Reporting method and apparatus used in node for wireless communication

Abstract

The present application discloses a reporting method and apparatus used in a node for wireless communication. A first node sends a first message, wherein the first message indicates a plurality of functions supported by the first node and J resources in the first node, J is a positive integer greater than 1, the plurality of functions are all based on AI, only some of the plurality of functions are used for channel information reporting, and any one of the plurality of functions occupies at least one of the J resources. In the method, a plurality of functions including channel information reporting are achieved on the basis of AI / ML, thereby improving system performance, and better supporting various different functions, application scenarios or terminals.

Description

A method and apparatus for reporting in a node for wireless communication Technical Field

[0001] This application relates to transmission methods and apparatus in wireless communication systems, and more particularly to reporting schemes and apparatus in wireless communication systems. Background Technology

[0002] In traditional wireless communication, the UE (User Equipment) calculates CSI (Channel State Information) by measuring downlink reference signals. This 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). 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 NRR (release) 19, research on AI (Artificial Intelligence) / ML (Machine Learning) technologies was initiated, one aspect of which is AI-based channel information reporting, including beam prediction, CSI prediction, and more.

[0003] In traditional wireless communication, channel coding is used to improve the reliability of wireless communication. Typical channel coding techniques include Turbo codes, LDPC (Low Density Parity Check) codes, polar codes, and so on. In the future technological evolution of 5G and 6G, AI / ML-based coding techniques, AI / ML-based decoding techniques, and AI / ML-based channel estimation techniques have become research hotspots.

[0004] Since the specifications of AI models may extend beyond the scope of 3GPP (except for reference models 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 Networks), or a hybrid model composed of multiple models.

[0005] In the future evolution of 5G or 6G, AI / ML can be used for functions including but not limited to CSI measurement / computation / prediction, encoding, decoding, demodulation, and channel estimation. AI / ML models, training, and inference for different functions all require significant computing and storage resources. For example, the storage requirements of an AI model depend primarily on the number of its parameters, potentially reaching hundreds of megabytes (MB). AI / ML training and inference require substantial computing power, with common computing resources including GPUs (graphics processing units). Central processing units (CPUs) may be used for AI models or functions with lower computational demands. Therefore, supporting AI / ML technology in the future evolution of 5G or 6G requires considering the resource requirements of AI / ML models, training, and inference. Summary of the Invention

[0006] The applicant's research revealed that the 5G standard defines a CSI processing unit specifically for calculating CSI reports. This unit explicitly defines the amount of CSI processing resources required by the UE to calculate different types of CSI, as well as the duration of these resource occupancy. In the future evolution of 5G or 6G, AI / ML can be used for various functions, including but not limited to channel information reporting. These functions will inevitably consume certain resources within the node. Identifying the relationship between different functions and resources is a key issue that needs to be addressed. In 5G systems, the UE indicates the supported feature set by sending UE capability information messages. In future 5G or 6G technologies, resource occupancy is a significant factor; therefore, how to report the supported resource status is a crucial issue. To address these issues, this application discloses a solution. It should be noted that while many embodiments of this application focus on AI / ML, this application is also applicable to other solutions, such as traditional reporting schemes. Furthermore, adopting a unified solution across different scenarios (including but not limited to AI / ML-based schemes and traditional reporting schemes) helps reduce 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] As an example, the interpretation of terms in this application is based on the definitions in the 3GPP specification protocol TS38 series.

[0008] As an example, the interpretation of the terms in this application is based on the definitions in the 3GPP specification protocol TS28 series.

[0009] When necessary, the interpretation of terms in this application shall be based on the definitions of the 3GPP specification protocol TS38 series, or the definitions of the 3GPP specification protocol TS28 series.

[0010] This application discloses a method used in a first node of wireless communication, characterized by comprising:

[0011] Send the first message;

[0012] The first message indicates multiple functions supported by the first node and J resources in the first node, where J is a positive integer greater than 1; the multiple functions are all based on AI, only some of the multiple functions are used for channel information reporting, and any one of the multiple functions occupies at least one of the J resources.

[0013] As an example, the problems to be solved by this application include: AI / ML can be used for multiple functions including but not limited to channel information reporting. These functions will inevitably occupy certain resources in the first node. How to identify the relationship between different functions and the resources of the first node is a key problem that needs to be solved.

[0014] As an example, the AI ​​includes ML (Machine Learning) inference.

[0015] As an example, the advantages of the above method include: implementing multiple functions based on AI / ML, thereby improving system performance.

[0016] As an example, the advantages of the above method include: the recipient of the first message is fully aware of the AI / ML-based functions that the first node can support and the total number of available resources in the first node.

[0017] As an example, the advantages of the above method include: better support for various functions, application scenarios or terminals.

[0018] As an example, the advantages of the above method include: high flexibility and strong adaptability.

[0019] As one example, the first node is a user equipment.

[0020] As an example, the first node is a relay node.

[0021] As one example, the first node is a terminal.

[0022] As one example, the terminal is a user equipment.

[0023] According to one aspect of this application, the first message indicates the amount of resources occupied by one of the plurality of functions.

[0024] As an example, the advantages of the above method include: the recipient of the first message can fully know the amount of resources in the first node required for an AI / ML-based function of the first node, which helps to reasonably schedule or configure different functions in the future, and reduces or avoids the situation where the first node does not have enough resources for a scheduled or configured function.

[0025] According to one aspect of this application, the plurality of functions include one or more of receiving a wireless channel, encoding a wireless channel, channel estimation on DMRS, and predicting wireless link failure.

[0026] As an example, the advantages of the above method include: implementing functions other than channel information reporting based on AI / ML, thereby improving system performance.

[0027] According to one aspect of this application, it is characterized by comprising:

[0028] Send the first information block;

[0029] Wherein, the first information block is used to indicate N1 applicable functions of the first node from N functions, all of which are based on AI, where N is a positive integer greater than 1 and N1 is a positive integer not greater than N; the amount of resources occupied by each of the N1 applicable functions is equal to or less than J.

[0030] As an example, in the above method, the reporting of the first information block indicates the function that the first node can perform, i.e., the applicable functionality. The advantages of the above method include: avoiding the first node being scheduled or configured with an unsuitable or unexecutable function, thus ensuring system performance.

[0031] According to one aspect of this application, the N1 applicable functions satisfy a first condition; the first condition includes: the total number of resources occupied by all applicable functions is equal to or less than J.

[0032] As an example, in the above method, the selection of applicable functions takes into account the total amount of available resources in the first node, avoiding the scheduling or configuration of excessive functions that exceed the capabilities of the first node.

[0033] According to one aspect of this application, whether one of the N functions is the applicable function of the first node depends on the purpose of the function.

[0034] According to one aspect of this application, the first function and the second function are two functions among the N functions, and under the requirement of satisfying a first condition, the second function is determined as the applicable function of the first node with priority over the first function; wherein, only the first function among the first function and the second function is used for channel information reporting; the first condition includes: the total number of resources occupied by all applicable functions is equal to or less than J.

[0035] As an example, in the above method, it is taken into consideration that the total amount of available resources in the first node is limited, and the selection of applicable functions should take into account the purpose of the functions. For example, the second function takes precedence over the channel information reporting function. That is, when resources are insufficient, certain specific functions are given priority, thus optimizing the selection of applicable functions and ensuring system performance.

[0036] According to one aspect of this application, it is characterized by comprising:

[0037] Perform inference;

[0038] The reasoning performed by the first node is used to implement at least one of the N1 applicable functions.

[0039] According to one aspect of this application, it is characterized by comprising:

[0040] Receive the second information block.

[0041] As one example, the second information block indicates the N functions.

[0042] As one embodiment, the second information block instructs the first node to provide the applicable functions of the first node.

[0043] As one embodiment, the second information block instructs the first node to report UAI (UE Assistance Information).

[0044] According to one aspect of this application, the first information block is triggered when a second condition is met; the second condition includes: the first node receiving an RRC message indicating that the training dataset has been reconfigured.

[0045] As an example, in the above method, the first information block is triggered, which helps to report the applicability of the function in a timely manner and reduces delay.

[0046] According to one aspect of this application, it is characterized by comprising:

[0047] Receive the second message;

[0048] The second message may indicate that the first node reports the supported functions, or the second message may query the capabilities of the first node.

[0049] This application discloses a method used in a second node for wireless communication, characterized by comprising:

[0050] Receive the first message;

[0051] Wherein, the first node is the sender of the first message, the first message indicates multiple functions supported by the first node and J resources in the first node, where J is a positive integer greater than 1; the multiple functions are all based on AI, only some of the multiple functions are used for channel information reporting, and any one of the multiple functions occupies at least one of the J resources.

[0052] According to one aspect of this application, the first message indicates the amount of resources occupied by one of the plurality of functions.

[0053] According to one aspect of this application, the plurality of functions include one or more of receiving a wireless channel, encoding a wireless channel, channel estimation on DMRS, and predicting wireless link failure.

[0054] According to one aspect of this application, it is characterized by comprising:

[0055] Receive the first information block;

[0056] Wherein, the first information block is used to indicate N1 applicable functions of the first node from N functions, all of which are based on AI, where N is a positive integer greater than 1 and N1 is a positive integer not greater than N; the amount of resources occupied by each of the N1 applicable functions is equal to or less than J.

[0057] According to one aspect of this application, the N1 applicable functions satisfy a first condition; the first condition includes: the total number of resources occupied by all applicable functions is equal to or less than J.

[0058] According to one aspect of this application, whether one of the N functions is the applicable function of the first node depends on the purpose of the function.

[0059] According to one aspect of this application, the first function and the second function are two functions among the N functions, and under the requirement of satisfying a first condition, the second function is determined as the applicable function of the first node with priority over the first function; wherein, only the first function among the first function and the second function is used for channel information reporting; the first condition includes: the total number of resources occupied by all applicable functions is equal to or less than J.

[0060] According to one aspect of this application, the first node performs reasoning; wherein the reasoning performed by the first node is used to implement at least one of the N1 applicable functions.

[0061] According to one aspect of this application, it is characterized by comprising:

[0062] Send the second information block.

[0063] As one example, the second information block indicates the N functions.

[0064] As one embodiment, the second information block instructs the first node to provide the applicable functions of the first node.

[0065] As one example, the second information block instructs the first node to report UAI.

[0066] According to one aspect of this application, the first information block is triggered when a second condition is met; the second condition includes: the first node receiving an RRC message indicating that the training dataset has been reconfigured.

[0067] According to one aspect of this application, it is characterized by comprising:

[0068] Send a second message;

[0069] The second message may indicate that the first node reports the supported functions, or the second message may query the capabilities of the first node.

[0070] This application discloses a first node used for wireless communication, characterized in that it comprises:

[0071] The first processor sends the first message;

[0072] The first message indicates multiple functions supported by the first node and J resources in the first node, where J is a positive integer greater than 1; the multiple functions are all based on AI, only some of the multiple functions are used for channel information reporting, and any one of the multiple functions occupies at least one of the J resources.

[0073] This application discloses a second node used for wireless communication, characterized in that it comprises:

[0074] The second processor receives the first message;

[0075] Wherein, the first node is the sender of the first message, the first message indicates multiple functions supported by the first node and J resources in the first node, where J is a positive integer greater than 1; the multiple functions are all based on AI, only some of the multiple functions are used for channel information reporting, and any one of the multiple functions occupies at least one of the J resources.

[0076] As an example, compared with conventional solutions, this application has the following advantages:

[0077] To better adapt to various different functions, application scenarios, or terminals;

[0078] Better adaptable to various terminal capabilities;

[0079] High flexibility;

[0080] Highly adaptable;

[0081] The design has been simplified;

[0082] Enhanced reliability and robustness;

[0083] Enhanced overall system performance.

[0084] This application discloses a method used in a first node of wireless communication, characterized by comprising:

[0085] Send a first message; wherein the first message indicates J resources, multiple feature sets, and the amount of resources occupied by each feature set in the multiple feature sets;

[0086] Wherein, J is a positive integer greater than 1; the feature set 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 candidates for features in the feature set include at least two different features.

[0087] As an example, the problem this application aims to solve includes: how to report the status of supported resources.

[0088] In the above method, the first node, in the message indicating the supported feature set, also reports the supported resources and the resource occupancy requirements (i.e., the amount of resources occupied) of the supported feature set. The advantages of this method include: the target recipient of the first message (i.e., the second node in this application) can fully understand the feature set that the first node can support and the resource occupancy requirements of the feature set, which helps the target recipient of the first message to reasonably schedule or configure the first node.

[0089] As an example, the advantages of the above method include: better support for different feature sets, application scenarios or terminals.

[0090] As an example, the advantages of the above method include: high flexibility and strong adaptability.

[0091] As one example, the first node is a user equipment.

[0092] As an example, the first node is a relay node.

[0093] As one example, the first node is a terminal.

[0094] As one example, the terminal is a user equipment.

[0095] According to one aspect of this application, the sum of the number of resources occupied by the plurality of feature sets is equal to or less than J.

[0096] According to one aspect of this application, the plurality of feature sets are respectively applicable to different frequency bands, or the plurality of feature sets are respectively applicable to different carriers.

[0097] According to one aspect of this application, the plurality of feature sets belong to the same set of feature sets in a plurality of feature sets; each set of feature sets in the plurality of feature sets includes a feature set applicable to each carrier in a frequency band.

[0098] According to one aspect of this application, the plurality of feature sets belong to the same set of feature sets in a plurality of feature sets; each set of feature sets in the plurality of feature sets includes a feature set applicable to each frequency band in a frequency band combination, the frequency band combination including one or more frequency bands.

[0099] According to one aspect of this application, the first message indicates the total number or maximum total number of resources occupied by the plurality of feature sets respectively, and the total number of resources occupied by the same feature set to which the plurality of feature sets belong is equal to or less than J.

[0100] In the above method, the total number or maximum total number of resources occupied by multiple feature sets may be different, depending on the capabilities of the first node. That is, the first node determines this based on its own capabilities and informs the target recipient of the first message by sending the first message.

[0101] As an example, the advantages of the above method include: better support for different feature sets, application scenarios or terminals.

[0102] As an example, the advantages of the above method include: high flexibility and strong adaptability.

[0103] According to one aspect of this application, the total number of resources occupied by each feature set in the plurality of feature sets is equal to or less than J.

[0104] In the above method, the total number or maximum total number of resources occupied by multiple feature sets is the same. The advantages of this method include: simplified design and reduced reporting overhead.

[0105] According to one aspect of this application, the first message indicates the amount of resources occupied by a feature in one of the plurality of feature sets.

[0106] In the above method, the first node also reports the amount of resources occupied by a feature. The advantages of this method include: the target recipient of the first message (i.e., the second node in this application) can fully understand the resource requirements of the feature, which helps the target recipient of the first message to reasonably schedule or configure the first node.

[0107] As an example, the advantages of the above method include: high flexibility and strong adaptability.

[0108] According to one aspect of this application, it is characterized by comprising:

[0109] Receive the second information block;

[0110] The second information block indicates at least one function, the configuration of which conforms to at least one of the plurality of feature sets.

[0111] In the above method, the configuration of functions follows the feature set supported by the first node, which avoids the first node being configured with an unsupported function and ensures system performance.

[0112] According to one aspect of this application, it is characterized by comprising:

[0113] Send the first information block;

[0114] The first information block is used to indicate the applicable function from the at least one function.

[0115] As an example, in the above method, the first information block indicates the applicable functionality, that is, the functionality that the first node can perform. The advantages of the above method include: preventing the first node from being scheduled or configured with a non-executable function, thus ensuring system performance.

[0116] According to one aspect of this application, it is characterized by comprising:

[0117] Perform inference;

[0118] The reasoning performed by the first node is used to implement a suitable function.

[0119] In the above method, reasoning is used to implement the function, which improves system performance.

[0120] According to one aspect of this application, it is characterized by comprising:

[0121] Receive the second message;

[0122] The first message is sent in response to the receipt of the second message; the second message queries the capabilities of the first node.

[0123] This application discloses a method used in a second node for wireless communication, characterized by comprising:

[0124] Receive a first message; wherein the first message indicates J resources, multiple feature sets, and the amount of resources occupied by each feature set in the multiple feature sets;

[0125] Wherein, J is a positive integer greater than 1; the feature set 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 candidates for features in the feature set include at least two different features.

[0126] According to one aspect of this application, the sum of the number of resources occupied by the plurality of feature sets is equal to or less than J.

[0127] According to one aspect of this application, the plurality of feature sets are respectively applicable to different frequency bands, or the plurality of feature sets are respectively applicable to different carriers.

