Method and apparatus in node used for wireless communication

Abstract

The invention discloses a method and an apparatus in a node used for wireless communication. The node receives the first request; generating a first result through the first model as a response to receiving the first request; the first model is used for at least one of training or reasoning; the identifier of the first model is a first identifier, and the first identifier is a non-negative integer; the first identifier is associated with a first index, and the first index is one of a plurality of indexes included in a first index set; the first index set is associated to a first sub-identifier set; the plurality of types of data indicated by the plurality of sub-identifiers included in the first sub-identifier set comprise at least one of channel quality information data, codebook data and position auxiliary information data; the sub-identifier associated with the first identifier depends on the first index. According to the method, the design of the model identifier and the matching of the data in the system adopting human intelligence and machine learning are optimized, so that the overall performance of the system is optimized.

Description

Technical Field

[0001] The present application relates to a transmission method and device in a wireless communication system, and more particularly to a scheme and device for model configuration in a wireless communication system. Background Art

[0002] The application scenarios of future wireless communication systems are becoming more and more diversified, and different application scenarios have different performance requirements for the system. In order to meet the different performance requirements of various application scenarios, the 3GPP (3rd Generation Partner Project) RAN (Radio Access Network) #72 plenary meeting decided to study the new air interface technology (NR, New Radio) (or 5G), and the WI (Work Item, Work Item) of the new air interface technology (NR, New Radio) was passed at the 3GPP RAN #75 plenary meeting, and the standardization work on NR began. In NR R (release) 18, the research on AI (Artificial Intelligence) / ML (Machine Learning) technology was established to explore its impact on system performance and system design. Compared with traditional processing methods, AI / ML has the characteristics of being based on training and requiring deployment. Summary of the invention

[0003] The applicant has found through research that when AI / ML functions are introduced, the existing data in the existing system can be applied to terminals or network devices with AI / ML capabilities to achieve training / inference functions. However, there are many different types of data in the existing system, as well as data for different application scenarios. How to combine this data with AI / ML is a problem that needs to be studied and solved.

[0004] In response to the above problems, the present application discloses a solution. It should be noted that although a large number of embodiments of the present application are developed for AI / ML, the present application is also applicable to other solutions, such as traditional non-AI / ML-based solutions. In addition, the use of a unified solution for different scenarios (including but not limited to AI / ML-based solutions) also helps to reduce hardware complexity and cost. In the absence of conflict, the embodiments in the first node of the present application and the features in the embodiments can be applied to the second node, and vice versa. In the absence of conflict, the embodiments of the present application and the features in the embodiments can be arbitrarily combined with each other.

[0005] In particular, the interpretation of the terms, nouns, functions, and variables in this application (if not otherwise specified) may refer to the definitions in the TS38 series and TS37 series in the technical specifications (TS) of 3GPP (the 3rd Generation Partnership Project). If necessary, reference may be made to TS38.211, TS38.212, TS38.213, TS38.214, TS38.215, TS38.300, TS38.304, TS38.305, TS38.321, TS38.331, TS37.355, TS38.423, TS28.105, TR28.908, and TS38.843 in the 3GPP technical standards to assist in understanding this application.

[0006] As an example, the interpretation of the terms in the present application refers to the definitions of the TS38 series of specification protocols of 3GPP.

[0007] As an example, the interpretation of the terms in the present application refers to the definitions of the TS37 series of specification protocols of 3GPP.

[0008] As an example, the interpretation of the terms in the present application refers to the definition of the TS40 series of specification protocols of 3GPP.

[0009] As an example, the interpretation of the terms in the present application refers to the definitions of the TS39 series of specification protocols of 3GPP.

[0010] As an example, the interpretation of the terms in the present application refers to the definitions of the TS28 series of specification protocols of 3GPP.

[0011] As an example, the interpretation of the terms in the present application refers to the definitions of the TR28 series of specification protocols of 3GPP.

[0012] As an example, the interpretation of the terms in the present application refers to the definitions in the Rel-17 version of the specification protocol of 3GPP.

[0013] As an example, the interpretation of the terms in the present application refers to the definitions in the Rel-18 version of the specification protocol of 3GPP.

[0014] As an example, the interpretation of the terms in the present application refers to the definitions in the Rel-19 version of the specification protocol of 3GPP.

[0015] As an example, the interpretation of the terms in the present application refers to the definitions in the Rel-20 version of the specification protocol of 3GPP.

[0016] The present application discloses a method in a first node used for wireless communication, which includes:

[0017] receiving a first request; generating a first result by a first model in response to receiving the first request;

[0018] Among them, the first model is for at least one of training or reasoning, and the first result is a training result or the first result is an reasoning result; the identifier of the first model is a first identifier, and the first identifier is a non-negative integer; the first identifier is associated with a first index, and the first index is one of a plurality of indexes included in a first index set; the plurality of indexes correspond to a plurality of training types, or the plurality of indexes correspond to a plurality of reasoning types, or the plurality of indexes correspond to a plurality of time windows; the first index set is associated with a first sub-identifier set, and any sub-identifier included in the first sub-identifier set indicates a class of data; the plurality of classes of data indicated by the plurality of sub-identifiers included in the first sub-identifier set include at least one of channel quality information data, codebook data, and location assistance information data; the sub-identifier associated with the first identifier depends on the first index.

[0019] As an embodiment, the problems to be solved by this application include: the problem of matching models and data in AI / ML scenarios.

[0020] As an embodiment, the problems to be solved by the present application include: matching different types of data for training or prediction in different application scenarios in AI / ML scenarios.

[0021] As an embodiment, the characteristics to be solved by the present application include: classifying data according to type, and different types of data are adopted by different application scenarios, thereby improving the performance and efficiency of the model.

[0022] As an embodiment, the characteristics of the above method include: configuring the model with identifiers and sub-identifier groups, thereby more flexibly de-associating the adopted data.

[0023] According to one aspect of the present application, the above method is characterized in that the first identifier is associated with a first sub-identifier group, any sub-identifier in the first sub-identifier group is a sub-identifier in the first sub-identifier set, and the data indicated by any sub-identifier included in the first sub-identifier group is used to generate the first result.

[0024] As an embodiment, the characteristics of the above method include: combining the first sub-identification group and the first index, and then selecting different types of data for different application scenarios to improve model accuracy.

[0025] As an embodiment, the characteristics of the above method include: one model can be applied to different scenarios, further improving the versatility of the model and reducing implementation costs.

[0026] According to one aspect of the present application, the above method is characterized in that at least one sub-identifier included in the first sub-identifier set is associated with a cell set, and any cell in the cell set can use the data indicated by the at least one sub-identifier.

[0027] As an embodiment, the characteristics of the above method include: the data indicated by the sub-identifier can be shared among different cells, thereby improving the utilization rate of the data and also improving the accuracy of the model.

[0028] According to one aspect of the present application, the above method is characterized in that at least one of the multiple sub-identifiers is associated with a second identifier, the second identifier identifies a second model, and the second model is different from the first model for at least one of training or reasoning.

[0029] As an embodiment, the characteristics of the above method include: the model identifier used to identify the model and the sub-identifier used to identify the data are independently configured to ensure maximum configuration flexibility.

[0030] According to one aspect of the present application, the above method is characterized in that the training of the first model and the training of a type of data indicated by any sub-identifier included in the first sub-identifier group are independent.

[0031] As an embodiment, the characteristics of the above method include: separating the training of data and the training of the model to ensure accuracy and independence.

[0032] According to one aspect of the present application, the above method is characterized in that the channel quality information data includes CSI (Channel State Information).

[0033] As an embodiment, the characteristics of the above method include: using the CSI of the physical layer as part of the reference data for model training / prediction to improve performance.

[0034] According to one aspect of the present application, the above method is characterized in that the codebook data is a statistical result of the codebook adopted by the first node in a given time window.

[0035] As an embodiment, the characteristics of the above method include: using the codebook information of the physical layer as part of the reference data for model training / prediction to improve performance.

[0036] According to one aspect of the present application, the above method is characterized in that the position assistance information data includes at least one of positioning data or mobility data.

[0037] As an embodiment, the characteristics of the above method include: using high-level information as part of the reference data for model training / prediction to improve performance.

[0038] As an embodiment, the characteristics of the above method include: a model is trained / predicted by physical layer data and high-level data at the same time to further improve the overall performance of the model.

[0039] According to one aspect of the present application, the above method is characterized in that it includes:

[0040] The first result is sent.

[0041] As an embodiment, the characteristics of the above method include: sending the first result to the sender of the first request to improve system performance by adopting AI / ML solution.

[0042] The present application discloses a method in a second node used for wireless communication, which includes:

[0043] Send the first request;

[0044] Among them, the recipient of the first request generates a first result through a first model as a response to receiving the first request; the first model is for at least one of training or inference, and the first result is a training result or the first result is an inference result; the identifier of the first model is a first identifier, and the first identifier is a non-negative integer; the first identifier is associated with a first index, and the first index is one of a plurality of indexes included in a first index set; the plurality of indexes correspond to a plurality of training types, or the plurality of indexes correspond to a plurality of inference types, or the plurality of indexes correspond to a plurality of time windows; the first index set is associated with a first sub-identifier set, and any sub-identifier included in the first sub-identifier set indicates a class of data; the plurality of classes of data indicated by the plurality of sub-identifiers included in the first sub-identifier set include at least one of channel quality information data, codebook data, and location assistance information data; the sub-identifier associated with the first identifier depends on the first index.

