EPRE (Energy Per Resource Element).

[1144] As an example, the output-related parameters included in the first inference configuration are used to determine the inference output.

[1145] As an example, the output-related parameters included in the first inference configuration are used to determine the physical layer channel obtained based on inference.

[1146] As a sub-implementation of the above embodiments, the physical layer channel includes PUSCH.

[1147] As an example, the output-related parameters included in the first inference configuration indicate the configuration information of the physical layer channel.

[1148] As an example, the model parameters included in the first inference configuration are used to determine a model.

[1149] As an example, the model parameters included in the first inference configuration include some or all of the parameters used to construct a model.

[1150] As an example, the model parameters included in the first inference configuration indicate the training dataset.

[1151] As an example, the model parameters included in the first inference configuration include an index of the training dataset.

[1152] As an example, the model parameters included in the first inference configuration include a first identifier, which is associated with inference.

[1153] As a sub-implementation of the above embodiments, the first identifier indicates a model.

[1154] As a sub-implementation of the above embodiments, the first identifier indicates a reference model or a nominal model.

[1155] As a sub-implementation of the above embodiments, the first identifier indicates a training dataset.

[1156] As a sub-implementation of the above embodiments, the first identifier indicates a reasoning.

[1157] As a sub-implementation of the above embodiments, the first identifier indicates an inference dataset.

[1158] As a sub-implementation of the above embodiments, the first identifier indicates an AI function or AI entity.

[1159] As a sub-implementation of the above embodiments, the first identifier indicates a performance testing dataset.

[1160] As a sub-implementation of the above embodiments, the first identifier indicates a function.

[1161] As a sub-implementation of the above embodiments, the first identifier is used to identify the first inference configuration.

[1162] As a sub-implementation of the above embodiments, the first identifier indicates the RS used to obtain the training dataset.

[1163] As a sub-implementation of the above embodiments, the first identifier indicates the RS used to obtain the inference dataset.

[1164] As a sub-example of the above embodiments, the first identifier indicates the RS used to obtain the performance monitoring dataset.

[1165] As a sub-example of the above embodiments, the first identifier indicates the characteristics of the RS used to obtain the training dataset, inference dataset, or performance monitoring dataset.

[1166] As a sub-example of the above embodiments, the first identifier indicates the data quality used to obtain the training dataset, inference dataset, or performance monitoring dataset.

[1167] As an example, the characteristics of the RS include one or more of delay spread, Doppler spread, Doppler shift, average delay, or spatial reception parameters.

[1168] As one embodiment, the characteristics of the RS include a downlink transmit beam or a set of downlink transmit beams.

[1169] As an example, the data quality includes one or more of the following: statistical characteristics, LCID, LCG ID (Logical Channel Group ID), logical channel type, MAC subheader, QoS flow, and PDU (Protocol Data Unit) session.

[1170] As an example, the unit of the first data rate is Mbps (megabits per second).

[1171] Considering future ultra-wideband transmission, the unit of the first data rate may also be Gbps or Mbpms (megabits per millisecond).

[1172] As an example, the first data rate is the uplink data rate.

[1173] As an example, the first data rate is the downlink data rate.

[1174] As one embodiment, the first data rate includes an uplink data rate and a downlink data rate.

[1175] As an example, the data rate is for bits in a TB (Transport Block).

[1176] As an example, the data rate depends on the rate of bits in the TB.

[1177] As an example, the data rate is the rate of bits in TB.

[1178] As an example, the data rate is for bits in a TB or CB (Code Block).

[1179] As an example, the data rate is the rate of bits in a TB or CB.

[1180] As a preferred embodiment, the idle data rate capability indicates the additional data rate that can be processed.

[1181] As a preferred embodiment, the idle data rate capability indicates the maximum additional data rate that can be processed.

[1182] In a preferred embodiment, the first inference configuration is used for data reception or data transmission, and the idle data rate capability indicates the additional data rate that can be processed based on the first inference configuration for data reception or data transmission.

[1183] In a preferred embodiment, the first inference configuration is used for data reception or data transmission, and the idle data rate capability indicates the maximum additional data rate that can be processed based on the first inference configuration for data reception or data transmission.

[1184] The benefits of the above methods include better support for inference functions and optimized resource utilization.

[1185] The benefits of the above methods include providing more and more targeted information to the network side, which is conducive to network optimization.

[1186] AI / ML model inference typically consumes significant computational and storage resources. AI / ML models, exemplified by the Transformer architecture, employ techniques such as MoE (Mixed Expert Model) and sparse activation to reduce inference complexity and improve speed. However, with these techniques, the resource consumption for inference depends on the input data; that is, the maximum data rate that inference can process varies depending on the input data. To address this, the aforementioned method proposes reporting the idle data rate capability of the first inference configuration. This allows the network to more accurately understand the resource usage of the first inference configuration, leading to better optimization of network-side scheduling and improved resource utilization of the first node.

[1187] As an example, "additional" refers to anything other than the scheduled data transmission.

[1188] As an example, the scheduled data transmission includes at least one of the following: data transmission based on a configured grant, semi-persistent scheduled data transmission, and dynamically scheduled data transmission.

[1189] As an example, "extra" refers to data transfers other than those being processed.

[1190] As an example, "extra" refers to data transmissions other than those being received or transmitted.

[1191] As an example, "additional" refers to data transmission other than that of the first inference process.

[1192] As an example, "additional" refers to anything other than the first channel.

[1193] As an example, the additional refers to data transmissions processed beyond the first reference time.

[1194] As an example, "additional" refers to data transmissions received or transmitted at the first reference time other than those received or transmitted at the first reference time.

[1195] As one example, the data transmission includes PDSCH (Physical Downlink Shared Channel) transmission.

[1196] As an example, the data transmission includes PUSCH (Physical Uplink Shared Channel) transmission.

[1197] As one example, the data transmission includes PDSCH transmission and PUSCH transmission.

[1198] As one example, the data transmission includes the transmission of TB or CB.

[1199] As one embodiment, data reception or data transmission based on the first inference configuration includes the data reception or data transmission following the instructions of the first inference configuration.

[1200] As one embodiment, data reception or data transmission based on the first inference configuration includes the data reception or data transmission employing inference-related parameters indicated by the first inference configuration.

[1201] As an example, data reception or data transmission based on the first inference configuration includes the data reception or data transmission following indicators indicated by the first inference configuration, the indicators including one or more of DMRS resources, DMRS type, MCS table, maximum modulation order, and maximum number of layers.

[1202] As one embodiment, data reception or data transmission based on the first inference configuration includes the data reception or data transmission depending on inference, the inference depending on the first inference configuration.

[1203] As a sub-implementation of the above embodiments, the data reception includes the reception of PDSCH, and the inference is used to recover the TB or CB carried by the PDSCH.

[1204] As a sub-implementation of the above embodiments, the data transmission includes the transmission of PUSCH, and the inference is used to generate the PUSCH.

[1205] As a sub-implementation of the above embodiments, the inference-related parameters included in the first inference configuration are used to determine at least one of the inference input, output, and model.

[1206] As one embodiment, the idle data rate capability for the first inference configuration includes indicating the maximum additional data rate that can be processed for data reception or data transmission based on the first inference configuration.

[1207] As one embodiment, the idle data rate capability for the first inference configuration includes the maximum additional data rate that can be processed under the conditions of the first inference configuration.

[1208] As one embodiment, the idle data rate capability for a first inference configuration includes the maximum additional data rate that can be processed under the conditions of inference-related parameters indicated by the first inference configuration.

[1209] As an example, the idle data rate capability for a first inference configuration includes the maximum additional data rate that can be processed under the conditions indicated by the first inference configuration; the indicators include one or more of DMRS type, DMRS resources, MCS table, maximum modulation order, and maximum number of layers.

[1210] As an example, if the data rate of the additionally scheduled PDSCH received by the first node does not exceed the first data rate, the additionally scheduled PDSCH reception is processed.

[1211] As an example, if the data rate received by the additionally scheduled PDSCH of the first node exceeds the first data rate, the first node does not process the PDSCH.

[1212] As an example, if the data rate of the PDSCH received by the first node based on the first inference configuration is not more than the first data rate, the PDSCH received by the first node based on the first inference configuration is processed.

[1213] As an example, if the data rate of the PDSCH received by the first node based on the first inference configuration exceeds the first data rate, the first node does not process the PDSCH.

[1214] As one embodiment, the first node not processing PDSCH includes the first node not being expected to process PDSCH.

[1215] As one embodiment, the first node not processing PDSCH includes the first node not being required to process PDSCH.

[1216] As one embodiment, the first node not processing the PDSCH includes the sender of the PDSCH not expecting the first node to process the PDSCH.

[1217] As an example, the first node not processing PDSCH includes the fact that the sender of the PDSCH cannot assume that the first node processes the PDSCH.

[1218] As one embodiment, the first node not processing PDSCH includes the first node determining whether to process PDSCH on its own.

[1219] As one embodiment, the first node not processing PDSCH includes the first node not processing PDSCH reception based on the first inference configuration.

[1220] As one embodiment, the first node not processing PDSCH includes the first node not being expected or required to process PDSCH reception based on the first inference configuration.

[1221] As an example, the first node not processing PDSCH includes the sender of the PDSCH not expecting or assuming that the first node processes PDSCH reception based on the first inference configuration.

[1222] As one embodiment, the first node not processing PDSCH includes the first node determining whether to process PDSCH reception based on the first inference configuration.

[1223] As an example, if the data rate of the additionally scheduled PUSCH transmissions of the first node does not exceed the first data rate, the additionally scheduled PUSCH transmissions are processed.