[0128] According to one aspect of this application, the plurality of feature sets belong to the same set of feature sets in a plurality of feature sets; each set of feature sets in the plurality of feature sets includes a feature set applicable to each carrier in a frequency band.

[0129] According to one aspect of this application, the plurality of feature sets belong to the same set of feature sets in a plurality of feature sets; each set of feature sets in the plurality of feature sets includes a feature set applicable to each frequency band in a frequency band combination, the frequency band combination including one or more frequency bands.

[0130] According to one aspect of this application, the first message indicates the total number or maximum total number of resources occupied by the plurality of feature sets respectively, and the total number of resources occupied by the same feature set to which the plurality of feature sets belong is equal to or less than J.

[0131] According to one aspect of this application, the total number of resources occupied by each feature set in the plurality of feature sets is equal to or less than J.

[0132] According to one aspect of this application, the first message indicates the amount of resources occupied by a feature in one of the plurality of feature sets.

[0133] According to one aspect of this application, it is characterized by comprising:

[0134] Send the second information block;

[0135] The second information block indicates at least one function, the configuration of which conforms to at least one of the plurality of feature sets.

[0136] According to one aspect of this application, it is characterized by comprising:

[0137] Receive the first information block;

[0138] The first information block is used to indicate the applicable function from the at least one function.

[0139] According to one aspect of this application, the first node is the sender of the first message, and the first node performs reasoning; wherein the reasoning performed by the first node is used to implement an applicable function.

[0140] According to one aspect of this application, it is characterized by comprising:

[0141] Send a second message;

[0142] In this context, the first node is the sender of the first message, and the sending of the first message is a response to receiving the second message; the second message queries the capabilities of the first node.

[0143] This application discloses a first node used for wireless communication, characterized in that it comprises:

[0144] A first processor sends a first message; wherein the first message indicates J resources, multiple feature sets, and the amount of resources occupied by each feature set in the multiple feature sets;

[0145] Wherein, J is a positive integer greater than 1; the feature set 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 candidates for features in the feature set include at least two different features.

[0146] This application discloses a second node used for wireless communication, characterized in that it comprises:

[0147] The second processor receives a first message; wherein the first message indicates J resources, multiple feature sets, and the amount of resources occupied by each feature set in the multiple feature sets;

[0148] Wherein, J is a positive integer greater than 1; the feature set 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 candidates for features in the feature set include at least two different features.

[0149] As an example, compared with conventional solutions, this application has the following advantages:

[0150] To better adapt to different feature sets, application scenarios, or terminals;

[0151] Better adaptability to various different abilities;

[0152] High flexibility;

[0153] Highly adaptable;

[0154] The design has been simplified;

[0155] Enhanced reliability and robustness;

[0156] Enhanced overall system performance. Attached Figure Description

[0157] 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:

[0158] Figure 1A shows a flowchart of a first message according to an embodiment of this application;

[0159] Figure 1B shows a flowchart of a first message according to an embodiment of this application;

[0160] Figure 2 shows a schematic diagram of a network architecture according to an embodiment of this application;

[0161] 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;

[0162] Figure 4 shows a schematic diagram of a first communication device and a second communication device according to an embodiment of this application;

[0163] Figure 5A illustrates the transmission between a first node and a second node according to an embodiment of this application;

[0164] Figure 5B illustrates the transmission between a first node and a second node according to an embodiment of this application;

[0165] Figures 6A-6B respectively illustrate schematic diagrams of a first message according to an embodiment of this application;

[0166] Figure 6C illustrates a schematic diagram of the relationship between multiple feature sets and J resources according to an embodiment of this application;

[0167] Figure 7A illustrates a schematic diagram of multiple functions according to an embodiment of this application;

[0168] Figures 7B-7C respectively illustrate schematic diagrams of multiple feature sets according to an embodiment of this application;

[0169] Figures 8A-8D respectively illustrate schematic diagrams of determining the applicable function according to an embodiment of this application;

[0170] Figures 8E-8F respectively illustrate schematic diagrams showing the relationship between multiple feature sets and multiple groups of feature sets according to an embodiment of this application;

[0171] Figure 9A shows a schematic diagram of the transmission conditions of a first information block according to an embodiment of this application;

[0172] Figures 9B-9C respectively illustrate the amount of resources occupied by multiple feature sets according to an embodiment of this application;

[0173] Figure 10A shows a schematic diagram of the transmission conditions of a first information block according to another embodiment of this application;

[0174] Figure 10B shows a schematic diagram of the amount of resources occupied by a feature in a feature set according to an embodiment of the present application;

[0175] Figure 11A illustrates a schematic diagram of the deployment of AI / ML functions in a RAN (Radio Access Network) domain according to an embodiment of this application;

[0176] Figure 11B shows a schematic diagram of the functions implemented by reasoning according to an embodiment of this application;

[0177] Figure 12A shows a schematic diagram of the deployment of AI / ML functions of a UE according to an embodiment of this application;

[0178] Figure 12B shows a schematic diagram of a first encoder according to an embodiment of this application;

[0179] Figure 12C shows a schematic diagram of a first decoder according to an embodiment of this application;

[0180] Figure 13 shows a schematic diagram of a processing system based on artificial intelligence or machine learning according to an embodiment of this application;

[0181] Figure 14A shows a structural block diagram of a processing apparatus for a first node according to an embodiment of the present application;

[0182] Figure 14B shows a structural block diagram of a processing apparatus for a first node according to an embodiment of the present application;

[0183] Figure 15A shows a structural block diagram of a processing apparatus for a second node according to an embodiment of this application;

[0184] Figure 15B shows a structural block diagram of a processing apparatus for a second node according to an embodiment of this application. Detailed Implementation

[0185] 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 1A and the embodiments in Figures 5A-15B, the embodiments in Figure 5A and the embodiments in Figures 6A-15B, etc.

[0186] Example 1A

[0187] Example 1A illustrates a flowchart of a first message according to an embodiment of this application, as shown in Figure 1A. In 100A shown in Figure 1A, 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.

[0188] In Example 1A, the first node sends a first message in step 101A; wherein the first message indicates multiple functions supported by the first node and J resources in the first node, where J is a positive integer greater than 1; the multiple functions are all based on AI, only some of the multiple functions are used for channel information reporting, and any one of the multiple functions occupies at least one of the J resources.

[0189] As an example, the first message includes an RRC message.

[0190] As an example, the first message includes a UECapabilityInformation message.

[0191] As one example, the first message includes the capability information of the first node.

[0192] As an example, J is the maximum number of resources in the first node.

[0193] As one example, the channel information reporting includes beam prediction reporting, CSI prediction, and CSI compression.

[0194] As one example, the channel information reporting includes one or more of beam prediction reporting, CSI prediction, and CSI compression.

[0195] As an example, each of the J resources includes one or more of storage resources, processing resources, energy consumption, and cost.

[0196] As an example, each of the J resources includes a storage resource.

[0197] As an example, each of the J resources includes a processing resource.

[0198] As an example, each of the J resources includes energy consumption.

[0199] As an example, each of the J resources includes a cost.

[0200] As an example, the amount of resources occupied by one of the multiple functions is predefined or configurable.

[0201] Example 1B

[0202] Example 1B illustrates a flowchart of a first message according to an embodiment of this application, as shown in Figure 1B. In Figure 1B, each block represents a step. In particular, the order of the steps in the blocks does not represent a specific temporal sequence between the steps.

[0203] In Example 1B, the first node sends a first message in step 101B; wherein the first message indicates J resources, multiple feature sets, and the number of resources occupied by each feature set in the multiple feature sets; wherein J is a positive integer greater than 1; the feature set 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 candidates for the features in the feature set include at least two different features.

[0204] As an example, the first message includes an RRC message.

[0205] As an example, the first message includes a UECapabilityInformation message.

[0206] As one example, the first message includes the capability information of the first node.

[0207] As an example, the first message is transmitted in PUSCH (Physical Uplink Shared Channel).

[0208] As an example, the first message indicates the J resources by indicating the number J of the resources.

[0209] As an example, each of the J resources includes one or more of storage resources, processing resources, energy consumption, and cost.

[0210] As an example, each of the J resources includes a storage resource.

[0211] As an example, each of the J resources includes a processing resource.

[0212] As an example, each of the J resources includes energy consumption.

[0213] As an example, each of the J resources includes a cost.

[0214] As one embodiment, the processing resources are used for at least one of processing, computation, or inference.

[0215] As an example, the processing resources are used for computation.

[0216] As an example, the processing resources are used for inference.

[0217] As one example, the processing resources include computing resources.

[0218] As one example, the processing resources are used for at least addition and multiplication operations.

[0219] As one example, the processing resources are used for at least convolution operations.

[0220] As one example, the processing resources include an AI processing unit (APU).

[0221] As one embodiment, the processing resource includes a central processing unit (CPU).

[0222] As one example, the processing resources include a GPU (graphics processing unit).

[0223] As one embodiment, the processing resources include general-purpose processing units.

[0224] As one example, the processing resources include general-purpose computing on graphics processing units (GPGPU).

[0225] As an example, the processing resources belong to the AI ​​processing unit (APU).

[0226] As an example, the processing resources belong to the Central Processing Unit (CPU).

[0227] As an example, the processing resources belong to the GPU (graphics processing unit).

[0228] As an example, the processing resources belong to a general-purpose processing unit.

[0229] As an example, the processing resource is a general-purpose graphics processing unit (GPGPU).

[0230] As one embodiment, a processing unit includes one or more of the aforementioned processing resources.

[0231] Regarding the processing unit described in the above embodiments, some typical but non-limiting implementations are described below:

[0232] As one embodiment, the processing unit is used for at least one of processing, calculation, or reasoning.

[0233] As one embodiment, the processing unit is an AI processing unit (APU).

[0234] As one embodiment, the processing unit is a central processing unit (CPU).

[0235] As an example, the processing unit is a GPU (graphics processing unit).

[0236] As one embodiment, the processing unit is a general-purpose processing unit.

[0237] As one embodiment, the processing unit is a general-purpose computing on graphics processing unit (GPGPU).

[0238] As one example, the storage resources are used for storage.

[0239] As one embodiment, the storage resource includes storage units or storage space.

[0240] As one example, the storage resources are used to store some or all of the parameters required for inference.

[0241] As one embodiment, the storage resources are used to store at least one of some or all of the inference intermediate results, or some or all of the inference outputs.

[0242] As one example, the storage resources are used to store some or all of the parameters of the AI ​​model.

[0243] As an example, the storage resources are used to store at least one of some or all of the parameters of the AI ​​model, some or all of the intermediate inference results, or some or all of the inference outputs.

[0244] As one embodiment, the storage resources are used to store 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, or number of feature maps.

[0245] As an example, the storage resources are used to store one or more of the following: convolution kernel, pooling kernel, pooling function, activation function, parameters of the pooling function, or parameters of the activation function.

[0246] As one example, the first message indicates all the functions supported by the first node.

[0247] As an example, the feature set is applicable to one or more functions supported by the first node, or the feature set includes features of one or more functions supported by the first node.

[0248] As an example, the feature set is applicable to one or more functions or the feature set includes features of one or more functions; the configuration of the function complies with the corresponding features in the feature set.

[0249] As an example, the feature set is also referred to as the functions supported by the first node or the function set, which includes one or more functions supported by the first node.

[0250] In some of the above embodiments, the candidate functions include one or more of the following: channel information reporting, uplink radio channel transmission, downlink radio channel reception, uplink radio channel encoding, downlink radio channel decoding, channel estimation on DMRS (Demodulation Reference Signal), and radio link failure prediction.

[0251] It should be noted that receiving (or transmitting) a wireless channel is a common expression in this field, meaning receiving (or transmitting) on ​​a wireless channel, or meaning receiving (or transmitting) signals (e.g., modulation symbols) on a wireless channel; the above expression is beneficial for maintaining consistency with the general expression in this field.

[0252] As a sub-example of the above embodiments, the channel information reporting includes one or more of the following: measured channel information reporting, beam prediction reporting, CSI prediction, and CSI compression.

[0253] As a sub-example of the above embodiments, the channel information reporting includes conventional (i.e., non-AI based or obtained by measurement calculation, such as those supported in 3GPP R18 and earlier versions) channel information reporting and AI-based channel information reporting. The AI-based channel information reporting includes one or more of beam prediction reporting, CSI prediction, and CSI compression.

[0254] As a sub-implementation of the above embodiments, the reception of the downlink wireless channel includes at least one of channel estimation, demodulation, and decoding.

[0255] As a sub-implementation of the above embodiments, the reception of the downlink wireless channel includes decoding.

[0256] As a sub-example of the above embodiments, the reception of the downlink wireless channel includes demodulation and decoding.

[0257] As a sub-example of the above embodiments, the reception of the downlink wireless channel includes the recovery of data carried on the wireless channel.

[0258] As a sub-example of the above embodiments, the transmission of the uplink wireless channel includes encoding and modulation.

[0259] As a sub-implementation of the above embodiments, the transmission of the uplink wireless channel includes the generation of signals on the first wireless channel.

[0260] As a sub-implementation of the above embodiments, the transmission of the uplink wireless channel includes the generation of complex-valued modulation symbols on the wireless channel.

[0261] As a sub-implementation of the above embodiment, the transmission of the uplink wireless channel includes processing the data to be carried on the wireless channel through at least encoding, rate matching, scrambling, modulation, layer mapping, precoding, mapping to resource element, OFDM baseband signal generation, and modulation and upconversion to obtain the transmitted signal on the wireless channel.

[0262] As an example, each of the plurality of feature sets includes one or more features.

[0263] As an example, the first message indicates all the features included in the plurality of feature sets respectively.

[0264] In the above method, the feature set includes one or more features from the candidate features (i.e., the at least two different features). The features included in any two feature sets can be the same or different, depending on the capabilities of the first node. That is, the first node determines and informs the target recipient (e.g., a base station) of the first message by sending the first message based on its own capabilities.

[0265] As an example, the number of resources occupied by each feature set in the plurality of feature sets is a positive integer.

[0266] As an example, the number of resources occupied by each feature set in the plurality of feature sets is a positive integer, and the number of resources occupied by each feature in each feature set in the plurality of feature sets is a positive integer.

[0267] As an example, the amount of resources occupied by each feature set in the plurality of feature sets is a positive integer, and the amount of resources occupied by at least one feature in each feature set in the plurality of feature sets is a positive integer.

[0268] As an example, the number of resources occupied by at least one feature set in the plurality of feature sets is a positive integer, and the number of resources occupied by each feature set in the plurality of feature sets is an integer equal to or greater than 0.

[0269] As an example, the amount of resources occupied by at least one feature set in the plurality of feature sets is a positive integer, and the amount of resources occupied by each feature set in the plurality of feature sets is 0 or a positive integer less than J.

[0270] As an example, the amount of resources occupied by at least one feature set in the plurality of feature sets is a positive integer, and the amount of resources occupied by each feature set in the plurality of feature sets is 0 or a positive integer not greater than J.

[0271] As an example, the amount of resources occupied by at least one feature in the plurality of feature sets is a positive integer, and the amount of resources occupied by each feature in the plurality of feature sets is 0 or a positive integer less than J.

[0272] As an example, the amount of resources occupied by at least one feature in the plurality of feature sets is a positive integer, and the amount of resources occupied by each feature in the plurality of feature sets is 0 or a positive integer not greater than J.

[0273] As an example, the amount of resources occupied by a feature in one feature set of the plurality of feature sets is predefined, either reported by the first node or indicated by the sender of the first message.

[0274] As an example, the feature set includes some or all of the information in a FeatureSetCombination.

[0275] As an example, the feature set includes some or all of the information in a FeatureSetsPerBand.

[0276] As an example, the feature set includes some or all of the information in a FeatureSet.

[0277] As an example, the feature set includes some or all of the information from at least one of FeatureSetDownlink, FeatureSetDownlinkPerCC, FeatureSetUplink, and FeatureSetUplinkPerCC.

[0278] As one embodiment, the feature set includes some or all of the information in a dummy.

[0279] As an example, the Dummy includes one or more of DummyA, DummyB, DummyC, DummyD, and DummyE.

[0280] As an example, the feature set includes at least one inference parameter.

[0281] As an example, the feature set includes at least one of supported encoders, supported decoders, inference parameters, and functions that support inference.

[0282] As an example, the candidate features in the feature set include at least one of supported encoders, supported decoders, inference parameters, and functions that support inference.

[0283] As an example, the candidates for features in the feature set include at least one inference parameter.