[0045] According to one aspect of the present application, the above method is characterized in that the first identifier is associated with a first sub-identifier group, any sub-identifier in the first sub-identifier group is a sub-identifier in the first sub-identifier set, and the data indicated by any sub-identifier included in the first sub-identifier group is used to generate the first result.

[0046] According to one aspect of the present application, the above method is characterized in that at least one sub-identifier included in the first sub-identifier set is associated with a cell set, and any cell in the cell set can use the data indicated by the at least one sub-identifier.

[0047] According to one aspect of the present application, the above method is characterized in that at least one of the multiple sub-identifiers is associated with a second identifier, the second identifier identifies a second model, and the second model is different from the first model for at least one of training or reasoning.

[0048] According to one aspect of the present application, the above method is characterized in that the training of the first model and the training of a type of data indicated by any sub-identifier included in the first sub-identifier group are independent.

[0049] According to one aspect of the present application, the above method is characterized in that the channel quality information data includes CSI.

[0050] According to one aspect of the present application, the above method is characterized in that the codebook data is a statistical result of the codebook adopted by the first node in a given time window.

[0051] According to one aspect of the present application, the above method is characterized in that the position assistance information data includes at least one of positioning data or mobility data.

[0052] According to one aspect of the present application, the above method is characterized in that it includes:

[0053] The first result is received.

[0054] The present application discloses a device for a first node used for wireless communication, comprising:

[0055] A first receiver receives a first request; in response to receiving the first request, generates a first result through a first model;

[0056] Among them, the first model is for at least one of training or reasoning, and the first result is a training result or the first result is an reasoning result; the identifier of the first model is a first identifier, and the first identifier is a non-negative integer; the first identifier is associated with a first index, and the first index is one of a plurality of indexes included in a first index set; the plurality of indexes correspond to a plurality of training types, or the plurality of indexes correspond to a plurality of reasoning types, or the plurality of indexes correspond to a plurality of time windows; the first index set is associated with a first sub-identifier set, and any sub-identifier included in the first sub-identifier set indicates a class of data; the plurality of classes of data indicated by the plurality of sub-identifiers included in the first sub-identifier set include at least one of channel quality information data, codebook data, and location assistance information data; the sub-identifier associated with the first identifier depends on the first index.

[0057] The present application discloses a device for a second node used for wireless communication, comprising:

[0058] A second transmitter sends a first request;

[0059] Among them, the recipient of the first request generates a first result through a first model as a response to receiving the first request; the first model is for at least one of training or inference, and the first result is a training result or the first result is an inference result; the identifier of the first model is a first identifier, and the first identifier is a non-negative integer; the first identifier is associated with a first index, and the first index is one of a plurality of indexes included in a first index set; the plurality of indexes correspond to a plurality of training types, or the plurality of indexes correspond to a plurality of inference types, or the plurality of indexes correspond to a plurality of time windows; the first index set is associated with a first sub-identifier set, and any sub-identifier included in the first sub-identifier set indicates a class of data; the plurality of classes of data indicated by the plurality of sub-identifiers included in the first sub-identifier set include at least one of channel quality information data, codebook data, and location assistance information data; the sub-identifier associated with the first identifier depends on the first index.

[0060] As an embodiment, compared with the traditional solution, the present application has the following advantages but not limited to:

[0061] - Match different types of data for training or prediction in different application scenarios under AI / ML scenarios;

[0062] -Classify data by type. Different types of data are used in different application scenarios, thereby improving the performance and efficiency of the model;

[0063] -Configure model identifiers and sub-identifier groups to more flexibly disassociate the adopted data, and data can be shared among multiple cells to improve data usage efficiency and thus improve overall system performance;

[0064] -Data training and model training are performed independently to ensure performance;

[0065] -The data of a model comes from data at multiple different layers to further improve the performance of the model. BRIEF DESCRIPTION OF THE DRAWINGS

[0066] Other features, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments with reference to the following drawings:

[0067] Figure 1 A flowchart showing first node transmission according to an embodiment of the present application is shown;

[0068] Figure 2 A schematic diagram of a network architecture according to an embodiment of the present application is shown;

[0069] Figure 3 A schematic diagram showing an embodiment of a wireless protocol architecture of a user plane and a control plane according to an embodiment of the present application;

[0070] Figure 4 A schematic diagram showing a first communication device and a second communication device according to an embodiment of the present application is shown;

[0071] Figure 5 A first flow chart showing transmission between a first node and a second node according to an embodiment of the present application is shown;

[0072] Figure 6 A schematic diagram of a first model according to an embodiment of the present application is shown;

[0073] Figure 7 A schematic diagram showing a first sub-identification group according to an embodiment of the present application is shown;

[0074] Figure 8 A schematic diagram of RAN (Radio Access Network) domain (Domain) AI / ML function deployment according to an embodiment of the present application is shown;

[0075] Fig. 9 A schematic diagram showing the deployment of AI / ML functions of a UE according to an embodiment of the present application is shown;

[0076] Fig.10 A schematic diagram of the structure of ML / AI of a UE according to an embodiment of the present application is shown;

[0077] Fig.11 A schematic diagram of a processing system based on artificial intelligence or machine learning according to an embodiment of the present application is shown;

[0078] Fig.12 A flowchart based on artificial intelligence or machine learning according to an embodiment of the present application is shown;

[0079] Fig.13 A structural block diagram of a processing device used in a first node according to an embodiment of the present application is shown;

[0080] Fig.14 A structural block diagram of a processing device used in a second node according to an embodiment of the present application is shown. DETAILED DESCRIPTION

[0081] The technical solution of the present application will be further described in detail below in conjunction with the accompanying drawings. It should be noted that, unless there is a conflict, the embodiments in the present application and the features in the embodiments can be combined with each other arbitrarily.

[0082] Example 1

[0083] Embodiment 1 illustrates a flowchart of the first node transmission according to an embodiment of the present application, as shown in the attached Figure 1 As shown in the attached Figure 1 In the example, each box represents a step. In particular, the order of the steps in the box does not represent a specific time sequence between the steps.

[0084] In step 101 , the first node receives a first request; in step 102 , in response to receiving the first request, the first node generates a first result through a first model.

[0085] In embodiment 1, the first model is for at least one of training or reasoning, and the first result is a training result or the first result is an reasoning result; the identifier of the first model is a first identifier, and the first identifier is a non-negative integer; the first identifier is associated with a first index, and the first index is one of a plurality of indexes included in a first index set; the plurality of indexes correspond to a plurality of training types, or the plurality of indexes correspond to a plurality of reasoning types, or the plurality of indexes correspond to a plurality of time windows; the first index set is associated with a first sub-identifier set, and any sub-identifier included in the first sub-identifier set indicates a class of data; the plurality of classes of data indicated by the plurality of sub-identifiers included in the first sub-identifier set include at least one of channel quality information data, codebook data, and location assistance information data; the sub-identifier associated with the first identifier depends on the first index.

[0086] As an embodiment, the first request triggers the first model.

[0087] As an embodiment, the first request includes a training request.

[0088] As an embodiment, the first request includes an inference request.

[0089] As an embodiment, the first request is a request for ML.

[0090] As an embodiment, the first request is a request for AI.

[0091] As an embodiment, the first node generates the first result through the first model.

[0092] As a sub-embodiment of this embodiment, the first node includes the first model.

[0093] As an embodiment, the first node generates the first result through the first model located in the network.

[0094] As a sub-embodiment of this embodiment, the first node and the first model are connected via a network.

[0095] As a sub-embodiment of this embodiment, the first model is located in the cloud.

[0096] As an embodiment, the first model includes an AI model.

[0097] As an embodiment, the first model includes an ML model.

[0098] As an embodiment, the first model is for AI.

[0099] As an embodiment, the first model is for ML.

[0100] As an embodiment, the first model includes an ML entity.

[0101] As an embodiment, the first model includes MnS Producer.

[0102] As an embodiment, the first result includes a training result.

[0103] As an embodiment, the first result includes a prediction result.

[0104] As an embodiment, the first model is for training.

[0105] As an embodiment, the first model is for reasoning.

[0106] As an embodiment, the first model is for training and inference.

[0107] As an embodiment, the first identifier is a non-negative integer.

[0108] As an embodiment, the first identifier is a positive integer.

[0109] As an embodiment, the first identifier is an Identity.

[0110] As an embodiment, the first identifier is an Identifier.

[0111] As an embodiment, the first identifier is an Index.

[0112] As an embodiment, the first identifier being associated with the first index means that: the signaling indicating the first identifier also indicates the first index.

[0113] As an embodiment, the first identifier being associated with the first index means that the first identifier and the first index are configured to the first model at the same time.

[0114] As an embodiment, the meaning of the first identifier being associated with the first index includes: the first identifier identifies the first model, and the first index indicates the application type of the first model.

[0115] As an embodiment, the first index is a non-negative integer.

[0116] As an embodiment, the indices included in the first index set are all non-negative integers.

[0117] As an embodiment, any two indexes included in the first index set are different.

[0118] As an embodiment, the multiple training types are divided according to different needs.

[0119] As an embodiment, the multiple reasoning types are divided according to different requirements.