[1224] As an example, if the data rate of the additionally scheduled PUSCH sent by the first node exceeds the first data rate, the first node does not process the PUSCH.

[1225] As an example, if the data rate of the additionally scheduled PUSCH transmission based on the first inference configuration of the first node does not exceed the first data rate, the additionally scheduled PUSCH transmission based on the first inference configuration is processed.

[1226] As an example, if the data rate of the PUSCH sent by the first node based on the first inference configuration exceeds the first data rate, the first node does not process the PUSCH.

[1227] As one embodiment, the first node not processing PUSCH includes the first node not being expected to process PUSCH.

[1228] As one embodiment, the first node not processing PUSCH includes the first node not being required to process PUSCH.

[1229] As one embodiment, the first node not processing the PUSCH includes the target recipient of the PUSCH not expecting the first node to process the PUSCH.

[1230] As an example, the first node not processing PUSCH includes the following: the target receiver of the PUSCH cannot assume that the first node processes the PUSCH.

[1231] As one embodiment, the first node not processing PUSCH includes the first node determining whether to process PUSCH on its own.

[1232] As one embodiment, the first node not processing PUSCH includes the first node not processing PUSCH transmissions based on the first inference configuration.

[1233] As one embodiment, the first node not processing PUSCH includes the first node not being expected or required to process PUSCH transmissions based on the first inference configuration.

[1234] As an example, the first node not processing PUSCH includes the target recipient of the PUSCH not expecting or assuming that the first node processes PUSCH transmissions based on the first inference configuration.

[1235] As one embodiment, the first node not processing PUSCH includes the first node determining on its own whether to process PUSCH transmission based on the first inference configuration.

[1236] As an example, the data rate of a PDSCH receiving or a PUSCH transmitting depends on the number of bits included in the TB carried by the PDSCH or PUSCH, the total number of CBs and the number of scheduled CBs in the TB carried by the PDSCH or PUSCH, and the numberology of the PDSCH or PUSCH.

[1237] As an example, the data rate of a PDSCH receiving or a PUSCH transmitting depends on the number of bits included in the TB carried by the PDSCH or PUSCH, the total number of CBs and the number of scheduled CBs of the TB carried by the PDSCH or PUSCH, the numberology of the PDSCH or PUSCH, and the number of symbols allocated to the PDSCH or PUSCH.

[1238] As an example, the data rate of a PDSCH receiving or a PUSCH transmitting increases with the increase of the number of bits included in the TB carried by the PDSCH or the PUSCH and the number of scheduled CBs.

[1239] As an example, the data rate of a PDSCH receiving or a PUSCH transmitting increases with the increase of the numberology of the PDSCH or the PUSCH.

[1240] As an example, the data rate of a PDSCH receiving or a PUSCH transmitting decreases as the number of symbols allocated to the PDSCH or PUSCH increases.

[1241] As an example, the data rate of a PDSCH receiving or a PUSCH transmitting is equal to A1 is the number of bits included in the TB carried by the PDSCH or the PUSCH, C1 is the total number of CBs in the TB carried by the PDSCH or the PUSCH, C′1 is the number of scheduled CBs in the TB carried by the PDSCH or the PUSCH, and T1 is the slot length of the numberology for the PDSCH or the PUSCH.

[1242] As an example, the data rate of a PDSCH receiving or a PUSCH transmitting is equal to A1 is the number of bits included in the TB carried by the PDSCH or PUSCH, C1 is the total number of CBs in the TB carried by the PDSCH or PUSCH, C′1 is the number of CBs scheduled for the TB carried by the PDSCH or PUSCH, T1 is the number of slot lengths in the numberology for the PDSCH or PUSCH, and L1 is the number of symbols allocated to the PDSCH or PUSCH.

[1243] As an example, the first node determines the first data rate itself.

[1244] As an example, the first node determines the idle data rate capability to be the first data rate.

[1245] Generally, how the first node determines the first data rate is determined by the hardware equipment manufacturer. Below are some non-limiting implementation methods:

[1246] As an example, the first data rate depends on data quality.

[1247] As an example, the first data rate depends on the quality of the scheduled data.

[1248] As an example, the first data rate depends on the data quality of data reception or data transmission based on the first inference configuration.

[1249] As an example, the data quality includes one or more of the following: statistical characteristics, LCID, LCG ID (Logical Channel Group ID), logical channel type, MAC subheader, QoS flow, and PDU (Protocol Data Unit) session.

[1250] As an example, the first node determines the first data rate based on the data quality of data reception or data transmission based on the first inference configuration.

[1251] As an example, the first node determines the first data rate based on the amount of the first type of resources occupied by the first inference.

[1252] As an example, the first node determines the first data rate based on the input of the first inference.

[1253] As an example, the first node determines the first data rate based on the data quality of the input from the first inference.

[1254] As an example, the higher the quality of the input data for the first inference, the greater the first data rate.

[1255] As an example, the first node determines the first data rate based on one or more of the logical channel type, LCID, LCG ID, MAC subheader, QoS flow, and PDU session of the input data of the first inference.

[1256] As an example, the first node determines the first data rate based on the output of the first inference.

[1257] As an example, the output of the first inference indicates the first data rate.

[1258] As an example, the output of the first inference explicitly indicates the first data rate.

[1259] As an example, the output of the first inference implicitly indicates the first data rate.

[1260] As an example, the output of the first inference is used to indicate the first data rate by indicative of other information.

[1261] As a sub-implementation of the above embodiments, the other information includes one or more of the following: data quality, the amount of first-type resources occupied, the amount of first-type resources required per unit data rate, and the maximum data rate.

[1262] As a sub-implementation of the above embodiments, the other information includes an index that indicates a subset of the training dataset of the model of the first inference.

[1263] As an example, the first node determines the first data rate based on the amount of a first type of resource that is not occupied during the execution of the first inference.

[1264] As an example, the first node determines the first data rate based on the data rate of the first inference process.

[1265] As an example, the higher the data rate of the first inference process, the lower the first data rate.

[1266] As an example, the first node determines the first data rate based on the data rate of the first inference process and the amount of the first type of resources occupied by the first inference.

[1267] As an example, the larger the first ratio, the smaller the first data rate; the first ratio is the ratio of the amount of the first type of resources occupied by the first inference to the data rate processed by the first inference.

[1268] As an example, the first node determines the first data rate based on the first channel.

[1269] As an example, the first node determines the first data rate based on the data quality carried by the first channel.

[1270] As an example, the higher the quality of the data carried by the first channel, the greater the first data rate.

[1271] As an example, the first node determines the first data rate based on one or more of the logical channel type, LCID, LCG ID, MAC subheader, QoS flow, and PDU session of the data carried by the first channel.

[1272] As an example, the first node determines the first data rate based on the amount of first-type resources occupied by the processing of the first channel.

[1273] As one embodiment, the first node determines the first data rate based on the amount of a first type of resource occupied by the inference process of the first channel.

[1274] As an example, the first node determines the first data rate based on the output of inference processed by the first channel.

[1275] As an example, the output of the inference process for the first channel explicitly indicates the first data rate.

[1276] As an example, the output of the inference process for the first channel implicitly indicates the first data rate.

[1277] As an example, the output of the inference process for the first channel indicates the first data rate by indicating a subset of the training dataset.

[1278] As one embodiment, the first node determines the first data rate based on the amount of a first type of resource that is not occupied during processing on the first channel.

[1279] As an example, the first node determines the first data rate based on the data rate of the first channel.

[1280] As an example, the higher the data rate of the first channel, the lower the first data rate.

[1281] As one embodiment, the first node determines the first data rate based on the data rate of the first channel and the amount of first-class resources occupied by the processing of the first channel.

[1282] As an example, the larger the first ratio, the smaller the first data rate; the first ratio is the ratio of the amount of first-type resources occupied by the processing of the first channel to the data rate of the first channel.

[1283] As one embodiment, the processing of the first channel includes at least one of receiving and transmitting.

[1284] As one embodiment, the first channel includes a PDSCH, and the processing of the first channel includes receiving.

[1285] As one embodiment, the first channel includes PUSCH, and the processing of the first channel includes transmission.

[1286] As an example, the first node determines the first data rate based on the amount of a first type of resource that is not occupied during the first reference time.

[1287] As an example, the first node determines the first data rate based on the amount of a first type of resource occupied during the first reference time.

[1288] As an example, the first node determines the first data rate based on the data rate processed at the first reference time.

[1289] As an example, the first node determines the first data rate based on the quality of the data processed at the first reference time.

[1290] As an example, the first node determines the first data rate based on one or more of the logical channel type, LCID, LCG ID, MAC subheader, QoS flow, and PDU session of the data processed at the first reference time.

[1291] As an example, the first node determines the first data rate based on at least one of the data rate and data quality of the data reception or data transmission processing based on the first inference configuration at the first reference time.

[1292] Example 25

[1293] Example 25 illustrates a transmission flowchart according to one embodiment of this application; as shown in Figure 25. In Figure 25, the second node U3 and the first node U4 are communication nodes transmitting via an air interface. In Figure 25, the steps in blocks F251 to F257 are respectively optional.

[1294] For the second node U3, in step S25301, a second report is received; in step S25302, a first message is sent; in step S25303, a second data rate is sent; in step S25304, a second message is received; in step S25305, a second data rate is received; in step S25306, a second signaling is sent; in step S25307, a first signaling is sent; and in step S2531, a first report is received.