[0284] As an example, the candidate features in the feature set include at least two different features among supported encoders, supported decoders, inference parameters, supported inference functions, maximum number of layers, maximum number of downlink RS resources, maximum number of uplink RS resources, supported bandwidth, maximum data rate, maximum modulation order, supported subcarrier spacing, and DMRS type or DMRS pattern.

[0285] In the above method, the candidates for inference parameters include the structure or parameters of the supported AI model, the structure or parameters of the AI-based encoder, the structure or parameters of the AI-based decoder, and functions that support inference.

[0286] AI models can be structured as Transformers, RNNs (Recurrent Neural Networks), CNNs (Conventional Neural Networks), or hybrid models composed of multiple models.

[0287] As an example, the reasoning-supporting functions include at least one of channel information reporting, downlink radio channel reception, uplink radio channel transmission, uplink radio channel encoding, downlink radio channel decoding, channel estimation on DMRS, and radio link failure prediction.

[0288] As an example, the candidate features in the feature set include supported encoders, supported decoders, inference parameters, supported inference functions, maximum number of layers, maximum number of downlink RS resources, maximum number of ports per downlink RS resource, supported codebook, supported number of panels, maximum number of RS resources in each RS resource set, maximum number of uplink RS resources, supported bandwidth, maximum data rate, maximum modulation order, supported subcarrier spacing, support for scells without synchronization signals, dummy, timeDurationForQCL (quasi-colocation), downlink DMRS type, uplink DMRS type, PDSCH (Physical Downlink Shared Channel) processing parameters, PUSCH processing parameters, and single DCI (downlink control information) TRP (Transmit / Receive Point) or SDM (spatial division) TRP (Transmit / Receive Point) or SDM (Spatial Division Point). Multiplexing (space division multiplexing) scheme, single TRP, multiple TRP, multiple DCI multiple TRP, SFN (Single Frequency Network) scheme, multiple TRP-PDCCH, PDCCH repetition, PDSCH repetition, PUSCH repetition, PUCCH repetition, frequency hopping scheme, uplink cancellation scheme, and at least two of the following distinct features.

[0289] In the above method, the DMRS type or DMRS pattern includes at least one of the following: the number of symbols occupied by the DMRS, the symbol positions occupied by the DMRS, the frequency domain density of the DMRS, whether additional DMRS is supported, whether data and DMRS overlap in the time domain, and whether data and DMRS include overlapping REs.

[0290] As an example, the candidates for the encoder include at least one of Polar encoding, LDPC encoding, and AI-based encoders.

[0291] The AI-based approach mentioned in this application includes AI / ML-based approaches, or reasoning-based approaches.

[0292] In the above method, the AI-based encoder can be based on classic models such as Transformer structure, RNN (Recurrent Neural Network), CNN (Conventional Neural Network), or a hybrid model composed of multiple models.

[0293] As an example, the candidates for the decoder include at least one of Polar decoding, LDPC decoding, and AI-based decoders.

[0294] In the above method, the AI-based decoder can be based on classic models such as Transformer structure, RNN (Recurrent Neural Network), CNN (Conventional Neural Network), or a hybrid model composed of multiple models.

[0295] As an example, the number of layers refers to the number of MIMO (Multiple Input Multiple Output) layers.

[0296] As an example, the number of layers refers to the number of transmission layers.

[0297] As an example, the number of downlink RS (Reference Signal) resources includes the number of CSI-RS (Channel State Information Reference Signal) resources.

[0298] As an example, the number of downlink RS resources includes the number of CSI-RS resource sets.

[0299] As an example, the number of downlink RS resources includes the number of TRS (Tracking Reference Signal) resource sets.

[0300] As an example, the number of uplink RS resources includes the number of SRS (Sounding Reference Signal) resources.

[0301] As an example, the number of uplink RS resources includes the number of RS resources in each downlink RS resource set.

[0302] As an example, the number of uplink RS resources includes the number of SRS resource sets.

[0303] As an example, the number of uplink RS resources includes the number of SRS resources in each SRS resource set.

[0304] As an example, the unit of bandwidth is Hz (Hertz), kHz (kilohertz), or MHz (megahertz).

[0305] Considering future ultra-wideband transmission, the unit of bandwidth may also be GHz.

[0306] As an example, the bandwidth is expressed as the number of RBs (Resource Blocks).

[0307] As an example, the unit of the data rate is Mbps (megabits per second).

[0308] Considering future ultra-wideband transmission, the unit of the data rate may also be Gbps or Mbpms (megabits per millisecond).

[0309] 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.

[0310] As an example, the maximum modulation order is one of 64QAM, 256QAM or 1024QAM.

[0311] 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 supported data rate, and the maximum supported modulation order, respectively.

[0312] Example 2

[0313] Example 2 illustrates a schematic diagram of a network architecture according to an embodiment of this application, as shown in Figure 2.

[0314] 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.

[0315] As an example, the first node includes the UE201.

[0316] As one embodiment, the second node includes the node 203.

[0317] As one embodiment, the second node includes the core network 210.

[0318] As one embodiment, the second node includes the node 203 and the core network 210.

[0319] As an example, the wireless link between the UE201 and the node203 includes a cellular link.

[0320] As an example, the storage resources described in this application are in the UE201.

[0321] As an example, the second message is generated in node 203.

[0322] As an example, the target recipient of the second message includes the UE201.

[0323] As an example, the first message is generated in the UE201.

[0324] As an example, the target recipient of the first message includes the node 203.

[0325] As an example, the second information block is generated in node 203.

[0326] As an example, the target recipient of the second information block includes the UE201.

[0327] As an example, the first information block is generated in the UE201.

[0328] As an example, the target recipient of the first information block includes the node 203.

[0329] Example 3

[0330] 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.

[0331] 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.).

[0332] As an example, the wireless protocol architecture in Figure 3 is applicable to the first node.

[0333] As an example, the wireless protocol architecture in Figure 3 is applicable to the second node.

[0334] As an example, the higher layer mentioned in this application refers to the layer above the physical layer.

[0335] As an example, the second message is generated in the RRC sublayer 306.

[0336] As an example, the first message is generated in the RRC sublayer 306.

[0337] As an example, the second information block is generated in the RRC sublayer 306.

[0338] As an example, the first information block is generated in the RRC sublayer 306.

[0339] As an example, the first information block is generated in the MAC sublayer 302 or the MAC sublayer 352.

[0340] As an example, the first information block is generated in the PHY301 or the PHY351.

[0341] As an example, the inference performed by the first node is carried out in the PHY301 or the PHY351.

[0342] As an example, the inference performed by the first node is carried out in at least one of the PHY301, the PHY351, the MAC sublayer 302, or the MAC sublayer 352.

[0343] Example 4

[0344] 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.

[0345] 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.

[0346] 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.

[0347] 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.

[0348] 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.

[0349] 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.

[0350] 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.

[0351] 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: sending a first message; wherein the first message indicates a plurality of functions supported by the first node and J resources in the first node, J being a positive integer greater than 1; the plurality of functions are all based on AI, only a portion of the plurality of functions are used for channel information reporting, and any one of the plurality of functions occupies at least one of the J resources.

[0352] As one embodiment, the second communication device 450 includes: a memory storing a computer-readable instruction program that generates an action when executed by at least one processor, the action including: sending a first message; wherein the first message indicates a plurality of functions supported by the first node and J resources in the first node, J being a positive integer greater than 1; the plurality of functions are all based on AI, only a portion of the plurality of functions are used for channel information reporting, and any one of the plurality of functions occupies at least one of the J resources.

[0353] 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 a first message; wherein a first node is the sender of the first message, the first message indicating multiple functions supported by the first node and J resources in the first node, J being a positive integer greater than 1; the multiple functions are all based on AI, only a portion of the multiple functions are used for channel information reporting, and any one of the multiple functions occupies at least one of the J resources.

[0354] As one embodiment, the first communication device 410 includes: a memory storing a computer-readable instruction program, the computer-readable instruction program generating an action when executed by at least one processor, the action including: receiving a first message; wherein a first node is the sender of the first message, the first message indicating multiple functions supported by the first node and J resources in the first node, J being a positive integer greater than 1; the multiple functions are all based on AI, only a portion of the multiple functions are used for channel information reporting, and any one of the multiple functions occupies at least one of the J resources.

[0355] As an example, the first node in this application includes the second communication device 450.

[0356] As an example, the second node in this application includes the first communication device 410.

[0357] 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 message in this application; 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 message in this application.

[0358] As an example, at least one of {the antenna 452, the transmitter 454, the transmitter processor 468, the multi-antenna transmitter processor 457, the controller / processor 459, the memory 460, and the data source 467} is used to transmit the first message in this application; at least one of {the antenna 420, the receiver 418, the receiver processor 470, the multi-antenna receiver processor 472, the controller / processor 475, and the memory 476} is used to receive the first message in this application.

[0359] 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 information block in this application; 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 information block in this application.

[0360] As an example, at least one of {the antenna 452, the transmitter 454, the transmitter processor 468, the multi-antenna transmitter processor 457, the controller / processor 459, the memory 460, and the data source 467} is used to transmit the first information block in this application; at least one of {the antenna 420, the receiver 418, the receiver processor 470, the multi-antenna receiver processor 472, the controller / processor 475, and the memory 476} is used to receive the first information block in this application.

[0361] As an example, at least one of {the antenna 452, the receiver / transmitter 454, the receiving processor 456, the multi-antenna receiving processor 458, the transmitting processor 468, the multi-antenna transmitting processor 457, the controller / processor 459, the memory 460, and the data source 467} is used for inference performed in the first node of this application;

[0362] As an example, at least one of {the antenna 420, the transmitter / receiver 418, the receiver processor 470, the multi-antenna receiver processor 472, the transmitter processor 416, the multi-antenna transmitter processor 471, the controller / processor 475, and the memory 476} is used for inference performed in the second node of this application.

[0363] 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: transmitting a first message; wherein the first message indicates J resources, a plurality of feature sets, and the number of resources occupied by each feature set in the plurality of feature sets; wherein J is a positive integer greater than 1; the feature sets include 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; candidates for the features in the feature sets include at least two different features.

[0364] As one embodiment, the second communication device 450 includes: a memory storing a computer-readable instruction program that, when executed by at least one processor, produces an action, the action including: sending a first message; wherein the first message indicates J resources, a plurality of feature sets, and the number of resources occupied by each feature set in the plurality of feature sets; wherein J is a positive integer greater than 1; the feature sets include at least one of a maximum number of layers, a maximum number of downlink RS resources, a maximum number of uplink RS resources, a supported bandwidth, a maximum data rate, and a maximum modulation order; and candidates for features in the feature sets include at least two different features.

[0365] 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 a first message; wherein the first message indicates J resources, a plurality of feature sets, and the number of resources occupied by each feature set in the plurality of feature sets; wherein J is a positive integer greater than 1; the feature sets include 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; candidates for features in the feature sets include at least two different features.

[0366] As one embodiment, the first communication device 410 includes: a memory storing a computer-readable instruction program that, when executed by at least one processor, produces an action, the action including: receiving a first message; wherein the first message indicates J resources, a plurality of feature sets, and the number of resources occupied by each feature set in the plurality of feature sets; wherein J is a positive integer greater than 1; the feature sets include at least one of a maximum number of layers, a maximum number of downlink RS resources, a maximum number of uplink RS resources, a supported bandwidth, a maximum data rate, and a maximum modulation order; and candidates for features in the feature sets include at least two different features.

[0367] As an example, the first node in this application includes the second communication device 450.

[0368] As an example, the second node in this application includes the first communication device 410.

[0369] 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 message in this application; 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 message in this application.

[0370] As an example, at least one of {the antenna 452, the transmitter 454, the transmitter processor 468, the multi-antenna transmitter processor 457, the controller / processor 459, the memory 460, and the data source 467} is used to transmit the first message in this application; at least one of {the antenna 420, the receiver 418, the receiver processor 470, the multi-antenna receiver processor 472, the controller / processor 475, and the memory 476} is used to receive the first message in this application.

[0371] 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 information block in this application; 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 information block in this application.

[0372] As an example, at least one of {the antenna 452, the transmitter 454, the transmitter processor 468, the multi-antenna transmitter processor 457, the controller / processor 459, the memory 460, and the data source 467} is used to transmit the first information block in this application; at least one of {the antenna 420, the receiver 418, the receiver processor 470, the multi-antenna receiver processor 472, the controller / processor 475, and the memory 476} is used to receive the first information block in this application.

[0373] As an example, at least one of {the antenna 452, the receiver / transmitter 454, the receiver processor 456, the multi-antenna receiver processor 458, the transmitter processor 468, the multi-antenna transmitter processor 457, the controller / processor 459, the memory 460, and the data source 467} is used for inference performed in the first node of this application.

[0374] As an example, at least one of {the antenna 420, the transmitter / receiver 418, the receiver processor 470, the multi-antenna receiver processor 472, the transmitter processor 416, the multi-antenna transmitter processor 471, the controller / processor 475, and the memory 476} is used for inference performed in the second node of this application.

[0375] Example 5A

[0376] Example 5A illustrates a flowchart of a transmission between a first node and a second node according to an embodiment of this application, as shown in Figure 5A. In Figure 5A, the second node N1A and the first node U1A are communication nodes transmitting via an air interface. In Figure 5A, the steps in blocks F51A to F54A are optional.

[0377] For the second node N1A, a second message is sent in step S511A; a first message is received in step S512A; a second information block is sent in step S513A; and a first information block is received in step S514A.

[0378] For the first node U1A, the second message is received in step S521A; the first message is sent in step S522A; the second information block is received in step S523A; the first information block is sent in step S524A; and inference is performed in step S525A.

[0379] In Embodiment 5A, the first message indicates multiple functions supported by the first node and J resources in the first node, where J is a positive integer greater than 1; the multiple functions are all based on AI, only a portion of the multiple functions are used for channel information reporting, and each of the multiple functions occupies at least one of the J resources; the first information block is used to indicate N1 applicable functions of the first node from N functions, where N functions are all based on AI, N ​​is a positive integer greater than 1, and N1 is a positive integer not greater than N; the number of resources occupied by each of the N1 applicable functions is equal to or less than J. The inference performed by the first node is used to implement at least one of the N1 applicable functions. The second message indicates that the first node reports the supported functions, or the second message queries the capabilities of the first node.

[0380] As one example, the second information block indicates the N functions.

[0381] As one embodiment, the second information block instructs the first node to provide the applicable functions of the first node.

[0382] As one embodiment, the second information block instructs the first node to report UAI (UE Assistance Information).

[0383] As one embodiment, the second information block instructing the first node to provide the applicable functions of the first node includes: the second information block instructing the first node to be allowed to provide the applicable functions of the first node.

[0384] As one embodiment, the second information block instructing the first node to perform UAI reporting includes: the second information block instructing the first node to perform UAI reporting.

[0385] As an example, the first node U1A is the first node in this application.

[0386] As an example, the second node N1A is the second node in this application.

[0387] As one embodiment, the air interface between the second node N1A and the first node U1A includes a wireless interface between the base station equipment and the user equipment.

[0388] As one embodiment, the air interface between the second node N1A and the first node U1A includes a wireless interface between the relay node device and the user equipment.

[0389] As one embodiment, the air interface between the second node N1A and the first node U1A includes a wireless interface between user equipment and user equipment.

[0390] As one example, the second node N1A is the serving cell sustaining base station of the first node U1A.

[0391] As one embodiment, the first information block is carried by higher-layer signaling.

[0392] As an example, the first information block is carried by an RRC message.

[0393] As an example, the first information block belongs to UAI (UE Assistance Information).

[0394] As an example, the first information block belongs to the UEAssistanceInformation message.

[0395] As one example, the first information block includes UAI.

[0396] As one embodiment, the first information block includes a UEAssistanceInformation message.

[0397] As an example, the UAI report is the reporting of the UEAssistanceInformation message.

[0398] As an example, the first information block includes a MAC CE.

[0399] As one embodiment, the first information block includes physical layer information.

[0400] As one embodiment, the first information block includes uplink control information.

[0401] As one example, the first information block is transmitted over a physical channel.

[0402] As an example, the first information block is transmitted on PUSCH (Physical Uplink Shared Channel).

[0403] As an example, the first information block is transmitted on PUCCH (Physical Uplink Control Channel).

[0404] As an example, the first information block indicates the index or identifier of the N1 applicable functions.

[0405] As an example, the first information block indicates the index of the N1 applicable functions among the N functions.

[0406] Typically, the N functions belong to the multiple functions supported by the first node.

[0407] As an example, the N functions are configured by the serving cell of the first node.

[0408] As an example, the N functions are configured by the serving base station of the first node.

[0409] As an example, the second information block in this application configures or indicates the N functions.