[0120] As an embodiment, the different requirements in this application are aimed at different application scenarios.

[0121] As an embodiment, the different requirements in this application are for different delays.

[0122] As an embodiment, the different requirements in this application are for different layers.

[0123] As an embodiment, the different requirements in this application are directed to different reliabilities.

[0124] As an embodiment, the different requirements in this application are for different confidence intervals.

[0125] As an embodiment, the multiple time windows correspond to multiple effective times respectively.

[0126] As an embodiment, the multiple time windows correspond to multiple processing delays respectively.

[0127] As an embodiment, the multiple time windows are respectively for multiple runtimes.

[0128] As an embodiment, the meaning of the first index set being associated with the first sub-identifier set includes: any index included in the first index set is associated with at least one sub-identifier in the first sub-identifier set.

[0129] As an embodiment, the meaning of the first index set being associated with the first sub-identifier set includes: the data used by any index included in the first index set are all data indicated by at least one sub-identifier in the first sub-identifier set.

[0130] As an embodiment, the meaning that the sub-identifier associated with the first identifier depends on the first index includes: the first index is used to determine the sub-identifier associated with the first identifier from the first sub-identifier set.

[0131] As an embodiment, the meaning that the sub-identifier associated with the first identifier depends on the first index includes: the first index is used to indicate the sub-identifier associated with the first identifier from the first sub-identifier set.

[0132] As an embodiment, the meaning of the sub-identifier associated with the first identifier being dependent on the first index includes: the multiple indexes included in the first index set are respectively associated with multiple sub-identifier groups, the sub-identifiers included in any sub-identifier group in the multiple sub-identifier groups belong to the first sub-identifier set, and the first index is used to determine the sub-identifier group associated with the first identifier from the multiple sub-identifier groups.

[0133] As an embodiment, the meaning that the sub-identifier associated with the first identifier depends on the first index includes: the training type corresponding to the first index is used to determine the sub-identifier associated with the first identifier from the first sub-identifier set.

[0134] As an embodiment, the meaning that the sub-identifier associated with the first identifier depends on the first index includes: the inference type corresponding to the first index is used to determine the sub-identifier associated with the first identifier from the first sub-identifier set.

[0135] As an embodiment, the meaning that the sub-identifier associated with the first identifier depends on the first index includes: the time window corresponding to the first index is used to determine the sub-identifier associated with the first identifier from the first sub-identifier set.

[0136] As an embodiment, any sub-identifier in the first sub-identifier set is a non-negative integer.

[0137] As an embodiment, any sub-identifier in the first sub-identifier set is a positive integer.

[0138] As an embodiment, the first sub-identifier set indicates the multiple data respectively.

[0139] As an embodiment, the multiple types of data correspond to multiple databases respectively.

[0140] As an embodiment, the first request indicates the first identifier.

[0141] As an embodiment, the first request indicates the first index associated with the first identifier.

[0142] As an embodiment, the input of the first model includes the multiple types of data indicated by the multiple sub-identifiers included in the first sub-identifier set.

[0143] As an embodiment, the input of the first model includes at least one of the multiple types of data indicated by the multiple sub-identifiers included in the first sub-identifier set.

[0144] As an embodiment, the generating of the first result by the first model depends on the multiple types of data.

[0145] As an embodiment, generating the first result by using the first model depends on at least one of the multiple types of data.

[0146] Example 2

[0147] Embodiment 2 illustrates a schematic diagram of a network architecture according to an embodiment of the present application, as shown in the attached Figure 2 shown.

[0148] Attached Figure 2A network architecture 200 is illustrated. The network architecture 200 is a network architecture of LTE (Long-Term Evolution), LTE-A (Long-Term Evolution Advanced), 5G system, 5G-Advanced and future 6G system. The network architecture of LTE, LTE-A, 5G system, 5G-Advanced and future 6G system is called EPS (Evolved Packet System). The 5GNR or LTE network architecture may be called 5GS (5G System) / EPS or some other suitable term; the 6G network architecture may be called 6GS (6G System) / EPS or some other suitable term. The network architecture 200 may include one or more UEs 201, RAN (Next Generation Radio Access Network) 202, a core network 210, an HSS (Home Subscriber Server) / UDM (Unified Data Management) 220 and an Internet service 230. . The network architecture 200 can be interconnected with other access networks, but for simplicity these entities / interfaces are not shown. Figure 2As shown, the network architecture 200 provides packet switching services, however, those skilled in the art will readily appreciate that the various concepts presented throughout this application can be extended to networks that provide circuit switching services. RAN 202 includes Node B 203 and other nodes 204. Node 203 provides user and control plane protocol terminations toward UE 201. Node 203 can be connected to other nodes 204 via an Xn interface (e.g., backhaul). Node 203 may also be referred to as a base station, a base transceiver station, a wireless base station, a wireless transceiver, a transceiver function, a Basic Service Set (BSS), an Extended Service Set (ESS), a TRP (Transmitter Receiver Point), or some other suitable term. Node 203 provides an access point to a core network 210 for UE 201; the core network 210 is a 5GC (5G Core Network, 5G core network) / EPC (Evolved Packet Core, Evolved Packet Core), or the core network 210 is a 6GC. Examples of UE 201 include cellular phones, smart phones, Session Initiation Protocol (SIP) phones, laptop computers, personal digital assistants (PDAs), satellite radios, global positioning systems, multimedia devices, video devices, digital audio players (e.g., MP3 players), cameras, game consoles, drones, aircraft, narrowband physical network devices, machine type communication devices, land vehicles, cars, wearable devices, or any other similar functional devices. Those skilled in the art may also refer to UE 201 as a mobile station, a subscriber station, a mobile unit, a subscriber unit, a wireless unit, a remote unit, a mobile device, a wireless device, a wireless communication device, a remote device, a mobile subscriber station, an access terminal, a mobile terminal, a wireless terminal, a remote terminal, a handset, a user agent, a mobile client, a client, or some other suitable term. Node 203 is connected to core network 210 via an S1 / NG interface.The core network 210 includes MME (Mobility Management Entity) / AMF (Authentication Management Field) / SMF (Session Management Function) 211, other MME / AMF / SMF 214, S-GW (Service Gateway) / UPF (User Plane Function) 212, and P-GW (Packet Data Network Gateway) / UPF 213. MME / AMF / SMF 211 is a control node that handles signaling between UE 201 and 5G-CN / EPC 210. In general, MME / AMF / SMF 211 provides bearer and connection management. All user IP (Internet Protocol) packets are transmitted through S-GW / UPF 212, which itself is connected to P-GW / UPF 213. P-GW provides UE IP address allocation and other functions. P-GW / UPF 213 is connected to Internet service 230. Internet service 230 includes operator-specific Internet protocol services, which may specifically include Internet, Intranet, IMS (IP Multimedia Subsystem) and packet switching services.

[0149] As an embodiment, the first node in the present application includes the UE 201.

[0150] As an embodiment, the second node in the present application includes the node 203.

[0151] As an embodiment, the node 203 is a macro cell base station.

[0152] As an embodiment, the node 203 is a micro cell base station.

[0153] As an embodiment, the node 203 is a pico cell base station.

[0154] As an embodiment, the node 203 is a home base station (Femtocell).

[0155] As an embodiment, the node 203 is a base station device that supports a large delay difference.

[0156] As an embodiment, the node 203 is a flying platform device.

[0157] As an embodiment, the node 203 is a satellite device.

[0158] As an embodiment, the node 203 is a test device (eg, a transceiver device simulating some functions of a base station, a signaling tester).

[0159] As an embodiment, the UE 201 includes a mobile phone.

[0160] As an embodiment, the UE 201 is a vehicle including a car.

[0161] As an embodiment, the wireless link from the UE 201 to the node 203 is an uplink, and the uplink is used to perform uplink transmission.

[0162] As an embodiment, the wireless link from the node 203 to the UE 201 is a downlink, and the downlink is used to perform downlink transmission.

[0163] As an embodiment, the wireless link between the node 203 and the UE 201 includes a cellular network link.

[0164] As an embodiment, the node 203 and the UE 201 are connected via a Uu air interface.

[0165] As an embodiment, the sender of the first request includes the node 203.

[0166] As an embodiment, the recipient of the first request includes the UE 201.

[0167] As an embodiment, the sender of the first result includes the UE 201.

[0168] As an embodiment, the recipient of the first result includes the node 203.

[0169] As an embodiment, the first model is maintained in the UE 201.

[0170] As an embodiment, the first model is maintained at the node 203 .

[0171] As an embodiment, the first model is maintained in both the UE 201 and the node 203 .

[0172] As an embodiment, the sender and the receiver of the first request are both located in the node 203 .

[0173] As an embodiment, the sender and receiver of the first request are both located in the UE 201.

[0174] As an embodiment, the UE 201 supports AI.

[0175] As an embodiment, the UE 201 supports ML.

[0176] As an embodiment, the gNB 203 supports AI.

[0177] As an embodiment, the gNB 203 supports ML.

[0178] As an embodiment, the UE 201 supports a 5G system.

[0179] As an embodiment, the node 203 supports a 5G system.

[0180] As an embodiment, the UE 201 at least supports the 6G system.

[0181] As an embodiment, the node 203 at least supports the 6G system.