[1295] For the first node U4, a second report is sent in step S25401; a first message is received in step S25402; a second data rate is received in step S25403; a second message is sent in step S25404; a second data rate is sent in step S25405; a second signaling is received in step S25406; a first signaling is received in step S25407; and a first report is sent in step S2541.

[1296] In Example 25, the first report indicates that the idle data rate capability is a first data rate, the idle data rate capability being configured for a first inference.

[1297] As an example, the first node U4 is the first node in this application.

[1298] As an example, the second node U3 is the second node in this application.

[1299] As one embodiment, the air interface between the second node U3 and the first node U4 includes a wireless interface between the base station equipment and the user equipment.

[1300] As one embodiment, the air interface between the second node U3 and the first node U4 includes a wireless interface between the relay node device and the user equipment.

[1301] As one embodiment, the air interface between the second node U3 and the first node U4 includes the interface between the core network equipment and the user equipment.

[1302] As one embodiment, the air interface between the second node U3 and the first node U4 includes the interface between the OTT server (Over-The-Top server) and the user equipment.

[1303] As one embodiment, the air interface between the second node U3 and the first node U4 includes the interface between the NAS (Network Access Server) device and the user equipment.

[1304] As one embodiment, the air interface between the second node U3 and the first node U4 includes a wireless interface between user equipment.

[1305] As one embodiment, the first node U4 includes a terminal.

[1306] As one embodiment, the first node U4 includes a user equipment.

[1307] As one embodiment, the second node U3 includes the serving cell sustaining base station of the first node U2.

[1308] As one embodiment, the second node U3 includes an OTT (Over-The-Top) server.

[1309] As an example, the second node U3 includes OAM (Operation Administration and Maintenance).

[1310] As one embodiment, the second node U3 includes a NAS device.

[1311] As one embodiment, the second node U3 includes core network equipment.

[1312] As an example, the first report is transmitted on PUSCH (Physical Uplink Shared Channel).

[1313] As an example, the first report is transmitted on PUCCH (Physical Uplink Control Channel).

[1314] As an example, the first data rate is for a first inference, which depends on the first inference configuration.

[1315] As an example, the first data rate is for a first channel, and the processing of the first channel depends on the first inference configuration.

[1316] As one embodiment, the first data rate is relative to a first reference time, which is associated with the first inference configuration.

[1317] As an example, the steps in blocks F252 and F254 of Figure 25 are both present, the first message indicates multiple inference configurations, and the second message indicates that a inference configuration in a first subset of the multiple inference configurations is available; wherein, the first subset of inference configurations includes the first inference configuration.

[1318] As an example, the first message is transmitted on PDSCH (Physical Downlink Shared Channel).

[1319] As an example, the second message is transmitted on PUSCH (Physical Uplink Shared Channel).

[1320] As an example, the step in block F256 of Figure 25 exists, whereby the second signaling activates the first inference configuration.

[1321] As one example, the reception of the second signaling is later than the transmission of the second message.

[1322] As an example, the second signaling is transmitted on the PDSCH.

[1323] As an example, the second signaling is transmitted on the PDCCH.

[1324] As an example, the step in block F253 of Figure 25 includes the first node receiving the second data rate, which is configured for the first inference.

[1325] As a sub-implementation of the above embodiments, the second data rate is the minimum data rate for the first inference configuration.

[1326] As a sub-implementation of the above embodiments, the second data rate is the maximum data rate for the first inference configuration.

[1327] As a sub-implementation of the above embodiment, the second data rate is carried by RRC signaling.

[1328] As a sub-implementation of the above embodiment, the second data rate is carried by RRC IE.

[1329] As a sub-implementation of the above embodiments, the second data rate is carried by an RRC message.

[1330] As a sub-implementation of the above embodiments, the second data rate is carried by the first message.

[1331] As a sub-implementation of the above embodiment, the second data rate is carried by the first message, and receiving the first message means receiving the content in the first message other than the second data rate.

[1332] As a sub-example of the above embodiment, the first node receives the second data rate on the PDSCH.

[1333] As an example, the step in block F255 of Figure 25 includes the first node sending the second data rate, the second data rate being configured for the first inference.

[1334] As a sub-implementation of the above embodiments, the second data rate is the maximum data rate for the first inference configuration.

[1335] As a sub-implementation of the above embodiment, the second data rate is carried by RRC signaling.

[1336] As a sub-implementation of the above embodiments, the second data rate is carried by an RRC message.

[1337] As a sub-implementation of the above embodiment, the second data rate is carried by the RRCReconfigurationComplete message.

[1338] As a sub-implementation of the above embodiments, the second data rate is carried by a UE Assistance Information message.

[1339] As a sub-implementation of the above embodiments, the second data rate is carried by a UE Capability Information message.

[1340] As a sub-implementation of the above embodiments, the second data rate is carried by the UE capability IE.

[1341] As a sub-implementation of the above embodiment, the second data rate is carried by the second message.

[1342] As a sub-implementation of the above embodiment, the second data rate is carried by the second message, and sending the second message means sending the content in the second message other than the second data rate.

[1343] As a sub-example of the above embodiment, the first node transmits the second data rate on the PUSCH.

[1344] As an example, the steps in blocks F253 and F255 of Figure 25 will not exist simultaneously.

[1345] As an example, the step in block F251 of Figure 25 includes the following: the second report indicates L0 first-class resources, where L0 is a positive integer greater than 1, and the L0 first-class resources are used for inference; wherein the first data rate depends on the number of first-class resources occupied by the first inference configuration.

[1346] As an example, the second report is transmitted on PUSCH.

[1347] As one example, the second report is sent earlier than the first report.

[1348] As an example, the second report and the first report are transmitted on the same physical layer channel.

[1349] As one example, the second report is sent before the first message is received.

[1350] As one example, the second report is sent later than the first message is received.

[1351] As one embodiment, the first report is triggered by an event in a first event set, the first event set including at least one of the following:

[1352] ●The change in the number of first-type resources occupied by the first inference configuration is greater than the first threshold;

[1353] ●The maximum data rate supported by the first inference configuration changes more than the second threshold.

[1354] ●The change in the number of first-class resources occupied per unit data rate processed by the first inference configuration is greater than the third threshold;

[1355] ● Whether at least one inference configuration is available changes;

[1356] ●At least one inference configuration is activated or deactivated;

[1357] ●The first timer expired.

[1358] As an example, the step in block F257 of Figure 25 exists, whereby the first signaling triggers the first report.

[1359] As an example, the first signaling is transmitted on the PDSCH.

[1360] As an example, the first signaling is transmitted on the PDCCH.

[1361] Example 26

[1362] Example 26 illustrates a schematic diagram of a first inference according to an embodiment of this application; as shown in Figure 26. In Example 26, the first inference depends on a first inference configuration.

[1363] As an example, the first inference is AI (Artificial Intelligence) inference.

[1364] As an example, the first inference is ML (Machine Learning) inference.

[1365] As an example, the first inference is AI inference or ML inference.

[1366] As an example, the first inference is used for data reception or data transmission.

[1367] As an example, the first inference is an inference corresponding to the first inference configuration.

[1368] As an example, the first inference configuration is used to configure the first inference.

[1369] As an example, the inference-related parameters included in the first inference configuration are used to determine at least one of the input, output, and model of the first inference.

[1370] In Figure 26(a), the first inference is used for data reception.

[1371] As an example, the data reception refers to the reception of the PDSCH.

[1372] As one embodiment, the data reception includes one or more of TBS determination, channel estimation, demodulation, multi-antenna reception, and channel decoding.

[1373] As one example, the data reception includes the recovery of TB or CB.

[1374] As an example, the first inference is used for data reception, and the inference-related parameters included in the first inference configuration are used to determine the input and model of the first inference.

[1375] As an example, the inputs to the first inference include at least one of PDSCH and DMRS.

[1376] As an example, the input to the first inference depends on at least one of PDSCH and DMRS.

[1377] As an example, at least one of PDSCH and DMRS is used to generate the input for the first inference.

[1378] As an example, the input to the first inference includes at least one of preprocessed DMRS and preprocessed PDSCH.

[1379] As an example, the first inference is used to recover TB or CB.

[1380] As an example, the output of the first inference includes the recovered TB or CB.

[1381] As an example, the output of the first inference is used to recover TB or CB.

[1382] As an example, the output of the first inference includes at least one of the following: channel estimation result, demodulated symbol, and decoded bit.

[1383] As an example, the inference-related parameters included in the first inference configuration are used to determine the PDSCH used as input for the first inference.

[1384] As an example, the first inference configuration includes inference-related parameters indicating the configuration information of the PDSCH used as the input for the first inference.

[1385] As an example, the configuration information of the PDSCH includes one or more of the following: time-domain resources, frequency-domain resources, DMRS configuration, MCS table, maximum modulation order, maximum number of layers, and TCI state.

[1386] As an example, the configuration information of the PDSCH includes a TBS scaling factor, which is used to determine the TBS.

[1387] In Figure 26(b), the first inference is used for data transmission.

[1388] As an example, the data transmission refers to the transmission of PUSCH.

[1389] As one example, the data transmission includes the preparation or generation of PUSCH.

[1390] As one example, the data transmission includes one or more of DMRS generation, DMRS insertion, channel coding, modulation, and precoding.

[1391] As an example, the first inference is used for data transmission, and the inference-related parameters included in the first inference configuration are used to determine the output and model of the first inference.

[1392] As an example, the input to the first inference depends on TB or CB.

[1393] As an example, TB or CB is used as input for the first inference.

[1394] As an example, the input to the first inference includes TB or CB.