[0410] As an example, the N functions are configured by RRC signaling.

[0411] As an example, the N functions are determined by the first node itself.

[0412] As an example, the N1 applicable functions are all the functions among the N functions that are applicable to the first node.

[0413] As an example, the N1 applicable functions are only a portion of all the functions applicable to the first node among the N functions.

[0414] As an example, the first information block is used to directly indicate the N1 applicable functions of the first node from N functions.

[0415] As an example, the first information block indicates at least one function that is not applicable to the first node, and the N1 applicable functions of the first node include the N1 functions other than the at least one function that is not applicable to the first node.

[0416] As one example, the second message triggers the first message.

[0417] As one embodiment, the second message instructs the first node to send the first message.

[0418] As one embodiment, the second message includes an RRC message, and the first message includes an RRC message.

[0419] As one embodiment, the second message instructs the first node to report the supported functionalities, and the first message includes the supported functionalities of the first node.

[0420] As one embodiment, the second message includes a UECapabilityEnqiry message, and the first message includes a UECapabilityInformation message, the UECapabilityInformation message including the functions supported by the first node.

[0421] As one embodiment, the first message includes functions supported by the first node, and the N functions are all functions supported by the first node.

[0422] As an example, the N functions are configured by higher-level signaling.

[0423] As an example, the first message indicates the N functions.

[0424] As an example, the first message indicates the N functions, and the second information block indicates M first-class identifiers, where M is a positive integer greater than 1; any one of the N1 applicable functions corresponds to one of the M first-class identifiers.

[0425] As an example, the first node determines the N functions on its own, and the second information block indicates M first-class identifiers, where M is a positive integer greater than 1; any one of the N1 applicable functions corresponds to one of the M first-class identifiers.

[0426] As one example, the indication of the second information block depends on the functions supported by the first node.

[0427] As one embodiment, the second message is used to query the capabilities of the first node, and the first message includes the capabilities of the first node.

[0428] As one embodiment, the second message includes a UECapabilityEnqiry message, the first message includes a UECapabilityInformation message, the second information block belongs to an RRCReconfiguration message, and the first information block belongs to a UEAssistanceInformation message.

[0429] As one embodiment, the second message includes a UECapabilityEnqiry message, the first message includes a UECapabilityInformation message, the second information block belongs to IE OtherConfig, and the first information block belongs to a UEAssistanceInformation message.

[0430] As one embodiment, the second information block is carried by higher-layer signaling.

[0431] As an example, the second information block belongs to an RRC message.

[0432] As an example, the second information block belongs to the RRCReconfiguration message.

[0433] As an example, the second information block belongs to IE OtherConfig.

[0434] As one embodiment, the second information block includes an RRCReconfiguration message.

[0435] As one example, the second information block includes IE OtherConfig.

[0436] As one embodiment, the second information block indicates M first-class identifiers, where M is a positive integer greater than 1.

[0437] As an example, the second information block indicates M first-class identifiers, where M is a positive integer greater than 1; any one of the N functions corresponds to one of the M first-class identifiers.

[0438] As an example, the second information block indicates M first-class identifiers, where M is a positive integer greater than 1; any one of the N1 applicable functions corresponds to one of the M first-class identifiers.

[0439] As one embodiment, the second information block indicates the N functions; any one of the N functions includes a first type of identifier.

[0440] As an example, the N1 applicable functions include one or more of channel information reporting, wireless channel reception, wireless channel coding, channel estimation on DMRS, and wireless link failure prediction; the reasoning performed by the first node is used to implement at least one of the N1 applicable functions.

[0441] As an example, channel information reporting is one of the N1 applicable functions, and the reasoning performed by the first node is used to generate the channel information.

[0442] As a sub-implementation of the above embodiment, the second node recovers the channel information based on inference.

[0443] As an example, receiving a wireless channel is one of the N1 applicable functions, and the reasoning performed by the first node is for receiving the wireless channel.

[0444] As an example, the encoding of the wireless channel is one of the N1 applicable functions, and the reasoning performed by the first node is used for the encoding of the wireless channel.

[0445] As a sub-implementation of the above embodiment, the second node receives the wireless signal based on inference.

[0446] As an example, channel estimation on DMRS is one of the N1 applicable functions, and the inference performed by the first node is used for channel estimation on DMRS.

[0447] As an example, wireless link failure prediction is one of the N1 applicable functions, and the inference performed by the first node is used to predict the wireless link failure.

[0448] As an example, the reasoning is based on training or AI.

[0449] As an example, the reasoning includes AI (Artificial Intelligence) inference.

[0450] As an example, the reasoning model is obtained through training.

[0451] As an example, the training of the inference in the first node is performed by the first node.

[0452] As an example, the training of the inference in the first node is performed by the second node.

[0453] As an example, the training of the inference in the first node is performed by the core network.

[0454] As an example, the training of the inference in the first node is performed by an AI training producer.

[0455] As an example, the training of the inference in the first node is performed by the MDA (Management Data Analytics Function).

[0456] As an example, the training of the inference in the first node is performed by the MDA function located in the first node.

[0457] As an example, the training of the inference in the first node is performed by the MDA function located in the second node.

[0458] As an example, the training of the inference in the first node is performed by NWDAF (Network Data Analytics Function).

[0459] As an example, the training of the inference in the first node is performed by the MDAS (Management Data Analytics Service) producer.

[0460] As an example, the training of the inference in the first node is performed by the MnS (Management Service) producer.

[0461] As an example, the inference in the first node needs to be deployed.

[0462] As an example, the reasoning in the first node is obtained by loading.

[0463] As an example, the inference in the first node is obtained from the serving cell of the first node.

[0464] As an example, the inference in the first node is obtained from the sustaining base station of the serving cell of the first node.

[0465] As an example, the first node deploys the inference.

[0466] As an example, the inference does not require deployment.

[0467] As an example, the inference is obtained from the core network.

[0468] As an example, the reasoning is based on artificial intelligence or machine learning.

[0469] As an example, the reasoning is based on a neural network.

[0470] As an example, the inference is based on CNN (Conventional Neural Networks).

[0471] As one example, the inference includes preprocessing.

[0472] As one example, the reasoning includes post-processing.

[0473] As one example, the post-processing includes DFT.

[0474] As one example, the post-processing includes quantization.

[0475] As an example, the post-processing 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, and frequency domain to time domain transformation.

[0476] As one example, the post-processing includes truncation and / or padding.

[0477] As an example, the inference includes one or more of convolution, pooling, cascading, and activation.

[0478] As one example, the inference includes a fully connected layer.

[0479] As one example, the inference includes a pooling layer.

[0480] As an example, the inference includes at least one convolutional layer.

[0481] As one example, the reasoning includes at least one coding layer.

[0482] As an example, an encoding layer includes at least one convolutional layer and one pooling layer.

[0483] As an example, in a convolutional layer, at least one convolutional kernel is used to convolve the input to generate a corresponding feature map, and at least one feature map output by the convolutional layer is reshaped into a vector and input to a fully connected layer; the fully connected layer transforms the vector into an output.

[0484] As an example, some or all of the following parameters in the inference—convolution kernel size, number of convolutional layers, convolution stride, pooling kernel size, pooling kernel stride, pooling function, activation function, and number of feature maps—are obtained through training.

[0485] As an example, some or all of the convolution kernel, pooling kernel, pooling function, activation function, parameters of the pooling function, and parameters of the activation function in the inference are obtained through training.

[0486] As one example, the AI ​​includes ML (Machine Learning).

[0487] As an example, the AI ​​includes AI and ML.

[0488] As one example, the AI ​​includes AI or ML.

[0489] As an example, the preprocessing includes one or more of the following: quantization, DFT (Discrete Fourier Transform), matrix decomposition, matrix transformation or projection, spatial-to-angular-domain transformation, angular-to-spatial-domain transformation, frequency-to-time-domain transformation, time-to-frequency-domain transformation, truncation, padding, mapping, or labeling.

[0490] As an example, the preprocessing includes DFT (Discrete Fourier Transform).

[0491] As one example, the preprocessing includes one or more of matrix decomposition, matrix transformation, or projection.

[0492] As an example, the preprocessing includes one or more of the following: quantization, spatial-to-angular-domain transformation, angular-to-spatial-domain transformation, frequency-to-time-domain transformation, or time-to-frequency-domain transformation.

[0493] As one example, the preprocessing includes truncation and / or padding.

[0494] As one example, the preprocessing includes mapping.

[0495] As one example, the preprocessing includes mapping to vectors.

[0496] As one example, the preprocessing includes labeling.

[0497] As an example, the label refers to a mark made with a label.

[0498] Example 5B

[0499] Example 5B illustrates a flowchart of a transmission between a first node and a second node according to an embodiment of this application, as shown in Figure 5B. In Figure 5B, the second node N1B and the first node U1B are communication nodes transmitting via an air interface. In Figure 5B, the steps in blocks F51B to F54B are optional.

[0500] For the second node N1B, a second message is sent in step S511B; a first message is received in step S512B; a second information block is sent in step S513B; and a first information block is received in step S514B.

[0501] For the first node U1B, the second message is received in step S521B; the first message is sent in step S522B; the second information block is received in step S523B; the first information block is sent in step S524B; and inference is performed in step S525B.

[0502] In Example 5B, the sending of the first message is a response to receiving the second message; the second message queries the capabilities of the first node. The first message indicates J resources, multiple feature sets, and the number of resources occupied by each feature set in the multiple feature sets; J is a positive integer greater than 1; the feature sets include at least one of 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; candidates for the features in the feature sets include at least two distinct features. The second information block indicates at least one function, the configuration of which conforms to at least one of the multiple feature sets. The first information block is used to indicate an applicable function from the at least one function. The reasoning performed by the first node is used to implement an applicable function.

[0503] As an example, the first node U1 is the first node in this application.

[0504] As an example, the second node N1B is the second node in this application.

[0505] As one embodiment, the air interface between the second node N1B and the first node U1B includes a wireless interface between the base station equipment and the user equipment.

[0506] As one embodiment, the air interface between the second node N1B and the first node U1B includes a wireless interface between the relay node device and the user equipment.

[0507] As one embodiment, the air interface between the second node N1B and the first node U1B includes a wireless interface between user equipment and user equipment.

[0508] As one example, the second node N1B is the serving cell sustaining base station of the first node U1B.

[0509] As an example, block F51B exists, where the sending of the first message is a response to receiving the second message.

[0510] As an example, block F51B is not present, and the sending of the first message is triggered by an event, or by the second node sending other signaling that triggers or instructs the first node to send the first message.

[0511] As one embodiment, the second information block is carried by higher-layer signaling.

[0512] As an example, the second information block belongs to an RRC message.

[0513] As an example, the second information block belongs to the RRCReconfiguration message.

[0514] As an example, the second information block belongs to IE OtherConfig.

[0515] As one embodiment, the second information block includes an RRCReconfiguration message.

[0516] As one example, the second information block includes IE OtherConfig.

[0517] As one embodiment, the second information block indicates at least one function, which is associated with a first type of identifier.

[0518] As one embodiment, the second information block indicates at least one function, the configuration information of which includes a first type of identifier.

[0519] As one embodiment, the second information block indicates at least one function, which is identified by a first type of identifier.

[0520] As one embodiment, the second information block indicates at least one function, the function being based on reasoning, and the reasoning on which the function is based is associated with a first type of identifier.

[0521] As one example, the second information block is transmitted on the PDSCH.

[0522] As one embodiment, the second information block indicates at least one function including: the second information block configures at least one function.

[0523] As one embodiment, the second information block includes some or all of the information in one or more RRC IEs (Information Elements).

[0524] As one embodiment, the second information block instructs the first node to provide the applicable functions of the first node.

[0525] As one embodiment, the second information block instructs the first node to report UAI (UE Assistance Information).

[0526] As one embodiment, the second information block instructing the first node to provide the applicable functions of the first node includes: the second information block instructing the first node to be allowed to provide the applicable functions of the first node.

[0527] As one embodiment, the second information block instructing the first node to perform UAI reporting includes: the second information block instructing the first node to perform UAI reporting.

[0528] As an example, the configuration of the at least one function complies with capabilities from at least one of the plurality of feature sets.

[0529] As an example, the at least one function includes only one function whose configuration conforms to the capability of a feature set from the plurality of feature sets.

[0530] As an example, the at least one function includes multiple functions, the configuration of which conforms to the capability of a feature set from the multiple feature sets.

[0531] As an example, the at least one function includes multiple functions, the configuration of which are respectively compatible with capabilities from the multiple feature sets.

[0532] As an example, the at least one function includes one or more of the following: channel information reporting, downlink radio channel reception, uplink radio channel coding, channel estimation on DMRS, and radio link failure prediction.

[0533] As an example, at least one of the functions is based on AI or reasoning.

[0534] As one example, the AI-based functionality includes: the functionality is implemented based on reasoning.

[0535] As one example, the AI-based function includes: the inference parameters of the function are obtained through training.

[0536] As one example, the AI-based function includes: the function is associated with a first type of identifier.

[0537] As one example, the AI-based function includes: the configuration information of the function includes a first type of identifier.

[0538] As one example, the AI-based function includes: the function being identified by a first type of identifier.

[0539] As one embodiment, the AI-based function includes: the function is based on reasoning, and the reasoning on which the function is based is associated with a first type of identifier.

[0540] As an example, the first type of identifier is a non-negative integer.

[0541] As an example, the first type of identifier is a string.

[0542] As an example, the first type of identifier is an associated identifier (associated ID).

[0543] As an example, the inference association with the first type of identifier includes: the parameters of the inference include the first type of identifier.

[0544] As an example, the inference association with the first type of identifier includes: the parameters of the inference are identified by the first type of identifier.

[0545] As an example, the first type of identifier associated with the inference includes: the AI ​​model used for the inference is identified by the first type of identifier.

[0546] As an example, the inference association with the first type of identifier includes: the AI ​​entity of the inference is identified by the first type of identifier.

[0547] As an example, the advantages of the above method include simplifying the design and unifying the understanding of different inference parameters / AI models / AI entities across multiple nodes by identifying inference parameters / AI models / AI entities through a first type of identifier.

[0548] As one embodiment, the inference association first type identifier includes: the first type identifier is used to identify or indicate a set of RS resources, and the measurement of the set of RS resources is used to obtain the training dataset for the inference.

[0549] As an example, the inference association first type of identifier includes: the dataset used to train the parameters of the inference is identified by the first type of identifier.

[0550] As an example, the advantages of the above method include identifying the AI ​​model / inference parameters trained by the AI ​​by identifying the RS resource set or training dataset used for AI training, establishing consensus among different AI model / inference parameters, and further simplifying the design.

[0551] In the above method, how a function's configuration conforms to a set of features is implementation-dependent, i.e., determined by the hardware vendor of the second node. Several typical but non-limiting implementations are described below:

[0552] As an example, the feature set includes a maximum number of layers; the configuration of the function cannot exceed the maximum number of layers in the feature set.

[0553] As an example, the feature set includes the maximum number of downlink RS resources; the configuration of the function cannot exceed the maximum number of downlink RS resources in the feature set.

[0554] As an example, the feature set includes the maximum number of uplink RS resources; the configuration of the function cannot exceed the maximum number of uplink RS resources in the feature set.

[0555] As an example, the feature set includes supported bandwidths; the configuration of the function cannot exceed the supported bandwidths in the feature set.

[0556] As an example, the feature set includes a maximum data rate; the configuration of the function cannot exceed the maximum data rate in the feature set.

[0557] As an example, the feature set includes at least one of the maximum modulation orders; the configuration of the function cannot exceed the maximum modulation order in the feature set.

[0558] As one embodiment, the feature set includes support for overlapping REs in the data and DMRS; the configuration of the function can be that the data and DMRS include overlapping REs, or the data and DMRS can exclude overlapping (i.e., orthogonal) REs.

[0559] As one example, the feature set includes an AI-based encoder, and the configuration of the function can be either an AI-based encoder or a traditional encoder.

[0560] As one example, the feature set includes an AI-based decoder, and the configuration of the function can be based on either an AI-based decoder or a traditional decoder.

[0561] As an example, the feature set includes functions that support reasoning, and these functions can be configured as the functions that support reasoning.

[0562] As one embodiment, the first information block is carried by higher-layer signaling.

[0563] As an example, the first information block is carried by an RRC message.

[0564] As an example, the first information block belongs to UAI (UE Assistance Information).

[0565] As an example, the first information block belongs to the UEAssistanceInformation message.

[0566] As one example, the first information block includes UAI.

[0567] As one embodiment, the first information block includes a UEAssistanceInformation message.

[0568] As an example, the UAI report is the reporting of the UEAssistanceInformation message.