[0182] Example 3

[0183] Embodiment 3 illustrates a schematic diagram of an embodiment of a wireless protocol architecture of a user plane and a control plane according to an embodiment of the present application, as shown in the attached figure. Figure 3 shown.

[0184] 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 3The wireless protocol architecture of the control plane 300 for the first communication node device (RSU (Road Side Unit) in UE or V2X (Vehicle to Everything), vehicle-mounted device or vehicle-mounted communication module) and the second node device (gNB, RSU in UE or V2X, vehicle-mounted device or vehicle-mounted communication module), or between two UEs is presented in three layers: Layer 1 (Layer 1, L1), Layer 2 (Layer 2, L2) and Layer 3 (Layer 3, L3). L1 is the lowest layer and implements various PHY (PHYsical layer) signal processing functions. L1 will be referred to as PHY 301 in this article. L2305 is above PHY 301 and is responsible for the link between the first node device and the second node device, or between two UEs through PHY301. L2305 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 node device. The PDCP sublayer 304 provides multiplexing between different radio bearers and logical channels. The PDCP sublayer 304 also provides security by encrypting data packets, and provides inter-zone mobility support for the first communication node device between the second communication node device. The RLC sublayer 303 provides segmentation and reassembly of upper layer data packets, retransmission of lost data packets, and reordering of data packets to compensate for out-of-order reception caused by HARQ (Hybrid Automatic Repeat Quest). The MAC sublayer 302 provides multiplexing between logical and transport channels. The MAC sublayer 302 is also responsible for allocating various wireless resources (e.g., resource blocks) in a cell between the first communication node devices. The MAC sublayer 302 is also responsible for HARQ operations. The RRC (Radio Resource Control) sublayer 306 in L3 in the control plane 300 is responsible for obtaining radio resources (ie, radio bearers) and configuring lower layers using RRC signaling between the second communication node device and the first communication node device.The wireless protocol architecture of the user plane 350 includes layer 1 (L1) and layer 2 (L2). The wireless protocol architecture for the first communication node device and the second communication node device in the user plane 350 is substantially the same as the corresponding layers and sublayers in the control plane 300 for the physical layer 351, the PDCP sublayer 354 in L2355, the RLC sublayer 353 in L2355, and the MAC sublayer 352 in L2355, but the PDCP sublayer 354 also provides header compression for upper layer data packets to reduce wireless transmission overhead. L2355 in the user plane 350 also includes a SDAP (Service Data Adaptation Protocol) sublayer 356, which is responsible for mapping between QoS (Quality of Service) flows and data radio bearers (DRBs) to support the diversity of services. Although not shown in the figure, the first communication node device may have several upper layers above L2355, including a network layer (e.g., IP (Internet Protocol) 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, a server, etc.).

[0185] As an example, Figure 3 The wireless protocol architecture in is applicable to the first node in this application.

[0186] As an example, Figure 3 The wireless protocol architecture in is applicable to the second node in this application.

[0187] As an embodiment, the first request is generated in the RRC 306.

[0188] As an embodiment, the first result is generated in the RRC 306.

[0189] As an embodiment, the first request is generated by the MAC 302 or MAC 352.

[0190] As an embodiment, the first result is generated by the MAC 302 or MAC 352.

[0191] As an embodiment, the first request is generated by the PHY301 or PHY351.

[0192] As an embodiment, the first result is generated by the PHY301 or PHY351.

[0193] As an embodiment, the higher layer mentioned in this application refers to a layer above the physical layer.

[0194] As an embodiment, the higher layer described in the present application includes a MAC layer.

[0195] As an embodiment, the higher layer described in the present application includes an RRC layer.

[0196] Example 4

[0197] Embodiment 4 illustrates a schematic diagram of a first communication device and a second communication device according to an embodiment of the present application, as shown in the attached Figure 4 Attached 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.

[0198] The first communication device 410 includes a controller / processor 475 , a memory 476 , a receive processor 470 , a transmit processor 416 , a multi-antenna receive processor 472 , a multi-antenna transmit processor 471 , a transmitter / receiver 418 and an antenna 420 .

[0199] The second communication device 450 includes a controller / processor 459, a memory 460, a data source 467, a transmit processor 468, a receive processor 456, a multi-antenna transmit processor 457, a multi-antenna receive processor 458, a transmitter / receiver 454 and an antenna 452.

[0200] In 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 the functionality of L2. In DL, 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 operations, 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 (i.e., physical layer). The transmit processor 416 implements coding and interleaving to facilitate forward error correction (FEC) at the second communication device 450, as well as mapping of signal constellations based on various modulation schemes (e.g., binary phase shift keying (BPSK), quadrature phase shift keying (QPSK), M-PSK, M-quadrature amplitude modulation (M-QAM)). The multi-antenna transmit processor 471 performs digital spatial precoding on the coded and modulated symbols, including codebook-based precoding and non-codebook-based precoding and beamforming processing, to generate one or more parallel streams. The transmit processor 416 then maps each parallel stream to a subcarrier, multiplexes the modulated symbols with a reference signal (e.g., pilot) in the time domain and / or frequency domain, and then uses an inverse fast Fourier transform (IFFT) to generate a physical channel carrying a time domain multi-carrier symbol stream. The multi-antenna transmit processor 471 then performs a transmit analog precoding / beamforming operation on the time domain multi-carrier symbol stream. Each transmitter 418 converts the baseband multi-carrier symbol stream provided by the multi-antenna transmit processor 471 into a radio frequency stream, and then provides it to different antennas 420.

[0201] 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 RF carrier and converts the RF stream into a baseband multi-carrier symbol stream and provides it to the receiving processor 456. The receiving processor 456 and the multi-antenna receiving processor 458 implement various signal processing functions of L1. The multi-antenna receiving processor 458 performs a receiving analog precoding / beamforming operation on the baseband multi-carrier symbol stream from the receiver 454. The receiving processor 456 uses a fast Fourier transform (FFT) to convert the baseband multi-carrier symbol stream after the receiving analog precoding / beamforming operation 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 receiving processor 456, where the reference signal will be used for channel estimation, and the data signal is recovered after multi-antenna detection in the multi-antenna receiving processor 458 to any parallel stream with the second communication device 450 as the destination. The symbols on each parallel stream are demodulated and recovered in the receiving processor 456, and soft decisions are generated. The receiving processor 456 then decodes and deinterleaves the soft decisions to recover the upper layer data and control signals transmitted by the first communication device 410 on the physical channel. The upper layer data and control signals are then provided to the controller / processor 459. The controller / processor 459 implements the functions of L2. The controller / processor 459 may be associated with a memory 460 storing program codes and data. The memory 460 may be referred to as a computer-readable medium. In DL, the controller / processor 459 provides multiplexing, packet reassembly, decryption, header decompression, and control signal processing between transmission and logical channels to recover the upper layer data packets from the core network. The upper layer data packets are then provided to all protocol layers above L2. Various control signals may also be provided to L3 for L3 processing. The controller / processor 459 is also responsible for error detection using confirmation (ACKnowledgement, ACK) and / or negative confirmation (NegativeACKnowledgement, NACK) protocols to support HARQ operations.

[0202] 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 L2. Similar to the transmission function at the first communication device 410 described in 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, and implements L2 functions for the user plane and the control plane. The controller / processor 459 is also responsible for HARQ operations, retransmission of lost packets, and signaling to the first communication device 410. The transmit processor 468 performs modulation mapping and channel coding processing, and the multi-antenna transmit processor 457 performs digital multi-antenna spatial precoding, including codebook-based precoding and non-codebook-based precoding, and beamforming processing. Then, the transmit processor 468 modulates the generated parallel stream into a multi-carrier / single-carrier symbol stream, which is then provided to different antennas 452 via the transmitter 454 after analog precoding / beamforming operations in the multi-antenna transmit processor 457. Each transmitter 454 first converts the baseband symbol stream provided by the multi-antenna transmit processor 457 into a radio frequency symbol stream, and then provides it to the antenna 452.

[0203] 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 reception 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 a radio frequency signal through its corresponding antenna 420, converts the received radio frequency signal into a baseband signal, and provides the baseband signal to the multi-antenna reception processor 472 and the reception processor 470. The reception processor 470 and the multi-antenna reception processor 472 jointly implement the functions of L1. The controller / processor 475 implements the L2 functions. The controller / processor 475 can be associated with a memory 476 storing program codes and data. The memory 476 can be referred to as a computer-readable medium. The controller / processor 475 provides demultiplexing between transmission and logical channels, packet reassembly, decryption, header decompression, control signal processing to recover the upper layer data packets from the second communication device 450. The upper layer data packets from the controller / processor 475 can be provided to the core network. The controller / processor 475 is also responsible for error detection using an ACK and / or NACK protocol to support HARQ operations.

[0204] As an embodiment, the second communication device 450 includes: at least one processor and at least one memory, the at least one memory includes 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 device receives at least a first request; and as a response to receiving the first request, generates a first result through a first model; the first model is for at least one of training or reasoning, the first result is a training result or the first result is a reasoning result; the identifier of the first model is a first identifier, the first identifier is a non-negative integer; the first identifier is associated with a first index, the first index is one of a plurality of indexes included in a first index set; the plurality of indexes correspond to a plurality of training types, or the plurality of indexes correspond to a plurality of reasoning types, or the plurality of indexes correspond to a plurality of time windows; the first index set is associated with a first sub-identifier set, any sub-identifier included in the first sub-identifier set indicates a type of data; the plurality of sub-identifiers included in the first sub-identifier set indicate a type of data; the plurality of sub-identifiers included in the first sub-identifier set indicate a type of data including at least one of channel quality information data, codebook data, and location auxiliary information data; the sub-identifier associated with the first identifier depends on the first index.