[1395] As an example, the input to the first inference includes the channel-coded bits of TB or CB.

[1396] As an example, the input to the first inference includes symbols of TB or CB after channel coding and modulation.

[1397] As an example, the input to the first inference includes symbols of TB or CB after channel coding, modulation, and precoding.

[1398] As an example, the first inference is used to generate PUSCH.

[1399] As an example, the output of the first inference is used to generate PUSCH.

[1400] As an example, the output of the first inference includes PUSCH.

[1401] As an example, the output of the first inference includes symbols transmitted on the PUSCH.

[1402] As an example, the output of the first inference is used to generate symbols transmitted on the PUSCH.

[1403] As an example, the symbols transmitted on the PUSCH include modulation symbols.

[1404] As an example, the symbols transmitted on the PUSCH include at least one of DMRS and PTRS.

[1405] As an example, the symbols transmitted on the PUSCH include OFDM (Orthogonal Frequency Division Multiplexing) symbols.

[1406] As an example, the output of the first inference includes at least one of DMRS, PTRS, modulated symbols, encoded bits, and pre-coded symbols.

[1407] As an example, the output of the first inference is processed by one or more of rate matching, modulation, layer mapping, precoding, mapping to resource blocks, OFDM modulation, and upconversion to generate a PUSCH.

[1408] Example 27

[1409] Example 27 illustrates a schematic diagram of a first data rate for a first inference according to an embodiment of this application; as shown in Figure 27.

[1410] As one embodiment, the idle data rate capability configured for the first inference includes the first data rate being configured for the first inference.

[1411] As an example, the first data rate is the idle data rate capability determined by the first node based on the first inference.

[1412] As an example, the first data rate is conditioned on the first inference.

[1413] As an example, the first data rate is the idle data rate capability conditioned on the first inference.

[1414] As an example, the first data rate indicates the additional data rate that can be processed based on the first inference.

[1415] As an example, the first data rate indicates the maximum additional data rate that can be processed based on the first inference.

[1416] As an example, the first data rate indicates the additional data rate that can be processed beyond the data processed in the first inference process.

[1417] As an example, the first data rate indicates the maximum additional data rate that can be processed beyond the data processed in the first inference process.

[1418] As an example, the first data rate indicates the data rate that can be additionally processed based on the data reception or data transmission configured in the first inference, in addition to the data processed in the first inference.

[1419] As an example, the first data rate indicates the maximum data rate that can be additionally processed based on the first inference configuration for data reception or data transmission, in addition to the data processed by the first inference.

[1420] As an example, the first data rate depends on the data rate of the first inference process.

[1421] As an example, the first node determines the first data rate based on the data rate of the first inference process.

[1422] As an example, the first data rate increases with the increase of the data rate of the first inference process.

[1423] As an example, the first data rate decreases as the data rate of the first inference process increases.

[1424] Example 28

[1425] Example 28 illustrates a schematic diagram of a first data rate for a first inference according to an embodiment of this application; as shown in Figure 28. In Example 28, the first data rate indicates the maximum data rate that can be additionally processed beyond the data processed by the first inference.

[1426] In Figures 28(a) and 28(b), if the data rate of the additionally scheduled PDSCH reception of the first node, in addition to the data of the first inference process, does not exceed the first data rate, the additionally scheduled PDSCH reception is processed.

[1427] In Figures 28(a) and 28(b), if the data rate of the PDSCH received by the first node in addition to the data of the first inference processing exceeds the first data rate, the first node does not process the PDSCH.

[1428] As an example, the additionally scheduled PDSCH reception refers to the additionally scheduled PDSCH reception based on the first inference configuration.

[1429] In Figure 28(a), the additionally scheduled PDSCH reception overlaps with the first inference in the time domain.

[1430] In Figure 28(b), the additionally scheduled PDSCH receives overlapping data in the time domain with the data processed by the first inference.

[1431] As a sub-implementation of the above embodiments, the data processed by the first inference includes at least one PDSCH.

[1432] As a sub-implementation of the above embodiments, the data processed by the first inference includes at least one TB or CB.

[1433] In Figures 28(c) and 28(d), if the data rate of the additionally scheduled PUSCH transmissions by the first node, in addition to the data from the first inference process, does not exceed the first data rate, the additionally scheduled PUSCH transmissions are processed.

[1434] In Figures 28(c) and 28(d), if the data rate of the PUSCH that is additionally scheduled by the first node in addition to the data of the first inference processing exceeds the first data rate, the first node does not process the PUSCH.

[1435] As an example, the additionally scheduled PUSCH transmission refers to the additionally scheduled PUSCH transmission based on the first inference configuration.

[1436] In Figure 28(c), the additionally scheduled PUSCH transmission has its allocated time-domain resources overlap with the first inference.

[1437] In Figure 28(d), the allocated time-domain resources of the additionally scheduled PUSCH transmission overlap with the data processed by the first inference.

[1438] As a sub-example of the above embodiments, the data processed by the first inference includes at least one PUSCH.

[1439] As a sub-implementation of the above embodiments, the data processed by the first inference includes at least one TB or CB.

[1440] Example 29

[1441] Example 29 illustrates a schematic diagram of a first channel according to an embodiment of this application; as shown in Figure 29. In Example 29, the processing of the first channel depends on first inference, which in turn depends on a first inference configuration.

[1442] As an example, the first channel is a physical layer channel.

[1443] As an example, the first channel is a physical layer channel for transmitting TB or CB.

[1444] As one example, the first channel includes PDSCH or PUSCH.

[1445] As an example, the first channel is PDSCH or PUSCH.

[1446] As an example, the first channel is a downlink channel.

[1447] As a sub-implementation of the above embodiment, the first node receives the first channel.

[1448] As a sub-example of the above embodiment, the first node receives TB or CB transmitted on the first channel.

[1449] As an example, the first channel is an uplink channel.

[1450] As a sub-implementation of the above embodiment, the first node transmits the first channel.

[1451] As a sub-example of the above embodiment, the first node transmits TB or CB on the first channel.

[1452] As an example, the first inference is used for data reception, and the first channel is PDSCH.

[1453] As an example, the first inference is used for data transmission, and the first channel is PUSCH.

[1454] As one embodiment, the first channel includes at least the former of DMRS and PTRS.

[1455] As an example, the first channel is a reference channel.

[1456] As an example, the reference channel is a channel that the first node has not scheduled.

[1457] As an example, the reference channel is a channel that the first node has never actually received or transmitted.

[1458] As an example, the reference channel is a hypothetical channel.

[1459] As an example, the first channel is a real channel.

[1460] As an example, the actual channel is the channel that the first node is scheduled to use.

[1461] As an example, the actual channel is the channel that the first node actually receives or transmits.

[1462] As one embodiment, the first channel is a downlink channel, and the processing of the first channel includes receiving.

[1463] As one embodiment, the receiving includes one or more of the following: TBS determination, channel estimation, demodulation, multi-antenna reception, channel decoding, and CRC check.

[1464] As an example, the downlink channel includes PDSCH.

[1465] As one embodiment, the first channel is an uplink channel, and the processing of the first channel includes transmission.

[1466] As an example, the transmission includes one or more of the following: TBS determination, DMRS generation, DMRS insertion, modulation, precoding, channel coding, and CRC bit generation.

[1467] As an example, the downlink channel includes PUSCH.

[1468] As an example, "processing of the first channel depends on the first inference configuration" means that the processing of the first channel depends on the inference configured by the first inference configuration.

[1469] As an example, the processing of the first channel depends on the first inference configuration, meaning that the inference configured by the first inference configuration is used to process the first channel.

[1470] As an example, an inference is used to process the first channel, and the first inference configuration indicates at least one of the input-related parameters, output-related parameters, and model parameters of the inference.

[1471] In Figure 29(a), the first channel is used as the input for the first inference.

[1472] As an example, the input to the first inference includes the first channel.

[1473] As an example, the input to the first inference includes symbols on the first channel.

[1474] As an example, the first channel is used to generate the input for the first inference.

[1475] As an example, the first channel is preprocessed and used to generate the input for the first inference.

[1476] As an example, symbols on the first channel are used to generate the input for the first inference.

[1477] As an example, the symbols on the first channel are preprocessed and used to generate the input for the first inference.

[1478] As an example, the input to the first inference includes preprocessed symbols on the first channel.

[1479] As an example, the symbols on the first channel include modulation symbols.

[1480] As an example, the symbols on the first channel include at least one of DMRS and PTRS.

[1481] As an example, the symbols on the first channel include OFDM symbols.

[1482] In Figure 29(b), the first channel depends on the first inference.

[1483] As an example, the first channel depends on the output of the first inference.

[1484] As an example, the generation of the first channel depends on the first inference.

[1485] As an example, the output of the first inference includes the first channel.

[1486] As an example, the output of the first inference includes symbols transmitted on the first channel.

[1487] As an example, the output of the first inference is used to generate the first channel.

[1488] As an example, the output of the first inference is used to generate symbols transmitted on the first channel.

[1489] As an example, the symbols transmitted on the first channel include modulation symbols.

[1490] As an example, the symbols transmitted on the first channel include at least one of DMRS and PTRS.

[1491] As an example, the symbols transmitted on the first channel include OFDM symbols.

[1492] As an example, the output of the first inference is processed by one or more of rate matching, modulation, layer mapping, precoding, resource block mapping, OFDM modulation, and upconversion to generate the first channel.

[1493] Example 30

[1494] Example 30 illustrates a schematic diagram of a first data rate for a first channel according to an embodiment of the present application; as shown in Figure 30.