[0569] As an example, the first information block includes a MAC CE.

[0570] As one embodiment, the first information block includes physical layer information.

[0571] As one embodiment, the first information block includes uplink control information.

[0572] As one example, the first information block is transmitted over a physical channel.

[0573] As an example, the first information block is transmitted on PUSCH (Physical Uplink Shared Channel).

[0574] As an example, the first information block is transmitted on PUCCH (Physical Uplink Control Channel).

[0575] As one embodiment, the first information block indicates an index or identifier of the applicable function.

[0576] As an example, the first information block indicates the index of the applicable function in the at least one function.

[0577] As an example, the first information block is used to directly indicate the applicable function of the first node from the at least one function.

[0578] As an example, the first information block is used to indirectly indicate the applicable function of the first node from the at least one function.

[0579] As one embodiment, the first information block is used to indicate an inapplicable function from the at least one function, and the applicable functions of the first node include functions other than the inapplicable functions among the at least one function.

[0580] In the above method, whether a function is applicable to the first node needs to take into account whether the first node has sufficient resources (such as storage resources, processing resources, energy consumption, or cost, etc.). An applicable function requires the first node to have sufficient storage resources, processing resources, battery or power, etc.

[0581] As one example, the second message instructs the first node to report the supported functions.

[0582] As one embodiment, the second message instructs the first node to report the supported feature set.

[0583] As one example, the second message triggers the first message.

[0584] As one embodiment, the second message instructs the first node to send the first message.

[0585] As one embodiment, the second message includes an RRC message, and the first message includes an RRC message.

[0586] As one embodiment, the second message instructs the first node to report the set of features or supported functionalities supported by the first node, and the first message includes the set of features or the supported functionalities supported by the first node.

[0587] As one embodiment, the second message includes a UECapabilityEnqiry message, and the first message includes a UECapabilityInformation message, wherein the UECapabilityInformation message includes the feature set or supported functions supported by the first node.

[0588] As one embodiment, the first message includes the feature set or supported functions supported by the first node.

[0589] As one embodiment, the second message is used to query the capabilities of the first node. The first message includes capability information of the first node, which includes a set of features or functions supported by the first node.

[0590] As one embodiment, the second message includes a UECapabilityEnqiry message, the first message includes a UECapabilityInformation message, the second information block belongs to an RRCReconfiguration message, and the first information block belongs to a UEAssistanceInformation message.

[0591] As one embodiment, the second message includes a UECapabilityEnqiry message, the first message includes a UECapabilityInformation message, the second information block belongs to IE OtherConfig, and the first information block belongs to a UEAssistanceInformation message.

[0592] Examples 6A-6B

[0593] Examples 6A-6B illustrate schematic diagrams of a first message according to an embodiment of this application, as shown in Figures 6A-6B respectively.

[0594] In embodiment 6A, the first message indicates multiple functions supported by the first node and J resources in the first node.

[0595] As an example, the first message indicates the J resources in the first node by indicating the number J of the resources.

[0596] In embodiment 6B, the first message indicates the amount of resources occupied by one of the plurality of functions. In Figure 6B, function #1, function #2, ... are the plurality of functions supported by the first node; positive integer #1 is the amount of resources occupied by function #1, positive integer #2 is the amount of resources occupied by function #2, ...; the first message indicates positive integer #1, positive integer #2, ...

[0597] As an example, the first message indicates the maximum amount of resources occupied by one of the plurality of functions.

[0598] Example 6C

[0599] Example 6C illustrates a schematic diagram of the relationship between multiple feature sets and J resources according to an embodiment of this application; as shown in Figure 6C.

[0600] In embodiment 6C, the sum of the number of resources occupied by the plurality of feature sets is equal to or less than J. In Figure 6C, feature set #1 occupies J1 of the J resources in this application, and feature set #2 occupies J2 of the J resources, where J1 and J2 are non-negative integers, and at least one of J1 and J2 is greater than 0.

[0601] As an example, the number of resources occupied by at least one feature set in the plurality of feature sets is a positive integer, the number of resources occupied by each feature set in the plurality of feature sets is 0 or a positive integer less than J; the sum of the number of resources occupied by the plurality of feature sets is equal to or less than J.

[0602] As an example, the number of resources occupied by at least one feature set in the plurality of feature sets is a positive integer, the number of resources occupied by each feature set in the plurality of feature sets is 0 or a positive integer not greater than J; the sum of the number of resources occupied by the plurality of feature sets is equal to or less than J.

[0603] As an example, the amount of resources occupied by at least one feature in the plurality of feature sets is a positive integer, the amount of resources occupied by each feature in the plurality of feature sets is 0 or a positive integer less than J; the sum of the amount of resources occupied by the plurality of feature sets is equal to or less than J.

[0604] As an example, the number of resources occupied by at least one feature in the plurality of feature sets is a positive integer, the number of resources occupied by each feature in the plurality of feature sets is 0 or a positive integer not greater than J; the sum of the number of resources occupied by the plurality of feature sets is equal to or less than J.

[0605] As an example, the number of resources occupied by each of the plurality of feature sets is a positive integer less than J, and the sum of the number of resources occupied by each of the plurality of feature sets is equal to or less than J.

[0606] Example 7A

[0607] Example 7A illustrates a schematic diagram of multiple functions according to one embodiment of the present application; as shown in Figure 7A.

[0608] In Embodiment 7A, only a portion of the plurality of functions are used for channel information reporting; the plurality of functions include one or more of the following: receiving a wireless channel, coding a wireless channel, channel estimation on a DMRS (Demodulation Reference Signal), and predicting wireless link failure.

[0609] As an example, only some of the multiple functions are used for channel information reporting, and the multiple functions also include one or more of the following: receiving wireless channels, encoding wireless channels, channel estimation on DMRS, and predicting wireless link failures.

[0610] As an example, channel information reporting is one of the many functions.

[0611] As an example, channel information reporting is one of the multiple functions. The reception of X4 CSI-RS resources occupies X5 of the J resources, where X4 is a positive integer and X5 is a positive integer.

[0612] As an example, channel information reporting is one of the multiple functions, and the reception of no more than X4 CSI-RS resources occupies X5 of the J resources, where X4 is a positive integer and X5 is a positive integer.

[0613] As an example, the advantages of the above method include: the acquisition and reporting of channel information based on AI / ML improves the accuracy of channel information estimation and enhances system performance.

[0614] As one example, receiving wireless channels is one of the many functions.

[0615] As an example, receiving wireless channels is one of the multiple functions. Receiving wireless channels of X1 layers or X2 RBs (Resource blocks) occupies X3 of the J resources, where X1 is a positive integer, X2 is a positive integer, and X3 is a positive integer.

[0616] As an example, receiving a wireless channel is one of the multiple functions. Receiving a wireless channel of no more than X1 layers or no more than X2 RBs occupies X3 of the J resources, where X1 is a positive integer, X2 is a positive integer, and X3 is a positive integer.

[0617] As an example, receiving the wireless channel includes at least one of channel estimation, demodulation, and decoding.

[0618] As one example, receiving the wireless channel includes decoding.

[0619] As one example, receiving the wireless channel includes demodulation and decoding.

[0620] As one embodiment, receiving the wireless channel includes recovering the data carried on the wireless channel.

[0621] In the above method, the specific algorithm for receiving the wireless channel can be implementation-dependent, i.e., determined by the hardware equipment vendor of the first node; several typical but non-limiting implementation methods are described below:

[0622] In one implementation, the first node demodulates the signal on the wireless channel and inputs it into an AI model for decoding. The output of the AI ​​model is then used to recover the data carried on the wireless channel.

[0623] In one implementation, the first node estimates the channel matrix based on the DMRS of the wireless channel, inputs the signal on the wireless channel and the estimated channel matrix into an AI model, and the output of the AI ​​model is used to recover the data carried on the wireless channel.

[0624] In another implementation, the REs occupied by the DMRS of the wireless channel and the REs carrying data in the wireless channel overlap; the first node performs channel estimation based on AI according to the signal on the RE occupied by the DMRS of the wireless channel; then, the first node demultiplexes the DMRS and the data, and inputs the signal of the demultiplexed data carrying data into the AI ​​model, and the output of the AI ​​model is used to recover the data carried on the wireless channel.

[0625] The structure and parameters of the AI ​​model in the above embodiments are known to the first node. For example, they may be obtained by downloading from a network device, or they may be specified in a standard, or they may be implementation-related to the first node (i.e., determined by the hardware device vendor of the receiver of the wireless channel).

[0626] As an example, the advantages of the above method include: improving the reception (e.g., decoding) performance of the wireless channel based on AI / ML, improving the reliability of the wireless channel, and improving transmission efficiency.

[0627] As an example, the encoding of the wireless channel is one of the many functions.

[0628] As an example, the advantages of the above method include: encoding the wireless channel based on AI / ML (channel coding, or source-channel joint coding) improves transmission reliability and transmission efficiency.

[0629] As an example, channel estimation on DMRS is one of the many functions mentioned.

[0630] As an example, the advantages of the above method include: improving the channel estimation accuracy on DMRS based on AI / ML, which can increase the transmission capacity of wireless channels (such as PDSCH, PUSCH, PDCCH, PUCCH, etc.) and reduce data interference to DMRS.

[0631] As an example, wireless link failure prediction is one of the many functions mentioned.

[0632] As an example, the benefits of the above method include: predicting wireless link failures based on AI / ML, reducing or avoiding actual wireless link failures, improving the reliability of wireless links, and enhancing system performance.

[0633] As one embodiment, receiving the wireless channel includes decoding the wireless channel.

[0634] As an example, the decoding of the wireless channel includes the decoding of the physical channel through which the data is transmitted.

[0635] As an example, the decoding of the wireless channel includes the decoding of PDSCH (Physical Downlink Shared Channel).

[0636] As one embodiment, the decoding of the wireless channel includes the decoding of the physical channel that transmits control information.

[0637] As an example, the decoding of the wireless channel includes the decoding of the PDCCH (Physical Downlink Control Channel).

[0638] As an example, the encoding of the wireless channel includes the encoding of the physical channel through which the data is transmitted.

[0639] As an example, the encoding of the wireless channel includes the encoding of PUSCH (Physical Uplink Shared Channel).

[0640] As one embodiment, the encoding of the wireless channel includes the encoding of the physical channel for transmitting control information.

[0641] As an example, the encoding of the wireless channel includes the encoding of PUCCH (Physical Uplink Control Channel).

[0642] As an example, the radio link failure prediction includes RLF (Radio Link Failure) prediction.

[0643] As one embodiment, the wireless link failure prediction includes beam failure prediction.

[0644] As an example, the multiple functions are all AI-based, including: each of the multiple functions is implemented based on inference.

[0645] As an example, the plurality of functions are all AI-based, including the fact that the inference parameters of each of the plurality of functions are obtained through training.

[0646] As an example, the plurality of functions are all AI-based, including: each of the plurality of functions is associated with a first type of identifier.

[0647] As an example, the plurality of functions are all AI-based and include: each of the plurality of functions includes a first type of identifier.

[0648] As an example, the plurality of functions are all AI-based, including: each of the plurality of functions is identified by a first type of identifier.

[0649] As an example, the plurality of functions are all AI-based, including: each of the plurality of functions is based on inference, and the inference on which each of the plurality of functions is based is associated with a first type of identifier.

[0650] As an example, the first type of identifier is a non-negative integer.

[0651] As an example, the first type of identifier is a string.

[0652] As an example, the first type of identifier is an associated identifier (associated ID).

[0653] As an example, the inference association with the first type of identifier includes: the parameters of the inference include the first type of identifier.

[0654] As an example, the inference association with the first type of identifier includes: the parameters of the inference are identified by the first type of identifier.

[0655] As an example, the first type of identifier associated with the inference includes: the AI ​​model used for the inference is identified by the first type of identifier.

[0656] As an example, the inference association with the first type of identifier includes: the AI ​​entity of the inference is identified by the first type of identifier.

[0657] As an example, the advantages of the above method include simplifying the design and unifying the understanding of different inference parameters / AI models / AI entities across multiple nodes by identifying inference parameters / AI models / AI entities through a first type of identifier.

[0658] As one embodiment, the inference association first type identifier includes: the first type identifier is used to identify or indicate a set of RS resources, and the measurement of the set of RS resources is used to obtain the training dataset for the inference.

[0659] As an example, the inference association first type of identifier includes: the dataset used to train the parameters of the inference is identified by the first type of identifier.

[0660] As an example, the advantages of the above method include identifying the AI ​​model / inference parameters trained by the AI ​​by identifying the RS resource set or training dataset used for AI training, establishing consensus among different AI model / inference parameters, and further simplifying the design.

[0661] Examples 7B-7C

[0662] Examples 7B-7C illustrate schematic diagrams of multiple feature sets according to an embodiment of this application, as shown in Figures 7B-7C respectively.

[0663] In embodiment 7B, the plurality of feature sets are respectively applicable to different frequency bands. In Figure 7B, feature set #1 is applicable to frequency band #1, and feature set #2 is applicable to frequency band #2.

[0664] As one example, the multiple feature sets are applicable to different supported bands simultaneously.

[0665] As one embodiment, the plurality of feature sets are respectively applicable to different frequency bands in a band combination, the band combination including multiple frequency bands.

[0666] As an example, the multiple feature sets are respectively applicable to different frequency bands in a band combination, which includes multiple simultaneously supported frequency bands.

[0667] As an example, the feature set includes some or all of the information in a FeatureSetsPerBand.

[0668] As an example, the feature set includes some or all of the information in a FeatureSet.

[0669] As an example, the feature set includes at least one of FeatureSetDownlink and FeatureSetUplink.

[0670] As an example, the feature set includes some or all of the information in a FeatureSetDownlink or FeatureSetUplink.

[0671] As an example, each of the multiple feature sets includes some or all of the information in a FeatureSetsPerBand within the same FeatureSetCombination.

[0672] As one embodiment, the multiple feature sets respectively include some or all of the information in FeatureSetsPerBand of different frequency bands in the same FeatureSetCombination.

[0673] As an example, the plurality of feature sets includes the combination of FeatureSets at the same position in the FeatureSetsPerBand across bands in a FeatureSetCombination.

[0674] As an example, the plurality of feature sets includes a combination of the first FeatureSet in a cross-band FeatureSetsPerBand within a FeatureSetCombination.

[0675] As one embodiment, the plurality of feature sets includes a combination of a second FeatureSet in a FeatureSetCombination that spans a frequency band within FeatureSetsPerBand.

[0676] In the above method, a FeatureSetCombination includes feature sets applicable to multiple frequency bands within a frequency band combination. Each frequency band's FeatureSetsPerBand comprises one or more FeatureSets arranged sequentially, with the position of a FeatureSet within its respective FeatureSetsPerBand determined by its order. The first node supports combinations of FeatureSets at the same position within FeatureSetsPerBand across frequency bands in a FeatureSetCombination.

[0677] In embodiment 7C, the plurality of feature sets are respectively applied to different carriers. In Figure 7C, feature set #1 is applied to carrier #1, and feature set #2 is applied to carrier #2. In Figure 7C(a), both carrier #1 and carrier #2 are in frequency band #1; in Figure 7C(b), carrier #1 is in frequency band #1, and carrier #2 is in frequency band #2.

[0678] As one example, the multiple feature sets are respectively applicable to different carriers in the same frequency band.

[0679] As an example, the multiple feature sets each include different FeatureSetDownlinkPerCCs within the same FeatureSetDownlink.

[0680] As an example, the multiple feature sets each include different FeatureSetUplinkPerCCs within the same FeatureSetUplink.

[0681] As an example, the plurality of feature sets are applicable to different carriers, and at least two feature sets in the plurality of feature sets are applicable to carriers in different frequency bands.

[0682] As an example, the multiple feature sets each include different FeatureSetDownlinkPerCC.

[0683] As one example, the plurality of feature sets each include different FeatureSetUplinkPerCC.

[0684] As an example, the plurality of feature sets each include different FeatureSetDownlinkPerCC and different FeatureSetUplinkPerCC.

[0685] As an example, the feature set includes at least one of a FeatureSetDownlinkPerCC and a FeatureSetUplinkPerCC.

[0686] As an example, the feature set includes some or all of the information in a FeatureSetDownlinkPerCC or FeatureSetUplinkPerCC.

[0687] Examples 8A-8D

[0688] Examples 8A-8D illustrate schematic diagrams of determining the applicable function according to an embodiment of this application, as shown in Figures 8A-8D respectively.

[0689] In Example 8A, the amount of resources occupied by each of the N1 applicable functions is equal to or less than that of J.