[0205] As an embodiment, the second communication device 450 includes: a memory storing a computer-readable instruction program, wherein the computer-readable instruction program generates an action when executed by at least one processor, wherein the action includes: receiving a first request; and generating a first result through a first model in response to receiving the first request.

[0206] As an embodiment, the first communication device 410 includes: at least one processor and at least one memory, the at least one memory includes 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 device at least sends a first request; the recipient of the first request generates a first result through a first model in response to receiving the first request; the first model is for at least one of training or reasoning, and the first result is a training result or the first result is a reasoning result; the identifier of the first model is a first identifier, and the first identifier is a non-negative integer; the first identifier is associated with a first index, and the first index is one of a plurality of indexes included in a first index set; the plurality of indexes correspond to a plurality of training types, or the plurality of indexes correspond to a plurality of reasoning types, or the plurality of indexes correspond to a plurality of time windows; the first index set is associated with a first sub-identifier set, and any sub-identifier included in the first sub-identifier set indicates a type of data; the plurality of sub-identifiers included in the first sub-identifier set indicate a type of data; the plurality of sub-identifiers included in the first sub-identifier set indicate a type of data including at least one of channel quality information data, codebook data, and location auxiliary information data; the sub-identifier associated with the first identifier depends on the first index.

[0207] As an embodiment, the first communication device 410 includes: a memory storing a computer-readable instruction program, wherein the computer-readable instruction program generates an action when executed by at least one processor, and the action includes: sending a first request.

[0208] As an embodiment, the first node in the present application includes the second communication device 450.

[0209] As an embodiment, the second node in the present application includes the first communication device 410.

[0210] As an embodiment, at least one of {the antenna 420, the transmitter 418, the transmit processor 416, the multi-antenna transmit processor 471, the controller / processor 475, and the memory 476} is used to send a first request; and at least one of {the antenna 452, the receiver 454, the receive processor 456, the multi-antenna receive processor 458, the controller / processor 459, the memory 460, and the data source 467} is used to receive a first request.

[0211] As an embodiment, at least one of {the antenna 420, the transmitter 418, the transmit processor 416, the multi-antenna transmit processor 471, the controller / processor 475, and the memory 476} is used to send the first result; and at least one of {the antenna 452, the receiver 454, the receive processor 456, the multi-antenna receive processor 458, the controller / processor 459, the memory 460, and the data source 467} is used to receive the first result.

[0212] Example 5

[0213] Embodiment 5 illustrates a first flow chart of transmission between a first node and a second node according to an embodiment of the present application, as shown in the attached Figure 5 As shown in the attached Figure 5 In the embodiment, the steps in block F51 are optional. It should be noted that the order in this embodiment does not limit the signal transmission order and implementation order in this application.

[0214] For the first node U1, a first request is received in step S510; a first result is generated by a first model in response to receiving the first request in step S511; and the first result is sent in step S512.

[0215] For the second node N2, a first request is sent in step S520; and a first result is received in step S521.

[0216] In Example 5, the first model is for at least one of training or reasoning, and the first result is a training result or the first result is an reasoning result; the identifier of the first model is a first identifier, and the first identifier is a non-negative integer; the first identifier is associated with a first index, and the first index is one of a plurality of indexes included in a first index set; the plurality of indexes correspond to a plurality of training types, or the plurality of indexes correspond to a plurality of reasoning types, or the plurality of indexes correspond to a plurality of time windows; the first index set is associated with a first sub-identifier set, and any sub-identifier included in the first sub-identifier set indicates a class of data; the plurality of classes of data indicated by the plurality of sub-identifiers included in the first sub-identifier set include at least one of channel quality information data, codebook data, and location assistance information data; the sub-identifier associated with the first identifier depends on the first index.

[0217] Typically, at least one sub-identifier included in the first sub-identifier set is associated with a cell set, and any cell in the cell set can use the data indicated by the at least one sub-identifier.

[0218] As an embodiment, the cell set includes at least one cell.

[0219] As an embodiment, the cell set includes multiple cells.

[0220] As an embodiment, a terminal in any cell in the cell set can use the data indicated by the at least one sub-identifier.

[0221] As an embodiment, the gNB corresponding to any cell in the cell set can use the data indicated by the at least one sub-identifier.

[0222] As an embodiment, the cell set depends on the at least one sub-identifier among the multiple sub-identifiers.

[0223] As an embodiment, the at least one sub-identifier included in the first sub-identifier set is used to determine a cell set.

[0224] As an embodiment, the data indicated by the at least one sub-identifier is shared among cells in the cell set.

[0225] Typically, the training of the first model and the training of a type of data indicated by any sub-identifier included in the first sub-identifier group are independent.

[0226] As an embodiment, the training of the first model does not rely on the training of a type of data indicated by any sub-identifier included in the first sub-identifier group.

[0227] As an embodiment, the training of the first model depends on a type of data indicated by any sub-identifier included in the first sub-identifier group.

[0228] As an embodiment, the training of a type of data indicated by any sub-identifier included in the first sub-identifier group does not depend on the training of the first model.

[0229] Typically, the channel quality information data includes CSI.

[0230] As an embodiment, the channel quality information data is based on statistics.

[0231] As an embodiment, the channel quality information data is periodic.

[0232] As an embodiment, the channel quality information data corresponds to a statistical time window.

[0233] As an embodiment, the channel quality information data corresponds to an effective time window.

[0234] As an embodiment, the CSI includes CQI (Channel Quality Indicator).

[0235] As an embodiment, the CSI includes PMI (Precoding Matrix Indicator).

[0236] As an embodiment, the CSI includes CRI (CSI-RS resource indicator, channel state information reference signal resource indicator).

[0237] As an embodiment, the CSI includes SSBRI (SS / PBCH Block Resource indicator, synchronization / physical broadcast channel block resource indication).

[0238] As an embodiment, the CSI includes LI (layer indicator).

[0239] As an embodiment, the CSI includes RI (rank indicator).

[0240] As an embodiment, the CSI includes L1-RSRP (Layer 1 Reference Signal Received Power, layer 1 reference signal received power).

[0241] As an embodiment, the channel quality information data includes CSI generated by the first node in a given time window.

[0242] As an embodiment, the sub-identifier corresponding to the channel quality information data is associated with a first cell set, and the channel quality information data includes the reported CSI in the cells included in the first cell set.

[0243] Typically, the codebook data is a statistical result of the codebook used by the first node in a given time window.

[0244] As an embodiment, the codebook data is based on statistics.

[0245] As an embodiment, the codebook data is periodic.

[0246] As an embodiment, the codebook data corresponds to a statistical time window.

[0247] As an embodiment, the codebook data corresponds to an effective time window.

[0248] As an embodiment, the codebook data includes a codebook adopted by the first node in a given time window.

[0249] As an embodiment, the sub-identifier corresponding to the codebook data is associated with a second cell set, and the codebook data includes a codebook used by terminals in cells included in the second cell set.

[0250] Typically, the location assistance information data includes at least one of positioning data or mobility data.

[0251] As an embodiment, the position assistance information data is based on statistics.

[0252] As an embodiment, the position assistance information data is periodic.

[0253] As an embodiment, the position assistance information data corresponds to a statistical time window.

[0254] As an embodiment, the position assistance information data corresponds to an effective time window.

[0255] As an embodiment, the location assistance information data includes location assistance information of the first node in a given time window.

[0256] As an embodiment, the sub-identifier corresponding to the location assistance information data is associated with a third cell set, and the location assistance information data includes location assistance information of terminals in cells included in the third cell set.

[0257] As an embodiment, the positioning data includes the accuracy and latitude of the terminal.

[0258] As an embodiment, the positioning data includes the cell where the terminal resides.

[0259] As an embodiment, the positioning data includes a base station providing services for the terminal.

[0260] As an embodiment, the positioning data includes a spatial transmission parameter or a spatial reception parameter adopted by the terminal.

[0261] As an embodiment, the positioning data includes a UEPositioningAssistanceInfo message.

[0262] As an embodiment, the positioning data includes location information of the terminal acquired by a positioning server (Location Server).

[0263] As an embodiment, the mobility data includes a movement trajectory of the terminal.

[0264] As an embodiment, the mobility data includes a list of cells where the terminal resides.

[0265] As an embodiment, the mobility data includes a list of base stations providing services for the terminal.

[0266] As an embodiment, the multiple types of data include UE variable VarLogMeasConfig.

[0267] As an embodiment, the multiple types of data include UE variable VarLogMeasReport.

[0268] As an embodiment, the multiple types of data include UE variable VarAppLayerIdleConfig.

[0269] As an embodiment, the multiple types of data include UE variable VarAppLayerPLMN-ListConfig.

[0270] As an embodiment, the multiple types of data include UE variable VarConditionalReconfig.

[0271] As an embodiment, the multiple types of data include UE variable VarConnEstFailReport.

[0272] As an embodiment, the multiple types of data include UE variable VarConnEstFailReportList.