[1495] As one embodiment, the idle data rate capability for the first inference configuration includes the first data rate for the first channel.

[1496] As an example, the first data rate is the idle data rate capability determined by the first node based on the first channel.

[1497] As an example, the first data rate is conditioned on the first channel.

[1498] As an example, the first data rate is the idle data rate capability conditioned on the first channel.

[1499] As an example, the first data rate indicates the additional data rate that can be processed on top of the first channel.

[1500] As an example, the first data rate indicates the maximum additional data rate that can be processed on top of the first channel.

[1501] As an example, the first data rate indicates the additional data rate that can be processed outside of the first channel.

[1502] As an example, the first data rate indicates the maximum additional data rate that can be processed outside of the first channel.

[1503] As an example, the first data rate indicates the data rate that can be additionally processed based on the first inference configuration for data reception or data transmission outside of the first channel.

[1504] As an example, the first data rate indicates the maximum data rate that can be additionally processed based on the first inference configuration for data reception or data transmission outside of the first channel.

[1505] As an example, the first data rate depends on the data rate of the first channel.

[1506] As an example, the first node determines the first data rate based on the data rate of the first channel.

[1507] As an example, the first data rate increases with the increase of the data rate of the first channel.

[1508] As an example, the first data rate decreases as the data rate of the first channel increases.

[1509] As an example, the data rate of the first channel depends on the number of bits included in the TB carried by the first channel, the total number of CBs in the TB carried by the first channel and the number of scheduled CBs, and the numberology of the first channel.

[1510] As a sub-implementation of the above embodiments, the data rate of the first channel also depends on the number of symbols allocated to the first channel.

[1511] Example 31

[1512] Example 31 illustrates a schematic diagram of a first data rate for a first channel according to an embodiment of this application; as shown in Figure 31. In Example 31, the first data rate indicates the maximum additional data rate that can be processed outside of the first channel.

[1513] In Figure 31(a), if the data rate of the PDSCH received by the first node outside the first channel does not exceed the first data rate, the additionally scheduled PDSCH reception is processed.

[1514] In Figure 31(a), if the data rate of the PDSCH received by the first node outside the first channel exceeds the first data rate, the first node does not process the PDSCH.

[1515] As an example, the additionally scheduled PDSCH reception overlaps with the first channel in the time domain.

[1516] As an example, the additionally scheduled PDSCH reception refers to the additionally scheduled PDSCH reception based on the first inference configuration.

[1517] In Figure 31(b), if the data rate of the additionally scheduled PUSCH transmissions outside the first channel does not exceed the first data rate, the additionally scheduled PUSCH transmissions are processed.

[1518] In Figure 31(b), if the data rate of the PUSCH that is additionally scheduled by the first node outside the first channel exceeds the first data rate, the first node does not process the PUSCH.

[1519] As an example, the additionally scheduled PUSCH transmission has its allocated time-domain resources overlap with the first channel.

[1520] As an example, the additionally scheduled PUSCH transmission refers to the additionally scheduled PUSCH transmission based on the first inference configuration.

[1521] Example 32

[1522] Example 32 illustrates a schematic diagram of a first reference time according to an embodiment of this application; as shown in Figure 32. In Example 32, the first reference time is associated with the first inference configuration.

[1523] In Figures 32(a) and 32(c), the first reference time associated with the first inference configuration includes an overlap between an inference configured by the first inference configuration and the first reference time.

[1524] As a sub-implementation of the above embodiments, the time domain resources occupied by the inference overlap with the first reference time.

[1525] In Figure 32(a), the first reference time is a time period.

[1526] In Figure 32(c), the first reference time is a point in time.

[1527] In Figures 32(b) and 32(d), the first reference time associated with the first inference configuration includes an overlap between a channel processed for data reception or data transmission based on the first inference configuration and the first reference time.

[1528] As a sub-implementation of the above embodiments, the channel is a physical layer channel.

[1529] In Figure 32(b), the first reference time is a time period.

[1530] In Figure 32(d), the first reference time is a point in time.

[1531] As an example, the first reference time is a point in time.

[1532] As an example, the first reference time is a time period.

[1533] As an example, the first reference time is a time slot.

[1534] As one embodiment, the first reference time includes a positive integer number of symbols.

[1535] As one embodiment, the first reference time includes the duration of a positive integer number of symbols.

[1536] As one embodiment, the first reference time includes a positive integer number of OFDM symbols.

[1537] As an example, the first reference time includes the duration of a positive integer number of OFDM symbols.

[1538] As an example, the time-domain resources occupied by the first inference overlap with the first reference time.

[1539] As an example, the first reference time is the time-domain resource occupied by the first inference.

[1540] As an example, the first reference time is a point in time within the time-domain resources occupied by the first inference.

[1541] As one embodiment, the time domain resources occupied by the first channel overlap with the first reference time.

[1542] As an example, the first reference time is the time-domain resource occupied by the first channel.

[1543] As an example, the first reference time is the time slot occupied by the first channel.

[1544] As an example, the first reference time is a point in time within the time-domain resources occupied by the first channel.

[1545] As an example, the first reference time is a point in time within the time slot occupied by the first channel.

[1546] As one embodiment, the processing time of the first channel overlaps with the first reference time.

[1547] As an example, the first channel includes a PDSCH, and the processing time of the first channel refers to the time that the first node processes the first channel.

[1548] As a sub-example of the above embodiments, the process includes one or more of the following: TBS determination, channel estimation, demodulation, multi-antenna reception, and channel decoding.

[1549] As a sub-example of the above embodiments, the process includes the recovery of TB or CB.

[1550] As an example, the first channel includes PUSCH, and the processing time of the first channel refers to the preparation time of the first node for the first channel.

[1551] As a sub-example of the above embodiments, the preparation includes one or more of the following: TBS determination, DMRS generation or insertion, modulation, precoding, and channel coding.

[1552] As an example, the first reference time depends on the time-domain resources of the first report.

[1553] As an example, the target recipient of the first report determines the first reference time based on the time-domain resources of the first report.

[1554] As an example, the first report indicates the first reference time.

[1555] As one embodiment, the first signaling indicates the first reference time.

[1556] As an example, the first reference time is a point in time, and overlapping with the first reference time means including the first reference time.

[1557] Example 33

[1558] Example 33 illustrates a schematic diagram of a first data rate relative to a first reference time according to an embodiment of the present application; as shown in Figure 33.

[1559] As one embodiment, the idle data rate capability for a first inference configuration includes the first data rate for a first reference time.

[1560] As an example, the first data rate indicates the additional data rate that can be processed during the first reference time.

[1561] As an example, the first data rate indicates the maximum additional data rate that can be processed during the first reference time.

[1562] As an example, the first data rate indicates the additional data rate that data reception or data transmission based on the first inference configuration can be processed during the first reference time, in addition to the data transmission scheduled during the first reference time.

[1563] As an example, the first data rate indicates the maximum additional data rate that data reception or data transmission based on the first inference configuration can handle during the first reference time, in addition to the data transmission scheduled during the first reference time.

[1564] As an example, the first data rate indicates the additional data rate that data reception or data transmission based on the first inference configuration can be processed during the first reference time, in addition to the data transmission processed during the first reference time.

[1565] As an example, the first data rate indicates the maximum data rate that can be additionally processed during the first reference time for data reception or data transmission based on the first inference configuration, in addition to the data transmission processed during the first reference time.

[1566] As one example, the data transmission includes PDSCH transmission.

[1567] As one example, the data transmission includes PUSCH transmission.

[1568] As one example, the data transmission includes the transmission of TB or CB.

[1569] As an example, the time-domain resources allocated to the data transmission scheduled at the first reference time overlap with the first reference time.

[1570] As one embodiment, the time slot allocated to the data transmission scheduled at the first reference time overlaps with the first reference time.

[1571] As an example, the data transmission scheduled at the first reference time is based on the first inference configuration.

[1572] As one embodiment, the data transmission processed at the first reference time is the data transmission that overlaps with the first reference time.

[1573] As an example, the data transmission processed at the first reference time is based on the first inference configuration.

[1574] As an example, the first data rate depends on the data rate processed during the first reference time based on the data reception or data transmission according to the first inference configuration.

[1575] As an example, the first node determines the first data rate based on the data rate processed during the first reference time for data reception or data transmission based on the first inference configuration.

[1576] Example 34

[1577] Example 34 illustrates a schematic diagram of a first data rate relative to a first reference time according to an embodiment of this application; as shown in Figure 34. In Example 34, the first data rate indicates the maximum data rate that can be additionally processed in addition to the data transmissions scheduled or processed during the first reference time.

[1578] In Figures 34(a) and 34(c), if the data rate of the additionally scheduled PDSCH reception of the first node at the first reference time does not exceed the first data rate, the additionally scheduled PDSCH reception is processed.

[1579] In Figures 34(a) and 34(c), if the data rate of the PDSCH received by the first node during the first reference time exceeds the first data rate, the first node does not process the PDSCH.

[1580] In Figure 34(a), the first reference time is a time period.

[1581] In Figure 34(c), the first reference time is a point in time.

[1582] As an example, the additionally scheduled PDSCH reception refers to the additionally scheduled PDSCH reception based on the first inference configuration.

[1583] As one embodiment, the additionally scheduled PDSCH reception overlaps with the first reference time in the time domain.

[1584] As one embodiment, the time slot allocated to the additionally scheduled PDSCH receiver overlaps with the first reference time.

[1585] As an example, the processing time of the additionally scheduled PDSCH reception overlaps with the first reference time in the time domain.