[0690] As an example, whether one of the N functions is an applicable function of the first node depends on whether the number of resources in the first node required by the one function is not greater than J.

[0691] As an example, a necessary condition for one of the N functions to be an applicable function of the first node includes: the number of resources in the first node required by the one function is not greater than J.

[0692] In Example 8B, the N1 applicable functions satisfy a first condition; the first condition includes: the total number of resources occupied by all applicable functions is equal to or less than J.

[0693] As an example, whether one of the N functions is the applicable function of the first node also depends on whether the total number of resources required by the one function and all functions in the first function group is not greater than J; the first function group includes each function that has been identified as an applicable function before determining whether the one function is the applicable function of the first node.

[0694] As an example, the number of resources required by the N1 applicable functions and any function other than the N1 applicable functions among the N functions is greater than J.

[0695] In embodiment 8C, whether one of the N functions is the applicable function of the first node depends on the purpose of the function.

[0696] In the above method, functions with different purposes may have different priorities. When selecting applicable functions, higher-priority functions should be chosen first. The benefits include allocating limited resources to higher-priority functions, thus improving system performance.

[0697] In embodiment 8D, the first function and the second function are two functions among the N functions. Under the requirement of satisfying the first condition, the second function is determined as the applicable function of the first node with priority over the first function. Among the first function and the second function, only the first function is used for channel information reporting. The first condition includes: the total number of resources occupied by all applicable functions is equal to or less than J.

[0698] As one embodiment, the second function includes one of decoding the wireless channel, encoding the wireless channel, channel estimation on DMRS, or wireless link failure prediction.

[0699] In the above method, to meet the requirement of the first condition, if only one of the second function and the first function can be determined as the applicable function, then the second function is determined as the applicable function. The benefits include: allocating limited resources to higher-priority functions, thus improving system performance.

[0700] Examples 8E-8F

[0701] Examples 8E-8F illustrate schematic diagrams illustrating the relationship between multiple feature sets and multiple groups of feature sets according to an embodiment of this application, as shown in Figures 8E-8F respectively. In Figures 8E-8F, group #1 and group #2 are two groups of feature sets, and the multiple feature sets include feature set #1 and feature set #2, with the multiple feature sets in group #1.

[0702] In embodiment 8E, the plurality of feature sets belong to the same set of feature sets within multiple sets of feature sets; wherein each set of feature sets within the multiple sets of feature sets includes a feature set applicable to each carrier in a frequency band. In Figure 8E(a), both group #1 and group #2 are applicable to frequency band #1, feature set #1 and feature set #2 in group #1 are applicable to carrier #1 and carrier #2 in frequency band #1, respectively, and feature set #3 and feature set #4 in group #2 are applicable to carrier #1 and carrier #2 in frequency band #1, respectively. In Figure 8E(b), group #1 and group #2 are applicable to frequency band #1 and frequency band #2, respectively, feature set #1 and feature set #2 in group #1 are applicable to carrier #1 and carrier #2 in frequency band #1, respectively, and feature set #3 and feature set #4 in group #2 are applicable to carrier #3 and carrier #4 in frequency band #2, respectively.

[0703] As one embodiment, the multiple feature sets each include a feature set applicable to each carrier in the same frequency band.

[0704] As one embodiment, the multiple feature sets each include a feature set applicable to each carrier in different frequency bands.

[0705] As one embodiment, the plurality of feature sets are respectively applicable to different carriers; the plurality of feature sets belong to the same set of feature sets in multiple sets of feature sets; each set of feature sets in multiple sets of feature sets includes a feature set applicable to each carrier in a frequency band.

[0706] As one embodiment, the plurality of feature sets are respectively applicable to different carriers in a frequency band; the plurality of feature sets belong to the same set of feature sets in multiple sets of feature sets, the multiple sets of feature sets are applicable to the same frequency band, and each set of feature sets in multiple sets of feature sets includes a feature set applicable to each carrier in the same frequency band.

[0707] As an example, the multiple feature sets are respectively applicable to different carriers in a frequency band; the multiple feature sets belong to the same set of feature sets in multiple feature sets, the multiple feature sets are respectively applicable to different frequency bands, and each set of feature sets in the multiple feature sets includes a feature set applicable to each carrier in the applicable frequency band.

[0708] In embodiment 8F, the plurality of feature sets belong to the same set of feature sets within a plurality of feature sets; wherein each set of feature sets in the plurality of feature sets includes a feature set applicable to each frequency band in a frequency band combination, the frequency band combination including one or more frequency bands. In Figure 8F(a), both group #1 and group #2 are applicable to frequency band combination #1, feature set #1 in group #1 and feature set #3 in group #2 are applicable to frequency band #1 in frequency band combination #1, and feature set #2 in group #1 and feature set #4 in group #2 are applicable to frequency band #2 in frequency band combination #1. In Figure 8F(b), group #1 and group #2 are applicable to frequency band combination #1 and frequency band combination #2, respectively, feature set #1 and feature set #2 in group #1 are applicable to frequency band #1 and frequency band #2 in frequency band combination #1, respectively, and feature set #3 and feature set #4 in group #2 are applicable to frequency band #3 and frequency band #4 in frequency band combination #2, respectively.

[0709] As an example, the multiple feature sets belong to the same set of feature sets in multiple feature sets, and the multiple feature sets respectively include the applicable feature sets for each frequency band in different band combinations.

[0710] As an example, the multiple feature sets belong to the same set of feature sets in multiple feature sets, and the multiple feature sets respectively include some or all of the information in different FeatureSetCombinations.

[0711] As an example, the multiple feature sets belong to the same set of feature sets in multiple feature sets, and the multiple feature sets each include combinations of feature sets applicable to different Band Combinations.

[0712] As an example, the multiple feature sets belong to the same set of feature sets in multiple feature sets, and the multiple feature sets include some or all of the information in the same FeatureSetCombination.

[0713] As an example, the multiple feature sets belong to the same set of feature sets in multiple feature sets, and each set of feature sets in multiple feature sets includes a combination of feature sets at the same position in FeatureSetsPerBand across a band in a FeatureSetCombination.

[0714] As an example, the multiple feature sets belong to the same set of feature sets in multiple feature sets. Each set of feature sets in the multiple feature sets includes a feature set applicable to each frequency band in the same frequency band combination. The multiple feature sets belong to the same set of feature sets in the multiple feature sets.

[0715] As an example, the multiple feature sets belong to the same set of feature sets in multiple feature sets, and each set of feature sets in the multiple feature sets includes a combination of feature sets at the same position in the feature sets per band across the same feature set combination; the positions of the feature sets included in the multiple feature sets in the feature sets per band are different.

[0716] As an example, the multiple feature sets belong to the same set of feature sets in two sets of feature sets. The two sets of feature sets respectively include the combination of the first feature set in the cross-band FeatureSetsPerBand of the same FeatureSetCombination, and the combination of the second feature set in the cross-band FeatureSetsPerBand of the same FeatureSetCombination.

[0717] As one embodiment, the multiple feature sets are respectively applicable to different frequency bands in the same frequency band combination; the multiple feature sets belong to the same set of feature sets in multiple sets of feature sets, and each set of feature sets in multiple sets of feature sets includes a combination of feature sets at the same position in feature sets per band across the same feature set combination; the positions of the feature sets included in the multiple sets of feature sets in feature sets per band are different.

[0718] As one embodiment, the multiple feature sets are applicable to different frequency bands; the multiple feature sets belong to the same set of feature sets in multiple sets of feature sets, and each set of feature sets in multiple sets of feature sets includes a feature set applicable to each frequency band in a frequency band combination, and the frequency band combination includes one or more frequency bands.

[0719] As an example, the multiple feature sets are respectively applicable to different frequency bands in the same frequency band combination; the multiple feature sets belong to the same set of feature sets in multiple feature sets, the multiple set of feature sets are respectively applicable to different frequency band combinations, each set of feature sets in the multiple set of feature sets includes the feature set applicable to each frequency band in the applicable frequency band combination, and the frequency band combination includes one or more frequency bands.

[0720] As an example, the multiple feature sets are respectively applicable to different carriers in the same frequency band combination; the multiple feature sets belong to the same set of feature sets in multiple feature sets, and the multiple set of feature sets are respectively applicable to different frequency band combinations. Each set of feature sets in the multiple set of feature sets includes the feature set applicable to each carrier in each frequency band of the applicable frequency band combination, and the frequency band combination includes one or more frequency bands.

[0721] As one embodiment, the multiple feature sets are respectively applicable to different frequency bands in the same frequency band combination; the multiple feature sets belong to the same set of feature sets in multiple feature sets, the multiple set of feature sets are applicable to the same frequency band combination, each set of feature sets in the multiple set of feature sets includes a feature set applicable to each frequency band in the same frequency band combination, and the frequency band combination includes one or more frequency bands.

[0722] As one embodiment, the multiple feature sets are respectively applicable to different carriers in the same frequency band combination; the multiple feature sets belong to the same set of feature sets in multiple sets of feature sets, the multiple sets of feature sets are applicable to the same frequency band combination, each set of feature sets in the multiple sets of feature sets includes a feature set applicable to each carrier in each frequency band in the same frequency band combination, and the frequency band combination includes one or more frequency bands.

[0723] Example 9A

[0724] Example 9A illustrates a schematic diagram of the transmission conditions of a first information block according to an embodiment of this application; as shown in Figure 9A.

[0725] In embodiment 9A, the first node in this application receives a second information block and sends a first information block; wherein, the first information block is used to indicate N1 applicable functions of the first node from N functions, wherein all N functions are based on AI, N ​​is a positive integer greater than 1, and N1 is a positive integer not greater than N; the amount of resources occupied by each of the N1 applicable functions is equal to or less than J; wherein,

[0726] The second information block indicates the N functions;

[0727] or,

[0728] The second information block instructs the first node to provide the applicable functions of the first node;

[0729] or,

[0730] The second information block instructs the first node to report UAI (UE Assistance Information).

[0731] Examples 9B-9C

[0732] Examples 9B-9C illustrate the amount of resources occupied by multiple feature sets according to an embodiment of this application, as shown in Figures 9B-9C. In Figures 9B-9C, group 1#, ..., group #Q represent the multiple feature sets; integers #1, ..., integer #Q are non-negative integers corresponding to group 1#, ..., group #Q.

[0733] In Example 9B, the first message indicates the total number or maximum total number of resources occupied by the multiple feature sets respectively, and the total number of resources occupied by the same feature set to which the multiple feature sets belong is equal to or less than J.

[0734] As an example, the first message indicates the J.

[0735] As an example, the first message also indicates that the total number or maximum total number of resources occupied by the same set of features to which the plurality of feature sets belong is J.

[0736] In Example 9C, the total number of resources occupied by each feature set in the plurality of feature sets is equal to or less than J.

[0737] As an example, the first message indicates the J.

[0738] As an example, the first message indicates that the total number or maximum total number of resources occupied by each group of features in the plurality of feature sets is J.

[0739] Example 10A

[0740] Example 10A illustrates a schematic diagram of the transmission conditions of a first information block according to another embodiment of this application; as shown in Figure 10A.

[0741] In Example 10A, the first information block is triggered when the second condition is met; the second condition includes: the first node receives an RRC message indicating that the training dataset has been reconfigured.

[0742] As an example, whether the training dataset is used to train an AI model or inference parameters applicable to only one cell or multiple cells depends on whether the training dataset is configured for only one cell.

[0743] As an example, when the training dataset is configured for only one cell, the AI ​​model or inference parameters used to train the training dataset are applicable to the only cell; when the training dataset is configured for multiple cells or the configuration of the training dataset is independent of the cell configuration, the AI ​​model or inference parameters used to train the training dataset are applicable to multiple cells.

[0744] As an example, whether the AI ​​model or inference parameters of one of the N1 applicable functions are still applicable after cell handover depends on the use of that function.

[0745] As an example, the AI ​​model or inference parameters used for CSI prediction remain applicable after cell handover.

[0746] As an example, the AI ​​model or inference parameters used for CSI compression functionality are still applicable after cell handover.

[0747] As an example, the AI ​​model or inference parameters used for CSI prediction are also applicable after switching between cells in a first cell group, which includes multiple cells.

[0748] As an example, the AI ​​model or inference parameters for the CSI compression function are still applicable after switching between cells in a first cell group, which includes multiple cells.

[0749] As an example, the AI ​​model or inference parameters used for beam prediction functionality are not applicable after cell handover.

[0750] As an example, when the serving cell of the first node switches from the first cell to the second cell, if the training dataset remains unchanged, the AI ​​model or inference parameters trained on the training dataset are still applicable to the second cell.

[0751] As an example, the training dataset includes data from one cell or data from multiple cells.

[0752] As one example, the training dataset includes data from the first node or data from other user devices.

[0753] As an example, the base station indicates which user devices the first node can train the AI ​​model or inference parameters with.

[0754] As one embodiment, the second condition includes: the first node receiving an RRC message indicating that the training dataset has been reconfigured, the training dataset being used to train the AI ​​model or inference parameters on which the applicable functions of the first node are based.

[0755] In the above method, when the training dataset is reconfigured, the transmission of the first information block allows the base station to know whether the AI ​​model or inference parameters trained on the original training dataset are still valid, or whether the AI ​​model or inference parameters trained on the new training dataset have been completed. The benefits include: allowing the base station to know whether the previously applicable functions are still applicable to the first node under the new training dataset.

[0756] Example 10B

[0757] Example 10B illustrates a schematic diagram of the amount of resources occupied by a feature in a feature set according to an embodiment of the present application; as shown in Figure 10B.

[0758] In embodiment 10B, the first message indicates the amount of resources occupied by a feature in one of the plurality of feature sets. In Figure 10B, a feature set includes features #1, ..., features #t, and the amount of resources occupied by them are integers #T1, ..., integers #Tt, where integers #T1, ..., integers #Tt are non-negative or positive integers.

[0759] As an example, the first message indicates the amount of resources occupied by only a portion of the features in a subset of the plurality of feature sets.

[0760] As an example, the first message indicates the amount of resources occupied by only a portion of the features in one of the plurality of feature sets.

[0761] As an example, the first message indicates the amount of resources occupied by each feature in each of the plurality of feature sets.

[0762] In the above method, the amount of resources occupied by different features may be the same or different. The advantages of this method include: the target recipient of the first message (i.e., the second node in this application) can fully understand the resource requirements of different features, which helps the target recipient of the first message to reasonably schedule or configure the first node.

[0763] Example 11A

[0764] Example 11A illustrates a schematic diagram of RAN (Radio Access Network) domain AI / ML function deployment according to one embodiment of this application; as shown in Figure 11A. The gNB in ​​Example 11A can be replaced with, for example, an eNB, or a network device such as a 6G base station.

[0765] AI / ML related functions include ML training (also known as AI training, or AI / ML training), ML testing, and ML inference (also known as AI inference, or AI / ML inference), etc. ML training, ML testing, and ML inference functions can be deployed independently or co-located. Deployment of AI / ML related functions can be implemented through software, such as downloading and / or running executable files; or it can be implemented through a combination of software and hardware, such as accelerating specific computing units through hardware to improve computing speed or save power.

[0766] ML training functionality can be deployed in a cross-domain management system or a domain-specific management system; the domain-specific management system is used to manage the RAN domain or the CN (Core Network) domain. For example, ML training functionality for MDA (Management Data Analytics) can be deployed on MDAF (MDA Function); ML training for network data analytics can be deployed on NWDAF (Network Data Analytics Function), meaning the ML training functionality is an MTLF (Model Training Logical Function).

[0767] The ML inference function can also be deployed in a cross-domain management system or a domain-specific management system; for example, the ML inference function is MDAF, or the ML inference function is AnLF (Analytics logical function) located in NWDAF.

[0768] Similarly, ML testing capabilities can also be deployed in cross-domain management systems or domain-specific management systems.

[0769] In Example 11A, the RAN domain ML training function 1402 is located in the RAN domain management function 1403; while the ML inference function is located in the base station, that is, the AI / ML inference function 1404 is located in gNB 1405, the AI / ML inference function 1406 is located in gNB 1407, and so on.

[0770] In Figure 11A, the management of ML inference functions of multiple base stations is completed by RAN domain management function 1403, that is, data interaction with RAN domain MnS (Management Service) consumers / cross-domain management 1401 (as shown by the dashed arrow in Figure 11A).

[0771] Optionally, the management of ML inference function can also be completed by the base station itself, that is, each base station can independently interact with the RAN domain MnS consumer / cross-domain management 1401.