[0273] As an embodiment, the multiple types of data include UE variable VarLTM-Config.

[0274] As an embodiment, the multiple types of data include UE variable VarLTM-ServingCellNoResetID.

[0275] As an embodiment, the multiple types of data include UE variable VarLTM-ServingCellUE-MeasuredTA-ID.

[0276] As an embodiment, the multiple types of data include UE variable VarMeasReportList.

[0277] As an embodiment, the multiple types of data include statistics of terminal RLF (Radio Link Failure).

[0278] As an embodiment, the multiple types of data include terminal BFR (Beam Failure Recovery) statistics.

[0279] As an embodiment, the multiple types of data include data for RLF.

[0280] As an embodiment, the multiple types of data include data for BFR.

[0281] As an embodiment, the physical layer channel occupied by the first request includes PDCCH (Physical Downlink Control Channel).

[0282] As an embodiment, the physical layer channel occupied by the first request includes PDSCH (Physical Downlink Shared Channel).

[0283] As an embodiment, the physical layer channel occupied by the first result includes PUCCH (Physical Uplink Control Channel).

[0284] As an embodiment, the physical layer channel occupied by the first result includes PUSCH (Physical Uplink Shared Channel).

[0285] Example 6

[0286] Example 6 illustrates a schematic diagram of a first model according to an embodiment of the present application, as shown in the attached Figure 6 As shown in the attached Figure 6 In the invention, the first model is identified by a first identifier, and the first identifier is associated with a first index, and the first index is one of index #1 to index #K; the index #1 to index #K constitute the first index set of this application; wherein K is a positive integer greater than 1.

[0287] As an embodiment, the index #1 to index #K correspond to K training types respectively.

[0288] As an embodiment, the index #1 to index #K correspond to K inference types respectively.

[0289] As an embodiment, the index #1 to index #K correspond to K time windows respectively.

[0290] As an embodiment, the first index depends on the capabilities of the first node.

[0291] As an embodiment, the first index depends on the categories of the first node.

[0292] Example 7

[0293] Embodiment 7 illustrates a schematic diagram of a first sub-identification group according to an embodiment of the present application, as shown in the attached figure. Figure 7 As shown in the attached Figure 7 In the figure, the first sub-identifier group includes multiple sub-identifiers, and the sub-identifier #1 to sub-identifier #M shown in the figure are the multiple sub-identifiers, and the sub-identifier #1 to sub-identifier #M correspond to data #1 to data #M respectively; wherein M is a positive integer greater than 1.

[0294] As an embodiment, any two sub-identifiers among the multiple sub-identifiers are different.

[0295] As an embodiment, the data #1 to data #M correspond to the multiple data in this application.

[0296] As an embodiment, at least two data among the data #1 to data #M come from different layers.

[0297] As an embodiment, there are at least two data among the data #1 to data #M, each of which is for different performance requirements.

[0298] As an embodiment, the multiple data depend on the capabilities of the first node.

[0299] As an embodiment, the multiple data depend on the categories of the first node.

[0300] Example 8

[0301] Embodiment 8 illustrates a schematic diagram of RAN (Radio Access Network) domain (Domain) AI / ML function deployment according to an embodiment of the present application; as shown in the attached Figure 8 The gNB in ​​Example 8 may be replaced by a network device such as an eNB or a 6G base station.

[0302] AI / ML related functions include ML training function (also called AI training, or AI / ML training), ML testing function, ML reasoning (also called AI reasoning, or AI / ML reasoning) function, etc. ML training function, ML testing function, and ML reasoning function can be deployed independently or co-located. The deployment of AI / ML related functions can be achieved through software, such as downloading and / or running executable files; it can also be achieved through a combination of software and hardware, such as accelerating specific computing units through hardware to increase computing speed or save power consumption.

[0303] For ML training functions, they 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, the ML training function for MDA (Management Data Analytics) can be deployed in MDAF (MDA function); ML training for network data analysis can be deployed in NWDAF (Network Data Analytics Function), that is, the ML training function is MTLF (Model Training logical function).

[0304] The ML reasoning function can also be deployed in a cross-domain management system or a domain-specific management system; for example, the ML reasoning function is MDAF, or the ML reasoning function is AnLF (Analytics logical function) located in NWDAF.

[0305] Similarly, ML testing capabilities can also be deployed in cross-domain management systems, or domain-specific management systems.

[0306] In Example 8, the RAN domain ML training function 802 is located in the RAN domain management function 803; and the ML reasoning function is located in the base station, that is, the AI / ML reasoning function 804 is located in the gNB805, and the AI / ML reasoning function 806 is located in the gNB807.

[0307] Attached Figure 8In the example, the management of the ML inference functions of multiple base stations is completed by the RAN domain management function 803, that is, data interaction with the RAN domain MnS (Management Service) consumer / cross-domain management 801 (see the attached Figure 8 ).

[0308] Optionally, the management of the ML reasoning function may also be performed by the base station itself, that is, each base station may independently interact with the RAN domain MnS consumer / cross-domain management 801 for data.

[0309] It should be noted that Example 8 is merely a non-restrictive implementation method; optionally, the ML training function of the RAN domain may also be deployed in the base station; or optionally, some base stations deploy the ML reasoning function and the ML training function of the RAN domain, while some base stations only deploy the ML reasoning function.

[0310] As an embodiment, a gNB (or base station) in Embodiment 8 is the second node of the present application.

[0311] As an embodiment, the first receiver in the present application includes an attached Figure 8 An AL / ML inference function in, i.e., 804 or 806.

[0312] As an embodiment, the first transmitter in the present application includes an attached Figure 8 An AL / ML inference function in, i.e., 804 or 806.

[0313] As an example, Figure 8 An AL / ML inference function performs ML training based on the target precoding vector; the target reference signal is different from the first reference signal; any reference signal among the multiple reference signals except the first reference signal is associated with a precoding vector, wherein the target precoding vector is associated with the target reference signal.

[0314] Example 9

[0315] Embodiment 9 illustrates a schematic diagram of the AI / ML function deployment of a UE according to an embodiment of the present application; Fig. 9 Attached Fig. 9 The RAN domain ML training functionality 905 is optional.

[0316] The UE function 904 is deployed in the first node of the present application, and the UE function 904 includes an AI / ML reasoning function 906; the AI / ML reasoning function 906 uses an ML model (also called an AI model) for reasoning; an ML model is usually trained before being used for AI / ML reasoning.

[0317] As an embodiment, the UE function 904 includes a RAN domain ML training function 905, and the RAN domain ML training function 905 runs the training data through the ML model to derive relevant losses, and adjusts the parameters of the ML model based on the calculated losses; the ML training includes at least one of ML initial training, ML re-training, and reinforcement learning.

[0318] 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 high demands on the processing capability of the UE side.

[0319] Optionally, the UE function 904 also includes a CN domain ML training function ( Fig. 9 not included).

[0320] Optionally, the UE function 904 further includes an AI / ML deployment function: Fig. 9 Not included, used to load ML models and data.

[0321] As an embodiment, the first node indicates whether the ML training function is supported (RAN domain or CN domain) through capability reporting, and the capability reporting is RRC signaling, or NAS (Non-Access Stratum) signaling.

[0322] As an embodiment, the ML model and related metadata are loaded by the first node from a network device or a remote server.

[0323] Optionally, the UE function 904 is a MnS (Management Service) producer (Producer), providing data to the CN domain MnF (Management Function) 901, and / or the RAN domain MnF902, and / or the cross-domain management system 903 for management or analysis (as shown by the double arrow 907).

[0324] Optionally, the UE function 904 is an MnS consumer, which loads data from the CN domain MnF (Management Function) 901, and / or the RAN domain MnF 902, and / or the cross-domain management system 903 for AI / ML related management, such as management data requests, ML model activation, and / or ML training, etc. (as shown by the double arrow 907).

[0325] As an embodiment, the first channel information in the present application is obtained through reasoning by the AI / ML reasoning function 906.

[0326] As an embodiment, the RAN domain ML training function 905 performs ML training according to a target precoding vector; wherein the target reference signal is different from the first reference signal; any reference signal among the multiple reference signals except the first reference signal is associated with a precoding vector, wherein the target precoding vector is associated with the target reference signal.

[0327] As an embodiment, the second processor includes an attached Fig. 9 An AL / ML reasoning function 906 in.

[0328] As an embodiment, the ML model is based on a neural network.

[0329] As an embodiment, the ML model is based on CNN (Conventional Neural Networks).

[0330] As an embodiment, the ML model is based on the Transformer architecture.

[0331] Example 10

[0332] Embodiment 10 illustrates a schematic diagram of the structure of ML / AI according to an embodiment of the present application, as shown in the attached Fig.10 As shown in the attached Fig.10 In the example, MLMnS Consumer and MLMnS Producer are connected through MnS, and MLMnS Producer is connected to the database.

[0333] As an embodiment, the MLMnS Consumer corresponds to the second node in this application.

[0334] As an embodiment, the MLMnS Producer corresponds to the first node in the present application.

[0335] As an embodiment, the MLMnS Producer includes the first model in the present application.

[0336] As an embodiment, the MLMnS Producer corresponds to the first model deployed in the first node in the present application.

[0337] As an embodiment, the MLMnS Consumer sends the first request and receives the first data.

[0338] As an embodiment, the MLMnS Producer receives the first request and sends the first data.