[1586] In Figures 34(b) and 34(d), the additionally scheduled PUSCH transmissions are processed if the data rate of the additionally scheduled PUSCH transmissions of the first node at the first reference time does not exceed the first data rate.

[1587] In Figures 34(b) and 34(d), if the data rate of the PUSCH sent by the first node during the first reference time exceeds the first data rate, the first node does not process the PUSCH.

[1588] In Figure 34(b), the first reference time is a time period.

[1589] In Figure 34(d), the first reference time is a point in time.

[1590] As an example, the additionally scheduled PUSCH transmission refers to the additionally scheduled PUSCH transmission based on the first inference configuration.

[1591] As one embodiment, the allocated time-domain resources of the additionally scheduled PUSCH transmission overlap with the first reference time.

[1592] As one example, the time slot allocated to the additionally scheduled PUSCH transmission overlaps with the first reference time.

[1593] As an example, the processing time of the additionally scheduled PUSCH transmission overlaps with the first reference time in the time domain.

[1594] Example 35

[1595] Example 35 illustrates a schematic diagram of a first message and a second message according to an embodiment of this application; as shown in Figure 35. In Example 35, the first message indicates a plurality of inference configurations, and the second message indicates that a inference configuration in a first subset of the plurality of inference configurations is available, the first subset of inference configurations including the first inference configuration.

[1596] As one example, the first message is carried by higher layer signaling.

[1597] As an example, the first message is carried by RRC signaling.

[1598] As an example, the first message is carried by one or more RRC IEs.

[1599] As an example, the first message includes all or part of the information from one or more of the following IEs: CSI-MeasConfig IE, CSI-ReportConfig IE, PDSCH-Config IE, PUSCH-Config IE, SPS-Config IE, PDSCH-ServingCellConfig IE, PUSCH-ServingCellConfig IE, ConfiguredGrantConfig IE, and ServingCellConfig IE.

[1600] As an example, the first message is carried by an RRC message.

[1601] As an example, the first message is used for RRC configuration.

[1602] As an example, the first message is used for RRC reconfiguration.

[1603] As an example, the first message is carried by an RRC reconfiguration message.

[1604] As an example, the first message is carried by an RRC recovery message (RRCResume message).

[1605] As one example, the first message includes some or all of the information in the higher-level parameter reconfigurationWithSync.

[1606] As an example, the first message is used for UE capability query (UECapabilityEnquiry).

[1607] As an example, the first message is carried by a UE Capability Enquiry message.

[1608] As an example, the first message is carried by DL-DCCH-Message.

[1609] As one example, the second message is carried by higher layer signaling.

[1610] As an example, the second message is carried by RRC signaling.

[1611] As one example, the second message is carried by one or more RRC IEs.

[1612] As an example, the second message is carried by an RRC message.

[1613] As an example, the second message is carried by the RRCReconfigurationComplete message.

[1614] As an example, the second message is carried by a UE Assistance Information message.

[1615] As an example, the second message is carried by a UE Capability Information message.

[1616] As an example, the second message is carried by the UE capability IE.

[1617] As one example, the second message includes the capability report of the first node.

[1618] As one example, the second message includes all or part of the information in OtherConfig IE.

[1619] As one example, the UE capabilities include AI or ML-related capabilities.

[1620] As an example, the UE capabilities include capabilities related to AI or ML supported functionality.

[1621] As an example, both the first message and the second message are used for LCM (life cycle management).

[1622] As an example, both the first message and the second message are used for the LCM of AI / ML models or AI / ML functions.

[1623] As one example, the second message indicates the first inference configuration subset.

[1624] As an example, the second message explicitly specifies a subset of the first inference configuration.

[1625] As an example, the second message indicates the first inference configuration subset from the plurality of inference configurations.

[1626] As an example, the second message implicitly includes a subset of the first inference configuration.

[1627] As an example, the second message indicates the first subset of inference configurations by indicating that an inference configuration is unavailable among the plurality of inference configurations.

[1628] As one embodiment, the plurality of inference configurations includes the first inference configuration.

[1629] As an example, any of the plurality of inference configurations includes higher layer signaling.

[1630] As an example, any of the plurality of inference configurations includes RRC signaling.

[1631] As an example, any one of the plurality of inference configurations is carried by one or more RRC IEs.

[1632] As an example, one of the plurality of inference configurations includes a MAC CE.

[1633] As an example, one of the plurality of inference configurations includes DCI.

[1634] As an example, one of the multiple inference configurations is carried by both RRC signaling and MAC CE.

[1635] As an example, one of the multiple inference configurations is carried by both RRC signaling and DCI.

[1636] As an example, any one of the plurality of inference configurations is used to configure inference.

[1637] As an example, any one of the plurality of inference configurations is used to determine at least one of the inference inputs and outputs.

[1638] As an example, any one of the plurality of inference configurations is used to determine at least one of the inference input, output, and model.

[1639] As an example, any of the plurality of inference configurations includes inference-related parameters.

[1640] As an example, the inference-related parameters include at least one of the following: input-related parameters, output-related parameters, and model parameters.

[1641] As one embodiment, the plurality of inference configurations includes a second inference configuration, which is used to generate channel information.

[1642] As an example, the output of the inference configured by the second inference configuration includes channel information.

[1643] As an example, the channel information includes one or more of CQI (Channel Quality Indicator), PMI (Precoding Matrix Indicator), CRI (CSI-RS Resource Indicator), RI (Rank Indicator), SSBRI (SS / PBCH Block Resource Indicator), RSRP (Reference Signal Received Power), and SINR (Signal-to-Interference and Noise Ratio).

[1644] As an example, the channel information includes one or more of CRI, SSBRI, RSRP, and SINR.

[1645] As an example, the channel information includes at least one of the following: predicted beam information, predicted CSI (Channel State Information), and compressed CSI.

[1646] As one example, the channel information includes compressed CSI.

[1647] As one example, the channel information includes precoding information.

[1648] As an example, the channel information is used to determine at least one precoding matrix.

[1649] As one embodiment, the plurality of inference configurations includes a third inference configuration, which is used for positioning.

[1650] As an example, the output of the inference configured by the third inference configuration includes location information.

[1651] As an example, the output of the inference configured by the third inference configuration includes location assistance information.

[1652] As an example, the output of the inference configured by the third inference configuration is used to assist in localization.

[1653] As an example, the input-related parameters are used to determine the RS (Reference Signal) resource for obtaining the inference dataset.

[1654] As an example, the RS resources include downlink RS resources.

[1655] As an example, the RS resources include at least one of CSI-RS (Channel State Information Reference Signal) resources and SS / PBCH (Synchronization Signal / Physical Broadcast Channel) block resources.

[1656] As an example, the RS resources include one or more of DMRS (Demodulation Reference Signal), PRS (Positioning Reference Signal) resources and PTRS (Phase-Tracking Reference Signal).

[1657] As one example, the RS resources include uplink RS resources.

[1658] As an example, the RS resource includes the SRS (Sounding Reference Signal) resource.

[1659] As an example, the input-related parameters are used to determine the configuration information of the RS resources used to obtain the inference dataset.

[1660] As an example, the configuration information of the RS resources includes one or more of the following: frequency domain resources, time domain resources, number of ports, CDM type, density, quasi co-location relationship, TCI status, time domain behavior, power control parameters, and BWP (Bandwidth Part) index.

[1661] As one example, the time-domain behavior includes periodic, semi-persistent, and aperiodic.

[1662] As an example, the input-related parameters are used to determine the number of RS transmission opportunities for obtaining the input for inference.

[1663] As an example, the input-related parameters are used to determine the physical layer channel for obtaining the inference dataset, the physical layer channel including at least one of PDSCH and PUSCH.

[1664] As an example, the input-related parameters are used to determine the configuration information of the physical layer channels used to obtain the inference dataset, the physical layer channels including at least one of PDSCH and PUSCH.

[1665] As an example, the output-related parameters are used to determine the content of the inference output.

[1666] As one example, the content includes channel information.

[1667] As one example, the content includes location information or auxiliary location information.

[1668] As an example, the auxiliary positioning information includes one or more of RSRP, RSRQ (Reference Signal Received Quality), RSSI (Received Signal Strength Indicator), TDOA (Time Difference Of Arrival), RTT (Round Trip Time), RSTD (Reference Signal Time Difference), AoA (Angle of Arrival), and AoD (Angle of Departure).

[1669] As one example, the content includes the recovered TB or CB.

[1670] As one example, the content includes physical layer channels.

[1671] As an example, the output-related parameters are used to determine the physical layer channel, which includes at least one of PDSCH and PUSCH.

[1672] As an example, the output-related parameters are used to determine the configuration information of the physical layer channel, which includes at least one of PDSCH and PUSCH.

[1673] As an example, the model parameters are used to determine a model.

[1674] As an example, the model parameters include some or all of the parameters used to construct a model.

[1675] As an example, the model parameters indicate the training dataset.

[1676] As an example, the model parameters include an identifier that is associated with inference.

[1677] As a sub-implementation of the above embodiments, the identifier indicates a model.

[1678] As a sub-implementation of the above embodiments, the identifier indicates a reference model or a nominal model.

[1679] As a sub-implementation of the above embodiments, the identifier indicates a training dataset.

[1680] As a sub-implementation of the above embodiments, the identifier indicates an inference dataset.

[1681] As a sub-implementation of the above embodiments, the identifier is used to identify a reasoning configuration.

[1682] As a sub-implementation of the above embodiments, the identifier indicates the RS used to obtain the training dataset.