[0772] It should be noted that Embodiment 11A is merely a non-limiting implementation; optionally, the ML training function of the RAN domain may also be deployed in the base station; or optionally, some base stations may deploy both the ML inference function and the ML training function of the RAN domain, while some base stations may only deploy the ML inference function.

[0773] As an example, one of the gNBs (or base stations) in Example 11A is the second node of this application.

[0774] As an example, the second processor in this application includes an AL / ML inference function, namely 1404 or 1406, in Figure 11A.

[0775] Example 11B

[0776] Example 11B illustrates a schematic diagram of the function implemented by reasoning according to an embodiment of the present application; as shown in Figure 11B.

[0777] In embodiment 11B, the reasoning performed by the first node in this application is used to implement one of the following functions: channel information reporting, downlink radio channel reception, uplink radio channel transmission, uplink radio channel encoding, downlink radio channel decoding, channel estimation on DMRS (Demodulation Reference Signal), and radio link failure prediction.

[0778] As an example, the reasoning performed by the first node is used to implement the function of channel information reporting, which includes one or more of beam prediction reporting, CSI prediction, and CSI compression.

[0779] As an example, the advantages of the above method include: the acquisition and reporting of channel information based on AI / ML improves the accuracy of channel information estimation and enhances system performance.

[0780] As an example, the reception of the downlink wireless channel includes at least one of channel estimation, demodulation, and decoding.

[0781] As one example, receiving the downlink wireless channel includes decoding.

[0782] As one example, receiving the downlink wireless channel includes demodulation and decoding.

[0783] As an example, receiving the downlink radio channel includes recovering the data carried on the downlink radio channel.

[0784] In the above method, the specific algorithm for receiving the downlink wireless channel can be implementation-dependent, i.e., determined by the hardware equipment vendor of the first node. Several typical but non-limiting implementation methods are described below:

[0785] In one implementation, the first node demodulates the signal on the downlink wireless channel and inputs it into an AI model for decoding. The output of the AI ​​model is used to recover the data carried on the downlink wireless channel.

[0786] In one implementation, the first node estimates the channel matrix based on the DMRS of the downlink wireless channel, inputs the signal on the downlink wireless channel and the estimated channel matrix into an AI model, and the output of the AI ​​model is used to recover the data carried on the downlink wireless channel.

[0787] In another implementation, the REs occupied by the DMRS of the downlink radio channel and the REs carrying data in the downlink radio channel overlap; the first node performs channel estimation based on AI according to the signal on the RE occupied by the DMRS of the downlink radio channel; then, the first node demultiplexes the DMRS and the data, and inputs the signal of the demultiplexed data carrying data into the AI ​​model, and the output of the AI ​​model is used to recover the data carried on the downlink radio channel.

[0788] The structure and parameters of the AI ​​model in the above embodiments are known to the first node. For example, they may be obtained by downloading from a network device, or they may be specified in a standard, or they may be implementation-related to the first node (i.e., determined by the hardware device vendor of the receiver of the downlink wireless channel).

[0789] As an example, the advantages of the above method include: improving the reception (e.g., decoding) performance of the downlink wireless channel based on AI / ML, improving the reliability of the downlink wireless channel, and improving transmission efficiency.

[0790] As an example, uplink wireless channel coding is one of the many functions.

[0791] As an example, the advantages of the above method include: encoding the uplink wireless channel based on AI / ML (channel coding, or source-channel joint coding) improves transmission reliability and transmission efficiency.

[0792] As an example, channel estimation on DMRS is one of the many functions mentioned.

[0793] As an example, the advantages of the above method include: improving the channel estimation accuracy on DMRS based on AI / ML, which can increase the transmission capacity of wireless channels (such as PDSCH, PUSCH, PDCCH, PUCCH, etc.) and reduce data interference to DMRS.

[0794] As an example, wireless link failure prediction is one of the many functions mentioned.

[0795] As an example, the benefits of the above method include: predicting wireless link failures based on AI / ML, reducing or avoiding actual wireless link failures, improving the reliability of wireless links, and enhancing system performance.

[0796] As one embodiment, receiving the downlink wireless channel includes decoding the downlink wireless channel.

[0797] As an example, the decoding of the downlink wireless channel includes the decoding of the physical channel through which the data is transmitted.

[0798] As an example, the decoding of the downlink radio channel includes the decoding of the PDSCH.

[0799] As one embodiment, the decoding of the downlink wireless channel includes the decoding of the physical channel that transmits control information.

[0800] As an example, the decoding of the downlink radio channel includes the decoding of the PDCCH (Physical Downlink Control Channel).

[0801] As an example, the encoding of the uplink wireless channel includes the encoding of the physical channel for transmitting data.

[0802] As an example, the encoding of the uplink wireless channel includes the encoding of PUSCH (Physical Uplink Shared Channel).

[0803] As one embodiment, the encoding of the uplink wireless channel includes the encoding of the physical channel for transmitting control information.

[0804] As an example, the encoding of the uplink wireless channel includes the encoding of PUCCH (Physical Uplink Control Channel).

[0805] As an example, the radio link failure prediction includes RLF (Radio Link Failure) prediction.

[0806] As one embodiment, the wireless link failure prediction includes beam failure prediction.

[0807] As an example, the reasoning is based on training or AI.

[0808] As an example, the reasoning includes AI (Artificial Intelligence) inference.

[0809] As an example, the reasoning model is obtained through training.

[0810] As an example, the training of the inference in the first node is performed by the first node.

[0811] As an example, the training of the inference in the first node is performed by the second node.

[0812] As an example, the training of the inference in the first node is performed by the core network.

[0813] As an example, the training of the inference in the first node is performed by an AI training producer.

[0814] As an example, the training of the inference in the first node is performed by the MDA (Management Data Analytics Function).

[0815] As an example, the training of the inference in the first node is performed by the MDA function located in the first node.

[0816] As an example, the training of the inference in the first node is performed by the MDA function located in the second node.

[0817] As an example, the training of the inference in the first node is performed by NWDAF (Network Data Analytics Function).

[0818] As an example, the training of the inference in the first node is performed by the MDAS (Management Data Analytics Service) producer.

[0819] As an example, the training of the inference in the first node is performed by the MnS (Management Service) producer.

[0820] As an example, the inference in the first node needs to be deployed.

[0821] As an example, the reasoning in the first node is obtained by loading.

[0822] As an example, the inference in the first node is obtained from the serving cell of the first node.

[0823] As an example, the inference in the first node is obtained from the sustaining base station of the serving cell of the first node.

[0824] As an example, the first node deploys the inference.

[0825] As an example, the inference does not require deployment.

[0826] As an example, the inference is obtained from the core network.

[0827] As an example, the reasoning is based on artificial intelligence or machine learning.

[0828] As an example, the reasoning is based on a neural network.

[0829] As an example, the inference is based on classic models such as Transformer architecture, RNN (Recurrent Neural Network), CNN (Conventional Neural Networks), or a hybrid model composed of multiple models.

[0830] As one example, the inference includes preprocessing.

[0831] As one example, the reasoning includes post-processing.

[0832] As one example, the post-processing includes DFT.

[0833] As one example, the post-processing includes quantization.

[0834] As an example, the post-processing 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, and frequency domain to time domain transformation.

[0835] As one example, the post-processing includes truncation and / or padding.

[0836] As an example, the inference includes one or more of convolution, pooling, cascading, and activation.

[0837] As one example, the inference includes a fully connected layer.

[0838] As one example, the inference includes a pooling layer.

[0839] As an example, the inference includes at least one convolutional layer.

[0840] As one example, the reasoning includes at least one coding layer.

[0841] As an example, an encoding layer includes at least one convolutional layer and one pooling layer.

[0842] As an example, in a convolutional layer, at least one convolutional kernel is used to convolve the input to generate a corresponding feature map, and at least one feature map output by the convolutional layer is reshaped into a vector and input to a fully connected layer; the fully connected layer transforms the vector into an output.

[0843] As an example, some or all of the following parameters in the inference—convolution kernel size, number of convolutional layers, convolution stride, pooling kernel size, pooling kernel stride, pooling function, activation function, and number of feature maps—are obtained through training.

[0844] As an example, some or all of the convolution kernel, pooling kernel, pooling function, activation function, parameters of the pooling function, and parameters of the activation function in the inference are obtained through training.

[0845] As one example, the AI ​​includes ML (Machine Learning).

[0846] As an example, the AI ​​includes AI and ML.

[0847] As one example, the AI ​​includes AI or ML.

[0848] As an example, the preprocessing includes one or more of the following: quantization, DFT (Discrete Fourier Transform), matrix decomposition, matrix transformation or projection, spatial-to-angular-domain transformation, angular-to-spatial-domain transformation, frequency-to-time-domain transformation, time-to-frequency-domain transformation, truncation, padding, mapping, or labeling.

[0849] As an example, the preprocessing includes DFT (Discrete Fourier Transform).

[0850] As one example, the preprocessing includes one or more of matrix decomposition, matrix transformation, or projection.

[0851] As an example, the preprocessing includes one or more of the following: quantization, spatial-to-angular-domain transformation, angular-to-spatial-domain transformation, frequency-to-time-domain transformation, or time-to-frequency-domain transformation.

[0852] As one example, the preprocessing includes truncation and / or padding.

[0853] As one example, the preprocessing includes mapping.

[0854] As one example, the preprocessing includes mapping to vectors.

[0855] As one example, the preprocessing includes labeling.

[0856] As an example, the label refers to a mark made with a label.

[0857] Example 12A

[0858] Example 12A illustrates a schematic diagram of the deployment of AI / ML functionality in a UE according to one embodiment of this application; as shown in Figure 12A. The RAN domain ML training function 1505 in Figure 12A is optional.

[0859] UE function 1504 is deployed in the first node of this application, and the UE function 1504 includes AI / ML inference function 1506; the AI / ML inference function 1506 uses an ML model (also called an AI model) for inference; an ML model is typically trained before being used for AI / ML inference.

[0860] As an example, the first processor in this application includes an AL / ML inference function 1506 in Figure 12A.

[0861] As an example, the UE function 1504 includes a RAN domain ML training function 1505, which runs training data through an ML model to obtain a relevant loss and adjusts the parameters of the ML model based on the calculated loss; the ML training includes at least one of ML initial training, ML re-training, and reinforcement learning.

[0862] The above embodiments can reduce the complexity of the base station, or save air interface resources caused by reporting training data; however, the above embodiments place higher demands on the processing capabilities of the UE side.

[0863] Optionally, the UE function 1504 also includes a CN domain ML training function (not shown in Figure 12A).

[0864] Optionally, the UE function 1504 also includes an AI / ML deployment function—not shown in Figure 12A—for loading ML models and data.

[0865] As an example, the first node indicates whether it supports ML training function (RAN domain or CN domain) through capability reporting. The capability reporting is RRC signaling or NAS (Non-Access Stratum) signaling.

[0866] As an example, the ML model and the associated metadata are loaded by the first node from a network device or a remote server.

[0867] Optionally, the UE function 1504 is an MnS (Management Service) producer that provides data to the CN domain MnF (Management Function) 1501, and / or the RAN domain MnF 1502, and / or the cross-domain management system 1503 for management or analysis (as shown by double arrow 1507).

[0868] Optionally, the UE function 1504 is an MnS consumer that loads data from the CN domain MnF (Management Function) 1501, and / or the RAN domain MnF 1502, and / or the cross-domain management system 1503 for AI / ML-related management, such as managing data requests, ML model activation, and / or ML training (as shown by double arrow 1507).

[0869] As an example, the ML model is based on a neural network.

[0870] As an example, the ML model is based on CNN (Conventional Neural Networks).

[0871] As an example, the ML model is based on the Transformer architecture.

[0872] Example 12B

[0873] Example 12B illustrates a schematic diagram of a first encoder according to an embodiment of this application, as shown in Figure 12B. In Figure 12B, the first encoder is an AI-based encoder, which includes P1 coding layers, namely coding layers #1, #2, ..., #P1.

[0874] As an example, P1 is 2, meaning the P1 encoding layers include encoding layer #1 and encoding layer #2, where encoding layer #1 is a convolutional layer and encoding layer #2 is a fully connected layer, respectively. In the convolutional layer, at least one convolutional kernel is used to convolve the input of the first encoder to generate a corresponding feature map. At least one feature map output from the convolutional layer is reshaped into a vector and input to the fully connected layer. The fully connected layer transforms the vector into the output of the first encoder. For a more detailed description, please refer to CNN-related technical literature, such as Chao-Kai Wen, Deep Learning for Massive MIMO CSI Feedback, IEEE WIRELESS COMMUNICATIONS LETTERS, VOL.7, NO.5, OCTOBER 2018, etc.

[0875] As an example, P1 is 3, that is, the P1 coding layers include fully connected layers, convolutional layers, and pooling layers.

[0876] Example 12C

[0877] Example 12C illustrates a schematic diagram of a first decoder according to an embodiment of this application, as shown in Figure 12C. In Figure 12C, the first decoder is an AI-based decoder, which includes a preprocessing layer and P2 decoding layer groups, namely decoding layer groups #1, #2, ..., #P2, each decoding layer group including at least one decoding layer.

[0878] As an example, the preprocessing layer is a fully connected layer.

[0879] As an example, any two decoding layer groups in the P2 decoding layer groups have the same structure, which includes the number of decoding layers, the size of the input parameters of each decoding layer, the size of the output parameters, etc.

[0880] As an example, the decoding layer group #j includes L layers, namely layers #1, #2, ..., #L; the decoding layer group is any one of the P2 decoding layer groups.

[0881] As an example, L is 4, the first layer in the L layer is the input layer, and the last three layers in the L layer are convolutional layers. For a more detailed description, please refer to CNN-related technical literature, such as Chao-Kai Wen, Deep Learning for Massive MIMO CSI Feedback, IEEE WIRELESS COMMUNICATIONS LETTERS, VOL.7, NO.5, OCTOBER 2018, etc.

[0882] As an example, the L layer includes at least one convolutional layer and one pooling layer.

[0883] Example 13

[0884] Example 13 illustrates a schematic diagram of a processing system based on artificial intelligence or machine learning according to an embodiment of this application; as shown in Figure 13. Figure 13(a) includes a third processor, a fourth processor, and a fifth processor, and Figure 13(b) includes a third processor, a fourth processor, a fifth processor, and a sixth processor.

[0885] In Example 13(a), 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-type parameter set based on the first dataset, and sends the generated target first-type parameter set to the fifth processor; the fifth processor processes the second dataset using the target first-type parameter set to obtain a first-type output. In Figure 13(a), the first-type feedback is optional.

[0886] In Example 13(b), 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-type parameter set based on the first dataset, and sends the generated target first-type parameter set to the fifth processor; the fifth processor processes the second dataset using the target first-type parameter set to obtain a first-type output, and sends the first-type output to the sixth processor. In Figure 13(b), the first-type feedback and the second-type feedback are optional.

[0887] As an example, in Figure 13(a), the fifth processor sends the first type of output to the second node in this application.

[0888] As an example, Figure 13(a) employs a single-side AI model, in which the fifth processor performs inference in the first node of this application.

[0889] As an example, Figure 13(b) employs a two-sided AI model, in which the fifth processor performs the inference in the first node of this application, and the sixth processor includes the inference in the second node of this application.

[0890] As an example, the AI ​​includes ML (Machine Learning) inference.

[0891] As an example, the fifth processor performs the inference in the first node of this application.

[0892] As one embodiment, the sixth processor includes the inference function in the second node of this application.

[0893] As an example, the fifth processor sends a first type of feedback to the fourth processor, and the first type of feedback is used to trigger a recalculation or update of the target first type of parameter group.

[0894] 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.

[0895] As one embodiment, the third processor generates the first dataset and the second dataset based on measurements of a first type of wireless signal, the first type of wireless signal including downlink RS.

[0896] As one embodiment, the fifth processor belongs to the first node, and the sixth processor belongs to the second node.

[0897] As an example, the second dataset includes the input for inference in the first node.

[0898] As an example, for inference in the first node of this application, the second dataset includes information obtained based on CSI reporting configuration.

[0899] As an example, the first dataset includes training data.

[0900] As an example, the fourth processor belongs to the inference producer in the first node.

[0901] As one embodiment, the fourth processor includes an AI training producer.

[0902] As one embodiment, the fourth processor includes an AI training function.

[0903] As an example, the fourth processor is used for model training, and the trained model is described by the target first class of parameter sets.

[0904] As an example, the fourth processor belongs to the first node.

[0905] The above embodiments avoid passing the first dataset to the second node.

[0906] As one example, the fourth processor belongs to the second node.

[0907] The above embodiments support joint training and optimize system performance.

[0908] As an example, the fourth processor belongs to the core network.