[0339] As an example, Fig.10 The database in corresponds to multiple data in this application.

[0340] As an embodiment, the MLMnS Consumer corresponds to the entity that requests the first result in the first node in the present application, and the MLMnS Producer corresponds to the entity that generates the first result in the first node in the present application.

[0341] As an embodiment, the first model implements the related generation of the first result.

[0342] As an embodiment, the first model generates the first result according to a maximum likelihood criterion.

[0343] As an embodiment, the first model generates the first result according to a maximum probability criterion.

[0344] As an embodiment, the first model generates the first result by itself.

[0345] As an embodiment, the first model generates the first result according to the training result.

[0346] As an embodiment, the first model generates the first result based on the multiple data training results in this application combined with prediction.

[0347] As an embodiment, the first result includes first data and a first confidence level, and the first confidence level is used to indicate a prediction of the accuracy of the first data.

[0348] As an embodiment, the first result includes first data and a first confidence interval, and the first confidence interval is used to indicate a prediction of the accuracy of the first data.

[0349] Embodiment 11

[0350] Embodiment 11 illustrates a schematic diagram of a processing system based on artificial intelligence or machine learning according to an embodiment of the present application; Fig.11 Attached Fig.11 The invention comprises processor A, processor B, processor C and processor D. In embodiment 11, processor A sends a first data set to processor B and sends a second data set to processor C; processor B generates a target first-category parameter group according to the first data set, and processor B sends the generated target first-category parameter group to processor C; processor C processes the second data set using the target first-category parameter group to obtain a first-category output, and processor C sends the first-category output to processor D. Fig.11 In the , first and second type of feedback are optional.

[0351] As an embodiment, the processor A includes a data collection module and a model training module.

[0352] As an embodiment, the processor B includes a data collection module and a model training module.

[0353] As an embodiment, the processor A generates the first result.

[0354] As an embodiment, the processor B generates the first result.

[0355] As an embodiment, the processor C generates the first result.

[0356] As an embodiment, the processor C sends the first type of feedback to the processor B, and the first type of feedback is used to trigger recalculation or update of the target first type parameter group.

[0357] As an embodiment, the processor D sends a second type of feedback to the processor A, and the second type of feedback is used to generate the first data set or the second data set, or the second type of feedback is used to trigger the sending of the first data set or the sending of the second data set.

[0358] As an embodiment, the processor C belongs to the first node, and the processor D belongs to the second node.

[0359] As an embodiment, the second data set includes the multiple data in this application.

[0360] As an embodiment, the first data set includes the multiple data in this application.

[0361] As an embodiment, the first data set includes training data (Training Data).

[0362] As an embodiment, the processor B belongs to the producer of the operation A.

[0363] As an embodiment, the processor B includes an AI training producer.

[0364] As an embodiment, the processor B includes an AI training function.

[0365] As an embodiment, the processor B is used for model training (Model Training), and the trained model is used to generate the first result.

[0366] As an embodiment, the processor B belongs to the first node.

[0367] The above embodiment avoids transmitting the plurality of data to the second node.

[0368] As an embodiment, the processor B belongs to the second node.

[0369] The above embodiments support joint training and optimize system performance.

[0370] As an embodiment, the processor B belongs to the core network.

[0371] The above embodiments support joint training of the entire network, further optimizing system performance.

[0372] As an embodiment, the plurality of data includes inference data.

[0373] As an embodiment, the processor C includes an AI inference producer.

[0374] As an embodiment, the processor C includes an AI inference function.

[0375] As an embodiment, the processor C belongs to the first node.

[0376] As an embodiment, the processor C constructs a model according to the multiple data, and then inputs the multiple data into the constructed model to obtain the first result.

[0377] As an embodiment, the processor C generates a restored data set according to the first type of output, and the error between the restored data set and the second data set is used to generate the first type of feedback.

[0378] As a sub-embodiment of the above embodiment, the generation of the restored data set adopts a method similar to the operation B.

[0379] As an embodiment, the first type of feedback is used to reflect the performance of the trained model; when the performance of the trained model cannot meet the requirements, the processor B will recalculate the target first type parameter group.

[0380] As an embodiment, when the error is too large or has not been updated for too long, the performance of the trained model is considered to be unable to meet the requirements.

[0381] As an embodiment, the target first category parameter group includes: one or more of: convolution kernel size, number of convolution layers, convolution step size, pooling kernel size, pooling kernel step size, pooling function, activation function, or number of feature maps.

[0382] As an embodiment, the target first-category parameter group includes: one or more of a convolution kernel, a pooling kernel, a pooling function, an activation function, parameters of a pooling function, or parameters of an activation function.

[0383] As an embodiment, the first model in the present application includes processor A.

[0384] As an embodiment, the first model in the present application includes processor B.

[0385] As an embodiment, the first model in the present application includes a processor C.

[0386] As an embodiment, the first model in the present application includes a processor D.

[0387] Example 12

[0388] Embodiment 12 illustrates a schematic diagram based on artificial intelligence or machine learning according to an embodiment of the present application; Fig.12 Attached Fig.12 Including operation A, operation B, operation C, operation D and operation E. In embodiment 12, the operation A and operation B belong to the first stage, the operation C belongs to the second stage, the operation D belongs to the third stage, and the operation E belongs to the fourth stage. Fig.12 In the diagram, the lines with arrows indicate the sequence of the processes.

[0389] As an embodiment, the operation A includes AI training, the operation B includes AI testing, the operation C includes AI emulation, the operation D includes AI entity loading, and the operation E includes AI inference.

[0390] As an embodiment, the first phase includes a training phase, the second phase includes an emulation phase, the third phase includes a deployment phase, and the fourth phase includes an emulation phase.

[0391] As an embodiment, the first stage includes AI model training.

[0392] As an embodiment, the first stage includes AI model training and AI testing.

[0393] As an embodiment, the AI ​​model training includes initial training and re-training of one or a group of AI entities.

[0394] As an embodiment, the AI ​​model training relies on training data.

[0395] As an embodiment, the AI ​​model training includes AI entity validation.

[0396] As an embodiment, the AI ​​entity verification is used to evaluate the performance of the AI ​​entity.

[0397] As an embodiment, the AI ​​entity verification relies on verification data.

[0398] As an embodiment, if the result of the AI ​​entity verification does not meet expectations, the AI ​​model will be retrained.

[0399] As an embodiment, the AI ​​testing includes testing the verified AI entity to estimate the performance of the trained AI model.

[0400] As an embodiment, if the result of the AI ​​test meets expectations, the AI ​​entity proceeds to the next stage; otherwise the AI ​​model will be retrained.

[0401] As an embodiment, the AI ​​test relies on test data.

[0402] As an embodiment, the second stage includes AI simulation, which performs inference of AI entities in a simulation environment.

[0403] As an embodiment, the AI ​​simulation is to estimate the performance of AI entity inference in a simulation environment before using the AI ​​entity.

[0404] As an embodiment, the second stage is optional.

[0405] As an embodiment, the third stage includes AI entity loading, and the AI ​​entity loading is to obtain the trained AI entity to obtain the desired AI inference function.

[0406] As an embodiment, the third stage is optional.

[0407] As an example, when the training function and the inference function are co-located, the third stage is no longer needed.

[0408] As an embodiment, the fourth stage includes AI inference.

[0409] As an embodiment, the operation E includes the operation A.

[0410] As an embodiment, the operation E includes the operation B.

[0411] Example 13

[0412] Embodiment 13 illustrates a structural block diagram of a processing device used in a first node according to an embodiment of the present application, as shown in the attached figure. Fig.13 As shown in the attached Fig.13 In the embodiment, the processing device 1300 in the first node includes a first receiver 1301 and a first transmitter 1302.

[0413] In Embodiment 13, the first receiver 1301 receives a first request, and in response to receiving the first request, generates a first result through a first model; and the first transmitter 1302 sends the first result.

[0414] In Example 13, the first model is for at least one of training or reasoning, and the first result is a training result or the first result is an reasoning result; the identifier of the first model is a first identifier, and the first identifier is a non-negative integer; the first identifier is associated with a first index, and the first index is one of a plurality of indexes included in a first index set; the plurality of indexes correspond to a plurality of training types, or the plurality of indexes correspond to a plurality of reasoning types, or the plurality of indexes correspond to a plurality of time windows; the first index set is associated with a first sub-identifier set, and any sub-identifier included in the first sub-identifier set indicates a class of data; the plurality of classes of data indicated by the plurality of sub-identifiers included in the first sub-identifier set include at least one of channel quality information data, codebook data, and location assistance information data; the sub-identifier associated with the first identifier depends on the first index.

[0415] As an embodiment, the first identifier is associated with a first sub-identifier group, any sub-identifier in the first sub-identifier group is a sub-identifier in the first sub-identifier set, and data indicated by any sub-identifier included in the first sub-identifier group is used to generate the first result.

[0416] As an embodiment, at least one sub-identifier included in the first sub-identifier set is associated with a cell set, and any cell in the cell set can use the data indicated by the at least one sub-identifier.

[0417] As an embodiment, at least one of the multiple sub-identifiers is associated with a second identifier, and the second identifier identifies a second model. The second model is for at least one of training or reasoning, and the second model is different from the first model.

[0418] As an embodiment, the training of the first model and the training of a type of data indicated by any sub-identifier included in the first sub-identifier group are independent.