[1683] As a sub-example of the above embodiments, the identifier indicates the RS used to obtain the inference dataset.

[1684] As a sub-implementation of the above embodiments, the identifier indicates the characteristics of the RS used to obtain the training dataset or inference dataset.

[1685] As a sub-example of the above embodiments, the first identifier indicates the data quality used to obtain the training dataset, inference dataset, or performance monitoring dataset.

[1686] As an example, any inference configuration in the first subset of inference configurations is one of the plurality of inference configurations.

[1687] As an example, the first inference configuration subset includes only the first inference configuration.

[1688] As an example, in addition to the first inference configuration, the first subset of inference configurations also includes at least one other inference configuration among the plurality of inference configurations.

[1689] As an example, any inference configuration in the first subset of inference configurations is available.

[1690] As an example, the availability of an inference configuration means that the inference configuration is ready to be activated or executed.

[1691] As an example, the meaning of "inference configuration is available" includes that the inference configured by the inference configuration is available.

[1692] As an example, the availability of an inference configuration means that the first node is ready to apply the inference configured by the inference configuration.

[1693] As an example, the availability of an inference configuration means that the inference configured by the inference configuration is ready to be activated or executed.

[1694] As an example, the availability of an inference configuration means that the inference model configured by the inference configuration is available.

[1695] As an example, the availability of an inference configuration means that the training of the inference model configured by the inference configuration has been completed.

[1696] As an example, the availability of an inference configuration means that the inference model configured by the inference configuration is ready to be activated.

[1697] Generally, how the first node determines whether an inference configuration is available is determined by the hardware vendor. Below are some non-limiting implementation methods:

[1698] As an example, the first node determines whether an inference configuration is available based on at least one of the additional conditions on the network side and the additional conditions on the UE side.

[1699] As an example, the first node determines whether a reasoning configuration is available based on whether the training of the reasoning model corresponding to a reasoning configuration is complete.

[1700] As an example, the first node determines whether an inference configuration is available based on the detection result or performance monitoring result of the inference model corresponding to the inference configuration.

[1701] As an example, the first node determines whether inference is available based on its own capabilities, such as, but not limited to, computing power, storage capacity, and power consumption.

[1702] As an example, the first node determines whether an inference is available based on whether the model's training dataset and an inference configuration are consistent.

[1703] Example 36

[1704] Example 36 illustrates a schematic diagram of a second signaling activation of a first inference configuration according to an embodiment of this application; as shown in Figure 36.

[1705] As one example, the second signaling includes MAC CE.

[1706] As an example, the second signaling is MAC CE.

[1707] As one example, the second signaling includes DCI.

[1708] As an example, the second signaling is DCI.

[1709] As an example, once the first inference configuration is indicated to be available, the activation of the first inference configuration depends on the activation command.

[1710] As an example, once the first inference configuration is indicated to be available, the first node cannot automatically assume that the first inference configuration is activated.

[1711] As an example, activating the first inference configuration includes activating the inference configured by the first inference configuration.

[1712] As one embodiment, activating the first inference configuration includes activating data reception or data transmission based on the first inference configuration.

[1713] As one embodiment, activating the first inference configuration includes activating inference configured using the first inference configuration to process data reception or data transmission.

[1714] As an example, activating the first inference configuration includes activating the AI / ML model indicated by the first inference configuration.

[1715] As one example, the second signaling depends on the indication of the second message.

[1716] As one example, the sender of the second signaling relies on the indication of the second message to determine the second signaling.

[1717] As an example, the sender of the second signaling determines which one or more of the plurality of inference configurations can be activated based on the first subset of inference configurations.

[1718] As an example, the sender of the second signaling activates only the inference configurations available among the plurality of inference configurations.

[1719] Example 37

[1720] Example 37 illustrates a schematic diagram of a second data rate for a first inference configuration according to an embodiment of the present application; as shown in Figure 37.

[1721] As an example, the unit of the second data rate is Mbps (megabits per second).

[1722] Considering future ultra-wideband transmission, the unit of the second data rate may also be Gbps or Mbpms (megabits per millisecond).

[1723] As an example, the second data rate is the maximum data rate.

[1724] As an example, the second data rate is the minimum data rate.

[1725] As one embodiment, the second data rate is the maximum uplink data rate or the maximum downlink data rate.

[1726] As one example, the second data rate is the minimum uplink data rate or the minimum downlink data rate.

[1727] As an example, the first node sends the second data rate, which is the maximum data rate.

[1728] As one embodiment, the first node receives the second data rate, which is the maximum data rate.

[1729] As one embodiment, the first node receives the second data rate, which is the minimum data rate.

[1730] As one embodiment, the second data rate for the first inference configuration includes the maximum data rate supported by the data reception or data transmission based on the first inference configuration.

[1731] As one embodiment, the second data rate for the first inference configuration includes the maximum data rate expected to be obtained based on the first inference configuration for data reception or data transmission.

[1732] As one embodiment, the second data rate for the first inference configuration includes the minimum data rate expected to be obtained for data reception or data transmission based on the first inference configuration.

[1733] As one example, the first data rate depends on the difference between the second data rate and the current data rate.

[1734] As an example, the first data rate increases as the difference between the second data rate and the current data rate increases.

[1735] As an example, the first data rate is equal to the difference between the second data rate and the current data rate.

[1736] As an example, the current data rate is the data rate of the first inference process.

[1737] As an example, the current data rate is the data rate of the first channel.

[1738] As an example, the current data rate is the data rate scheduled at the first reference time.

[1739] As an example, the current data rate is the data reception or data transmission rate scheduled based on the first inference configuration at the first reference time.

[1740] As an example, the current data rate is the data rate processed at the first reference time.

[1741] As an example, the current data rate is the data reception or data transmission rate based on the first inference configuration processed at the first reference time.

[1742] Example 38

[1743] Example 38 illustrates a schematic diagram of a second report according to one embodiment of this application; as shown in Figure 38. In Example 38, the illustrated second report indicates L0 first-class resources.

[1744] As an example, the second report is carried by a higher-layer message.

[1745] As an example, the second report is carried by an RRC message.

[1746] As an example, the second report is carried by RRC signaling.

[1747] As an example, the second report is carried by MAC CE.

[1748] As an example, the second message is carried by a UE Assistance Information message.

[1749] As an example, the second message is carried by a UE Capability Information message.

[1750] As one example, the second report includes UE capability information.

[1751] As an example, the second report is carried by the UE capability IE.

[1752] As one example, the second report includes the capability report of the first node.

[1753] As an example, the second report includes the UE processing capability of the first node.

[1754] As one example, the second report includes the UE capability indication of the first node.

[1755] As one example, the second report applies only to one carrier or one serving cell of the first node.

[1756] As one example, the second report applies to all carriers of the first node.

[1757] As one example, the second report applies to all carriers belonging to the same cell group of the first node.

[1758] As one embodiment, the second report applies to all carriers belonging to the same band or band combination of the first node.

[1759] As one example, the second report applies to all serving cells of the first node.

[1760] As one example, the second report applies to all serving cells belonging to the same cell group as the first node.

[1761] As one example, the second report applies to all serving cells of the first node that belong to the same frequency band or frequency band combination.

[1762] Typically, the same cell group is either an MCG (Master Cell Group) or an SCG (Secondary Cell Group).

[1763] As an example, L0 is the maximum number of the first type of resources.

[1764] As an example, L0 is the maximum number of the first type of resources supported by the first node.

[1765] As an example, L0 is the maximum number of the first type of resources that the first node supports simultaneously.

[1766] As an example, L0 is the maximum number of first-class resources supported by the first node on a carrier.

[1767] As an example, L0 is the maximum number of first-class resources supported by the first node on all carriers.

[1768] As an example, "all carriers" refers to all component carriers belonging to the same cell group.

[1769] As an example, all carriers refer to all component carriers belonging to the same frequency band or combination of frequency bands.

[1770] As an example, L0 is the maximum number of first-class resources supported by the first node on a serving cell.

[1771] As an example, the serving cell is either SpCell (Special Cell) or SCell (Secondary Cell).

[1772] As an example, L0 is the maximum number of first-class resources supported by the first node across all serving cells.

[1773] As an example, all serving cells refer to all serving cells belonging to the same cell group.

[1774] As an example, all serving cells refer to all serving cells belonging to the same frequency band or combination of frequency bands.

[1775] As an example, any one of the L0 first-class resources is used for inference.

[1776] As an example, the L0 first-class resources are used for computation or processing.

[1777] As an example, the L0 first-class resources are used for storage.

[1778] As an example, the L0 first-class resources are used for computing and storage.

[1779] As an example, the L0 first-class resources include computing resources.

[1780] As one embodiment, the L0 first-class resources include CSI processing units.

[1781] As one example, the L0 first-class resources include storage resources.

[1782] As one example, the storage resource includes memory.

[1783] As one example, the storage resources include video memory.

[1784] As one example, the storage resources are used to store some or all of the parameters of the AI / ML model.

[1785] As one example, the storage resources are used to store some or all of the intermediate results of the inference.

[1786] As one example, the storage resources are used to store part or all of the inference output.

[1787] As an example, the parameters of the AI / ML model include one or more of the following: convolution kernel size, number of convolutional layers, convolution stride, pooling kernel size, pooling kernel stride, pooling function, activation function, and number of feature maps.

[1788] As an example, the parameters of the AI / ML model include one or more of the following: storage convolution kernel, pooling kernel, pooling function, activation function, parameters of the pooling function, and parameters of the activation function.