[0909] The above embodiments support network-wide joint training, further optimizing system performance.

[0910] As an example, the second dataset includes inference data.

[0911] As one embodiment, the fifth processor includes an AI inference producer.

[0912] As one embodiment, the fifth processor includes an AI inference function.

[0913] As an example, the fifth processor belongs to the first node.

[0914] 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.

[0915] As an example, the reasoning in the first node is described by the target first type of parameter group.

[0916] As an example, the target first type of parameter group is used to construct the inference in the first node.

[0917] As one embodiment, the fifth processor includes inference from the second node.

[0918] As an example, the fifth processor generates a recovery dataset based on the first type of output, and the error between the recovery dataset and the second dataset is used to generate the first type of feedback.

[0919] As a sub-example of the above embodiment, the generation of the recovery dataset employs inference similar to that in the second node.

[0920] 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.

[0921] 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.

[0922] 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.

[0923] 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.

[0924] Example 14A

[0925] Example 14A illustrates a structural block diagram of a processing apparatus for a first node according to an embodiment of the present application; as shown in Figure 14A. In Figure 14A, the processing apparatus 1800A in the first node includes a first processor 1801A.

[0926] As one example, the first node is a user equipment.

[0927] As an example, the first node is a relay node device.

[0928] As an example, the first processor 1801A includes at least one of the following in embodiment 4: {antenna 452, receiver / transmitter 454, receiver processor 456, transmitter processor 468, multi-antenna receiver processor 458, multi-antenna transmitter processor 457, controller / processor 459, memory 460, data source 467}.

[0929] As an example, the first processor 1801A includes the antenna 452, receiver / transmitter 454, receiver processor 456, transmitter processor 468, multi-antenna receiver processor 458, multi-antenna transmitter processor 457, controller / processor 459, memory 460, and data source 467 as in Example 4.

[0930] The first processor 1801A sends the first message;

[0931] In Example 14A, the first message indicates multiple functions supported by the first node and J resources in the first node, where J is a positive integer greater than 1; the multiple functions are all based on AI, only some of the multiple functions are used for channel information reporting, and any one of the multiple functions occupies at least one of the J resources.

[0932] As an example, the first message indicates the amount of resources occupied by one of the plurality of functions.

[0933] As one example, the multiple functions include one or more of the following: receiving a wireless channel, encoding a wireless channel, channel estimation on DMRS, and predicting wireless link failures.

[0934] As an example, the first processor 1801A sends a first information block; wherein the first information block is used to indicate N1 applicable functions of the first node from N functions, all of which are based on AI, where N is a positive integer greater than 1 and N1 is a positive integer not greater than N; the amount of resources occupied by each of the N1 applicable functions is equal to or less than J.

[0935] As an example, the N1 applicable functions satisfy a first condition; the first condition includes: the total number of resources occupied by all applicable functions is equal to or less than J.

[0936] As an example, whether one of the N functions is the applicable function of the first node depends on the purpose of the function.

[0937] As an example, the first function and the second function are two functions among the N functions. Under the requirement of satisfying a first condition, the second function is determined as the applicable function of the first node with priority over the first function; wherein, only the first function is used for channel information reporting; the first condition includes: the total number of resources occupied by all applicable functions is equal to or less than J.

[0938] As one embodiment, the first processor 1801A performs inference; wherein the inference performed by the first node is used to implement at least one of the N1 applicable functions.

[0939] As one embodiment, the first processor 1801A receives a second information block; wherein,

[0940] The second information block indicates the N functions;

[0941] or,

[0942] The second information block instructs the first node to provide the applicable functions of the first node;

[0943] or,

[0944] The second information block instructs the first node to report UAI (UE Assistance Information).

[0945] As an example, the first information block is triggered when the second condition is met; the second condition includes: the first node receives an RRC message indicating that the training dataset has been reconfigured.

[0946] As one embodiment, the first processor 1801A receives a second message; wherein the second message instructs the first node to report the supported functions, or the second message queries the capabilities of the first node.

[0947] Example 14B

[0948] Example 14B illustrates a structural block diagram of a processing apparatus for a first node according to an embodiment of the present application; as shown in Figure 14B. In Figure 14B, the processing apparatus 1800B in the first node includes a first processor 1801B.

[0949] As one example, the first node is a user equipment.

[0950] As an example, the first node is a relay node device.

[0951] As an example, the first processor 1801B includes at least one of the following in embodiment 4: {antenna 452, receiver / transmitter 454, receiver processor 456, transmitter processor 468, multi-antenna receiver processor 458, multi-antenna transmitter processor 457, controller / processor 459, memory 460, data source 467}.

[0952] As an example, the first processor 1801B includes the antenna 452, receiver / transmitter 454, receiver processor 456, transmitter processor 468, multi-antenna receiver processor 458, multi-antenna transmitter processor 457, controller / processor 459, memory 460, and data source 467 as described in Example 4.

[0953] The first processor 1801B sends the first message;

[0954] In Example 14B, the first message indicates J resources, multiple feature sets, and the number of resources occupied by each feature set in the multiple feature sets; J is a positive integer greater than 1; the feature set 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 candidates for the features in the feature set include at least two different features.

[0955] As an example, the sum of the number of resources occupied by the plurality of feature sets is equal to or less than J.

[0956] As one embodiment, the plurality of feature sets are applicable to different frequency bands, or the plurality of feature sets are applicable to different carriers.

[0957] As an example, the plurality of feature sets belong to the same set of feature sets in multiple sets of feature sets; each set of feature sets in multiple sets of feature sets includes a feature set applicable to each carrier in a frequency band.

[0958] As an example, the plurality of feature sets belong to the same set of feature sets in multiple sets of feature sets; each set of feature sets in multiple sets of feature sets includes a feature set applicable to each frequency band in a frequency band combination, the frequency band combination including one or more frequency bands.

[0959] As an example, the first message indicates the total number or maximum total number of resources occupied by the multiple feature sets respectively, and the total number of resources occupied by the same feature set to which the multiple feature sets belong is equal to or less than J.

[0960] As an example, the total number of resources occupied by each feature set in the plurality of feature sets is equal to or less than J.

[0961] As an example, the first message indicates the amount of resources occupied by a feature in one of the plurality of feature sets.

[0962] As one embodiment, the first processor 1801B receives a second information block; wherein the second information block indicates at least one function, the configuration of which conforms to at least one of the plurality of feature sets.

[0963] As one embodiment, the first processor 1801B sends a first information block; wherein the first information block is used to indicate the applicable function from the at least one function.

[0964] As one embodiment, the first processor 1801B performs inference; wherein the inference performed by the first node is used to implement a suitable function.

[0965] As one embodiment, the first processor 1801B receives a second message; wherein the sending of the first message is a response to receiving the second message; and the second message queries the capabilities of the first node.

[0966] Example 15A

[0967] Example 15A 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 15A. In Figure 15A, the processing apparatus 1900A in the second node includes a second processor 1901A.

[0968] In one embodiment, the second node is a base station device.

[0969] In one embodiment, the second node is a user equipment.

[0970] As one embodiment, the second node is a relay node device.

[0971] As an example, the second processor 1901A includes at least one of the following in embodiment 4: {antenna 420, receiver / transmitter 418, receiver processor 470, transmitter processor 416, multi-antenna receiver processor 472, multi-antenna transmitter processor 471, controller / processor 475, memory 476}.

[0972] As one embodiment, the second processor 1901A includes the antenna 420, receiver / transmitter 418, receiver processor 470, transmitter processor 416, multi-antenna receiver processor 472, multi-antenna transmitter processor 471, controller / processor 475, and memory 476 as in embodiment 4.

[0973] The second processor 1901A receives the first message;

[0974] In Example 15A, the first node is the sender of the first message, which indicates multiple functions supported by the first node and J resources in the first node, where J is a positive integer greater than 1; the multiple functions are all based on AI, only some of the multiple functions are used for channel information reporting, and any one of the multiple functions occupies at least one of the J resources.

[0975] As an example, the first message indicates the amount of resources occupied by one of the plurality of functions.

[0976] As one example, the multiple functions include one or more of the following: receiving a wireless channel, encoding a wireless channel, channel estimation on DMRS, and predicting wireless link failures.

[0977] As one embodiment, the second processor 1901A receives the first information block;

[0978] Wherein, the first information block is used to indicate N1 applicable functions of the first node from N functions, all of which are based on AI, where N is a positive integer greater than 1 and N1 is a positive integer not greater than N; the amount of resources occupied by each of the N1 applicable functions is equal to or less than J.

[0979] As an example, the N1 applicable functions satisfy a first condition; the first condition includes: the total number of resources occupied by all applicable functions is equal to or less than J.

[0980] As an example, whether one of the N functions is the applicable function of the first node depends on the purpose of the function.

[0981] As an example, the first function and the second function are two functions among the N functions. Under the requirement of satisfying a first condition, the second function is determined as the applicable function of the first node with priority over the first function; wherein, only the first function is used for channel information reporting; the first condition includes: the total number of resources occupied by all applicable functions is equal to or less than J.

[0982] As an example, the first node performs inference; wherein the inference performed by the first node is used to implement at least one of the N1 applicable functions.

[0983] As one embodiment, the second processor 1901A sends a second information block; wherein,

[0984] The second information block indicates the N functions;

[0985] or,

[0986] The second information block instructs the first node to provide the applicable functions of the first node;

[0987] or,

[0988] The second information block instructs the first node to report UAI.

[0989] As an example, the first information block is triggered when the second condition is met; the second condition includes: the first node receives an RRC message indicating that the training dataset has been reconfigured.

[0990] As one embodiment, the second processor 1901A sends a second message;

[0991] The second message may indicate that the first node reports the supported functions, or the second message may query the capabilities of the first node.

[0992] Example 15B

[0993] Example 15B 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 15B. In Figure 15B, the processing apparatus 1900B in the second node includes a second processor 1901B.

[0994] In one embodiment, the second node is a base station device.

[0995] In one embodiment, the second node is a user equipment.

[0996] As one embodiment, the second node is a relay node device.

[0997] As an example, the second processor 1901B includes at least one of the following in embodiment 4: {antenna 420, receiver / transmitter 418, receiver processor 470, transmitter processor 416, multi-antenna receiver processor 472, multi-antenna transmitter processor 471, controller / processor 475, memory 476}.

[0998] As one embodiment, the second processor 1901B includes the antenna 420, receiver / transmitter 418, receiver processor 470, transmitter processor 416, multi-antenna receiver processor 472, multi-antenna transmitter processor 471, controller / processor 475, and memory 476 as in embodiment 4.

[0999] The second processor 1901B receives the first message;

[1000] In Example 15B, the first message indicates J resources, multiple feature sets, and the number of resources occupied by each feature set in the multiple feature sets; J is a positive integer greater than 1; the feature set 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 candidates for the features in the feature set include at least two different features.

[1001] As an example, the sum of the number of resources occupied by the plurality of feature sets is equal to or less than J.

[1002] As one embodiment, the plurality of feature sets are applicable to different frequency bands, or the plurality of feature sets are applicable to different carriers.

[1003] As an example, the plurality of feature sets belong to the same set of feature sets in multiple sets of feature sets; each set of feature sets in multiple sets of feature sets includes a feature set applicable to each carrier in a frequency band.

[1004] As an example, the plurality of feature sets belong to the same set of feature sets in multiple sets of feature sets; each set of feature sets in multiple sets of feature sets includes a feature set applicable to each frequency band in a frequency band combination, the frequency band combination including one or more frequency bands.

[1005] As an example, the first message indicates the total number or maximum total number of resources occupied by the multiple feature sets respectively, and the total number of resources occupied by the same feature set to which the multiple feature sets belong is equal to or less than J.

[1006] As an example, the total number of resources occupied by each feature set in the plurality of feature sets is equal to or less than J.

[1007] As an example, the first message indicates the amount of resources occupied by a feature in one of the plurality of feature sets.

[1008] As one embodiment, the second processor 1901B sends a second information block; wherein the second information block indicates at least one function, the configuration of which conforms to at least one of the plurality of feature sets.

[1009] As one embodiment, the second processor 1901B receives a first information block; wherein the first information block is used to indicate the applicable function from the at least one function.

[1010] As one embodiment, the first node is the sender of the first message, and the first node performs reasoning; wherein the reasoning performed by the first node is used to implement an applicable function.

[1011] As one embodiment, the second processor 1901B sends a second message; wherein the first node is the sender of the first message, and the sending of the first message is a response to receiving the second message; the second message queries the capabilities of the first node.

[1012] Those skilled in the art will understand that all or part of the steps in the above methods can be implemented by a program instructing related hardware, and the program can be stored in a computer-readable storage medium, such as a read-only memory, hard disk, or optical disk. Optionally, all or part of the steps in the above embodiments can also be implemented using one or more integrated circuits. Accordingly, each module unit in the above embodiments can be implemented in hardware or in the form of software functional modules. This application is not limited to any specific combination of software and hardware. The user equipment, terminal, and UE in this application include, but are not limited to, drones, communication modules on drones, remote-controlled aircraft, aircraft, small aircraft, mobile phones, tablets, laptops, vehicle-mounted communication equipment, vehicles, RSUs, wireless sensors, internet access cards, IoT terminals, RFID terminals, NB-IoT terminals, MTC (Machine Type Communication) terminals, eMTC (enhanced MTC) terminals, data cards, internet access cards, vehicle-mounted communication equipment, low-cost mobile phones, low-cost tablets, and other wireless communication devices. The base stations or system equipment in this application include, but are not limited to, macrocell base stations, microcell base stations, small cell base stations, home base stations, relay base stations, eNBs, gNBs, TRPs (Transmitter Receiver Points), GNSS, relay satellites, satellite base stations, airborne base stations, RSUs (Road Side Units), drones, and testing equipment, such as transceivers or signaling testers that simulate some functions of a base station, and other wireless communication equipment.

[1013] Those skilled in the art will understand that the present invention can be practiced in other specified forms without departing from its core or essential characteristics. Therefore, the embodiments disclosed herein should in any way be considered descriptive rather than restrictive. The scope of the invention is defined by the appended claims rather than the foregoing description, and all modifications within their equivalent meaning and scope are considered to be included therein.

Claims

1. A first node used for wireless communication, characterized in that, include: The first processor sends the first message; The first message indicates multiple functions supported by the first node and J resources in the first node, where J is a positive integer greater than 1; the multiple functions are all based on AI, only some of the multiple functions are used for channel information reporting, and any one of the multiple functions occupies at least one of the J resources.

2. The first node according to claim 1, characterized in that, The first message indicates the amount of resources occupied by one of the plurality of functions.

3. The first node according to claim 1 or 2, characterized in that, The multiple functions include one or more of the following: receiving a wireless channel, coding a wireless channel, channel estimation on DMRS, and predicting wireless link failures.

4. The first node according to any one of claims 1 to 3, characterized in that, include: The first processor sends the first information block; Wherein, the first information block is used to indicate N1 applicable functions of the first node from N functions, all of which are based on AI, where N is a positive integer greater than 1 and N1 is a positive integer not greater than N; the amount of resources occupied by each of the N1 applicable functions is equal to or less than J.

5. The first node according to claim 4, characterized in that, The N1 applicable functions satisfy the first condition; the first condition includes: the total number of resources occupied by all applicable functions is equal to or less than J.

6. The first node according to claim 4 or 5, characterized in that, Whether one of the N functions is the applicable function of the first node depends on the purpose of the function.

7. The first node according to any one of claims 4 to 6, characterized in that, include: The first processor performs inference; The reasoning performed by the first node is used to implement at least one of the N1 applicable functions.

8. A second node used for wireless communication, characterized in that, include: The second processor receives the first message; Wherein, the first node is the sender of the first message, the first message indicates multiple functions supported by the first node and J resources in the first node, where J is a positive integer greater than 1; the multiple functions are all based on AI, only some of the multiple functions are used for channel information reporting, and any one of the multiple functions occupies at least one of the J resources.

9. A method used in a first node of wireless communication, characterized in that, include: Send the first message; The first message indicates multiple functions supported by the first node and J resources in the first node, where J is a positive integer greater than 1; the multiple functions are all based on AI, only some of the multiple functions are used for channel information reporting, and any one of the multiple functions occupies at least one of the J resources.

10. A method used in a second node for wireless communication, characterized in that, include: Receive the first message; Wherein, the first node is the sender of the first message, the first message indicates multiple functions supported by the first node and J resources in the first node, where J is a positive integer greater than 1; the multiple functions are all based on AI, only some of the multiple functions are used for channel information reporting, and any one of the multiple functions occupies at least one of the J resources.

Images