[1789] Example 39

[1790] Example 39 illustrates a schematic diagram of a first data rate according to an embodiment of this application; as shown in Figure 39. In Example 39, the first data rate depends on a first quantity and a first rate, the first quantity being the quantity of a first type of resource occupied by the first inference configuration, and the first rate being the data rate of data reception or data transmission based on the first inference configuration.

[1791] The benefits of the above method include more accurate and dynamic reporting of the current idle data rate, further optimization of resource allocation, and improved resource utilization.

[1792] The advantages of the above method include improved data transmission performance.

[1793] The advantages of the above methods include ensuring QoS for various services.

[1794] As an example, any of the first type of resources occupied by the first inference configuration is one of the L0 first type of resources.

[1795] As an example, the first type of resources occupied by the first inference configuration includes the first type of resources occupied by the inference configured by the first inference configuration.

[1796] As an example, the first type of resources occupied by the first inference configuration includes the first type of resources occupied by data reception or data transmission based on the first inference configuration.

[1797] As an example, the first quantity is the quantity of the first type of resources occupied by the first inference.

[1798] As an example, the first quantity is the quantity of the first type of resources occupied by the processing of the first channel.

[1799] As an example, the first quantity is the number of first-type resources occupied by the first inference configuration at the first reference time.

[1800] As an example, the first quantity is the number of first-class resources occupied by the inference configured in the first inference configuration at the first reference time.

[1801] As an example, the first quantity is the number of first-class resources occupied by data reception or data transmission based on the first inference configuration at the first reference time.

[1802] As an example, the first quantity is less than or equal to L0.

[1803] As an example, the first node determines the first data rate based on the first quantity.

[1804] As an example, the first data rate decreases as the first quantity increases.

[1805] As one example, the first data rate depends on the difference between L0 and the first quantity.

[1806] As an example, the first data rate increases as the difference between L0 and the first quantity increases.

[1807] As an example, the first quantity is the number of first-type resources occupied by the first inference, and the first rate is the data rate of the first inference processing.

[1808] As an example, the first quantity is the amount of a first type of resource occupied by the processing of the first channel, and the first rate is the data rate of the first channel.

[1809] As one embodiment, the first quantity is the amount of a first type of resource occupied by data reception or data transmission in the first reference time based on the first inference configuration, and the first rate is the data rate processed by data reception or data transmission in the first reference time based on the first inference configuration.

[1810] As an example, the first quantity is the number of first-class resources occupied by the inference configured by the first inference configuration at the first reference time, and the first rate is the data rate processed by the inference configured by the first inference configuration at the first reference time.

[1811] As an example, the first data rate increases as the first rate increases.

[1812] As an example, the first data rate decreases as the first rate increases.

[1813] As an example, the first data rate decreases as the first quantity increases.

[1814] As one example, the first data rate depends on the ratio of the first quantity to the first rate.

[1815] As an example, the first data rate decreases as the ratio of the first quantity to the first rate increases.

[1816] As an example, the first data rate depends on L0, the first quantity, and the first rate.

[1817] Example 40

[1818] Example 40 illustrates a schematic diagram of a first data rate according to an embodiment of this application; as shown in Figure 40. In Example 40, the first data rate depends on the difference between L0 and a first quantity divided by a first ratio, the first ratio being the ratio between the first quantity and the first rate; the first quantity being the quantity of a first type of resource occupied by the first inference configuration, and the first rate being the data rate for data reception or d...

Claims

1. A first node used for wireless communication, characterized in that, include: A first receiver receives a first signaling, the first signaling including a first inference configuration; A first transmitter sends a first report, the first report indicating support for L1 feature sets, where L1 is a positive integer; Wherein, the L1 feature sets depend on the indication of the first signaling; any feature set in the L1 feature sets includes at least one of the following: maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, maximum data rate, and maximum modulation order; the feature corresponding to at least one indicator included in the first inference configuration does not belong to the first feature set, and the first feature set is a feature set in the L1 feature sets.

2. The first node according to claim 1, characterized in that, The first transmitter sends a second report indicating support for L0 feature sets, where L0 is a positive integer greater than 1; wherein the second report is sent before the first report, and the L1 feature sets depend on the L0 feature sets.

3. The first node according to claim 2, characterized in that, The first receiver receives a second signaling, which indicates first configuration information; wherein the L0 feature sets depend on the indication of the second signaling.

4. The first node of claim 2 or 3, wherein, The L0 special 5. The first node according to claim 2 or 3, characterized in that, The L0 feature sets are conditional on a first hypothesis; the first hypothesis includes the association between the inference configuration and the training dataset. A second hypothesis is also conditional; the second hypothesis includes not using inference.

6. The first node according to any one of claims 1 to 5, characterized in that, The first report is triggered by an event in a first event set, the first event set including at least one state change of inference configuration, the state change including one or more of being activated, deactivated, configured and reconfigured.

7. The first node of any of claims 1 to 5, wherein, The first receiver receives a third signaling; wherein the third signaling triggers the first report.

8. A second node for use in wireless communication, characterized by include: The second transmitter transmits a first signaling message, the first signaling message including a first inference configuration; The second receiver receives a first report, which indicates support for L1 feature sets, where L1 is a positive integer. Wherein, the L1 feature sets depend on the indication of the first signaling; any feature set in the L1 feature sets includes at least one of the following: maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, maximum data rate, and maximum modulation order; the feature corresponding to at least one indicator included in the first inference configuration does not belong to the first feature set, and the first feature set is a feature set in the L1 feature sets.

9. The second node of claim 8, wherein, The second receiver receives a second report indicating support for L0 feature sets, where L0 is a positive integer greater than 1; wherein the second report is received before the first report, and the L1 feature sets depend on the L0 feature sets.

10. The second node according to claim 9, characterized in that, The second transmitter sends a second signaling message, which indicates first configuration information; wherein the L0 feature sets depend on the indication of the second signaling message.

11. The second node of claim 8 or 9, wherein, The L0 feature sets are conditional on a first hypothesis; the first hypothesis includes the association between the inference configuration and the training dataset.

12. The second node of claim 8 or 9, characterized by, The L0 feature sets are conditional on a second hypothesis; the second hypothesis includes not using reasoning.

13. The second node of any of claims 8-12, wherein, The first report is triggered by an event in a first event set, the first event set including at least one state change of inference configuration, the state change including one or more of being activated, deactivated, configured and reconfigured.

14. The second node of any of claims 8-12, wherein, The second transmitter sends a third signaling message; wherein the third signaling message triggers the first report.

15. A method in a first node used for wireless communication, characterized by, include: Receive a first signaling, the first signaling including a first inference configuration; Send a first report indicating support for L1 feature sets, where L1 is a positive integer; Wherein, the L1 feature sets depend on the indication of the first signaling; any feature set in the L1 feature sets includes at least one of the following: maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, maximum data rate, and maximum modulation order; the feature corresponding to at least one indicator included in the first inference configuration does not belong to the first feature set, and the first feature set is a feature set in the L1 feature sets.

16. The method in the first node according to claim 15, characterized in that, include: Send a second report indicating support for L0 feature sets, where L0 is a positive integer greater than 1; The second report is sent before the first report, and the L1 feature sets depend on the L0 feature sets.

17. A method in a first node according to claim 16, characterised by, include: Receive a second signaling message, the second signaling message indicating the first configuration information; The L0 feature sets depend on the indication of the second signaling.

18. The method in the first node according to claim 16 or 17, characterized in that, The L0 feature sets are conditional on a first hypothesis; the first hypothesis includes the association between the inference configuration and the training dataset.

19. A method in a first node according to claim 16 or 17, characterized by, The L0 feature sets are conditional on a second hypothesis; the second hypothesis includes not using reasoning.

20. A method in a first node according to any of claims 15 to 19, characterized by, The first report is triggered by an event in a first event set, the first event set including at least one state change of inference configuration, the state change including one or more of being activated, deactivated, configured and reconfigured.

21. A method in a first node according to any of claims 15 to 19, characterized by, include: Receive third signaling; The third signaling triggers the first report.

22. A method in a second node used for wireless communication, characterized by, include: Send a first signaling message, the first signaling message including a first inference configuration; Receive a first report indicating support for L1 feature sets, where L1 is a positive integer; Wherein, the L1 feature sets depend on the indication of the first signaling; any feature set in the L1 feature sets includes at least one of the following: maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, maximum data rate, and maximum modulation order; the feature corresponding to at least one indicator included in the first inference configuration does not belong to the first feature set, and the first feature set is a feature set in the L1 feature sets.

23. A method in a second node according to claim 22, characterised by, include: Receive a second report indicating support for L0 feature sets, where L0 is a positive integer greater than 1; The second report is received before the first report, and the L1 feature sets depend on the L0 feature sets.

24. A method in a second node according to claim 23, characterised by, include: Send a second signaling message, the second signaling message indicating the first configuration information; The L0 feature sets depend on the indication of the second signaling.

25. A method in a second node according to claim 23 or 24, characterized by, The L0 feature sets are conditional on a first hypothesis; the first hypothesis includes the association between the inference configuration and the training dataset.

26. A method in a second node according to claim 23 or 24, characterized by, The L0 feature set is conditioned on a second hypothesis; the second hypothesis includes not employing reasoning.

27. A method in a second node according to any of claims 22 - 26, characterized by, The first report is triggered by an event in a first set of events, the first set of events including a state change of at least one reasoning configuration, the state change including one or more of being activated, being deactivated, being configured, and being reconfigured.

28. A method in a second node according to any of claims 22 - 26, characterized by, comprising: sending third signaling; wherein the third signaling triggers the first report.

Images