[0419] As an embodiment, the channel quality information data includes CSI.

[0420] As an embodiment, the codebook data is a statistical result of the codebook adopted by the first node in a given time window.

[0421] As an embodiment, the location assistance information data includes at least one of positioning data or mobility data.

[0422] As an embodiment, the first node is user equipment.

[0423] As an embodiment, the first node is a relay node device.

[0424] As an embodiment, the first receiver 1301 includes 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} in Example 4.

[0425] As an embodiment, the first transmitter 1302 includes at least one of {the antenna 452, the transmitter 454, the transmit processor 468, the multi-antenna transmit processor 457, the controller / processor 459, the memory 460, and the data source 467} in Embodiment 4.

[0426] Embodiment 14

[0427] Embodiment 14 illustrates a structural block diagram of a processing device used in a second node according to an embodiment of the present application, as shown in the attached figure. Fig.14 As shown in the attached Fig.14 In the embodiment, the processing device 1400 in the second node includes a second transmitter 1401 and a second receiver 1402.

[0428] In Embodiment 14, the second transmitter 1401 sends a first request; and the second receiver 1402 receives a first result.

[0429] In Example 14, the recipient of the first request generates a first result through a first model as a response to receiving the first request; the first model is for at least one of training or inference, and the first result is a training result or the first result is an inference result; the identifier of the first model is a first identifier, and the first identifier is a non-negative integer; the first identifier is associated with a first index, and the first index is one of a plurality of indexes included in a first index set; the plurality of indexes correspond to a plurality of training types, or the plurality of indexes correspond to a plurality of inference types, or the plurality of indexes correspond to a plurality of time windows; the first index set is associated with a first sub-identifier set, and any sub-identifier included in the first sub-identifier set indicates a class of data; the plurality of classes of data indicated by the plurality of sub-identifiers included in the first sub-identifier set include at least one of channel quality information data, codebook data, and location assistance information data; the sub-identifier associated with the first identifier depends on the first index.

[0430] As an embodiment, the first identifier is associated with a first sub-identifier group, any sub-identifier in the first sub-identifier group is a sub-identifier in the first sub-identifier set, and data indicated by any sub-identifier included in the first sub-identifier group is used to generate the first result.

[0431] As an embodiment, at least one sub-identifier included in the first sub-identifier set is associated with a cell set, and any cell in the cell set can use the data indicated by the at least one sub-identifier.

[0432] As an embodiment, at least one of the multiple sub-identifiers is associated with a second identifier, and the second identifier identifies a second model. The second model is for at least one of training or reasoning, and the second model is different from the first model.

[0433] As an embodiment, the training of the first model and the training of a type of data indicated by any sub-identifier included in the first sub-identifier group are independent.

[0434] As an embodiment, the channel quality information data includes CSI.

[0435] As an embodiment, the codebook data is a statistical result of the codebook adopted by the first node in a given time window.

[0436] As an embodiment, the location assistance information data includes at least one of positioning data or mobility data.

[0437] As an embodiment, the second node is a base station device.

[0438] As an embodiment, the second node is user equipment.

[0439] As an embodiment, the second node is a TRP.

[0440] As an embodiment, the second transmitter 1401 includes at least one of {the antenna 420, the transmitter 418, the transmit processor 416, the multi-antenna transmit processor 471, the controller / processor 475, and the memory 476} in Embodiment 4.

[0441] As an embodiment, the second receiver 1402 includes at least one of {the antenna 420, the receiver 418, the receiving processor 470, the multi-antenna receiving processor 472, the controller / processor 475, and the memory 476} in Embodiment 4.

[0442] A person of ordinary skill in the art can understand that all or part of the steps in the above method can be completed by instructing the relevant hardware through a program, and the program can be stored in a computer-readable storage medium, such as a read-only memory, a hard disk or an optical disk. Optionally, all or part of the steps in the above embodiment can also be implemented using one or more integrated circuits. Accordingly, each module unit in the above embodiment can be implemented in the form of hardware or in the form of a software function module, and the present application is not limited to any specific form of software and hardware combination. The user equipment, terminal and UE in the present application include but are not limited to drones, communication modules on drones, remote-controlled aircraft, aircraft, small aircraft, mobile phones, tablet computers, notebooks, vehicle-mounted communication equipment, transportation tools, vehicles, RSUs, wireless sensors, Internet cards, Internet of Things terminals, RFID (Radio Frequency Identification, Radio Frequency Identification Technology) terminals, NB-IoT (Narrow Band Internet of Things, Narrow Band Internet of Things) terminals, MTC (Machine Type Communication, Machine Type Communication) terminals, eMTC (enhanced MTC, enhanced MTC) terminals, data cards, Internet cards, vehicle-mounted communication equipment, low-cost mobile phones, low-cost tablet computers 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, eNB (evolved Node B), gNB, TRP, GNSS (Global Navigation Satellite System), relay satellites, satellite base stations, aerial base stations, RSU, drones, test equipment, such as wireless communication equipment such as transceivers or signaling testers that simulate some functions of base stations.

[0443] It should be understood by those skilled in the art that the present invention may be implemented in other specified forms without departing from its core or essential features. Therefore, the embodiments disclosed herein should be considered illustrative rather than restrictive in any way. The scope of the invention is determined by the appended claims rather than the preceding description, and all modifications within their equivalent meanings and regions are considered to be included therein.

Claims

1. A first node for wireless communication, characterized in that: include: A first receiver receives a first request; In response to receiving the first request, generating a first result by a first model; Among them, the first model is for at least one of training or reasoning, and the first result is a training result or the first result is an reasoning result; the identifier of the first model is a first identifier, and the first identifier is a non-negative integer; the first identifier is associated with a first index, and the first index is one of a plurality of indexes included in a first index set; the plurality of indexes correspond to a plurality of training types, or the plurality of indexes correspond to a plurality of reasoning types, or the plurality of indexes correspond to a plurality of time windows; the first index set is associated with a first sub-identifier set, and any sub-identifier included in the first sub-identifier set indicates a class of data; the plurality of classes of data indicated by the plurality of sub-identifiers included in the first sub-identifier set include at least one of channel quality information data, codebook data, and location assistance information data; the sub-identifier associated with the first identifier depends on the first index.

2. The first node according to claim 1, characterized in that: The first identifier is associated with a first sub-identifier group, any sub-identifier in the first sub-identifier group is a sub-identifier in the first sub-identifier set, and data indicated by any sub-identifier included in the first sub-identifier group is used to generate the first result.

3. The first node according to claim 1 or 2, characterized in that: At least one sub-identifier included in the first sub-identifier set is associated with a cell set, and any cell in the cell set can use the data indicated by the at least one sub-identifier.

4. The first node according to claim 2 or 3, characterized in that: The training of the first model and the training of a type of data indicated by any sub-identifier included in the first sub-identifier group are independent.

5. The first node according to any one of claims 1 to 4, characterized in that: The channel quality information data includes CSI.

6. The first node according to any one of claims 1 to 5, characterized in that: The codebook data is a statistical result of the codebook used by the first node in a given time window.

7. The first node according to any one of claims 1 to 6, characterized in that: The location assistance information data includes at least one of positioning data or mobility data.

8. The first node according to any one of claims 1 to 7, characterized in that include: The first transmitter sends the first result.

9. A second node for wireless communication, characterized in that: include: A second transmitter sends a first request; wherein the receiver of the first request generates a first result through the first model as a response to receiving the first request; The first model is for at least one of training or inference, and the first result is a training result or the first result is an inference result; the identifier of the first model is a first identifier, and the first identifier is a non-negative integer; the first identifier is associated with a first index, and the first index is one of a plurality of indexes included in a first index set; the plurality of indexes correspond to a plurality of training types, or the plurality of indexes correspond to a plurality of inference types, or the plurality of indexes correspond to a plurality of time windows; the first index set is associated with a first sub-identifier set, and any sub-identifier included in the first sub-identifier set indicates a class of data; the plurality of classes of data indicated by the plurality of sub-identifiers included in the first sub-identifier set include at least one of channel quality information data, codebook data, and location assistance information data; the sub-identifier associated with the first identifier depends on the first index.

10. A method in a first node in wireless communication, characterized in that: include: receiving a first request; In response to receiving the first request, generating a first result by a first model; Among them, the first model is for at least one of training or reasoning, and the first result is a training result or the first result is an reasoning result; the identifier of the first model is a first identifier, and the first identifier is a non-negative integer; the first identifier is associated with a first index, and the first index is one of a plurality of indexes included in a first index set; the plurality of indexes correspond to a plurality of training types, or the plurality of indexes correspond to a plurality of reasoning types, or the plurality of indexes correspond to a plurality of time windows; the first index set is associated with a first sub-identifier set, and any sub-identifier included in the first sub-identifier set indicates a class of data; the plurality of classes of data indicated by the plurality of sub-identifiers included in the first sub-identifier set include at least one of channel quality information data, codebook data, and location assistance information data; the sub-identifier associated with the first identifier depends on the first index.

11. A second node for wireless communication, characterized in that: include: Send the first request; wherein the receiver of the first request generates a first result through the first model as a response to receiving the first request; The first model is for at least one of training or inference, and the first result is a training result or the first result is an inference result; the identifier of the first model is a first identifier, and the first identifier is a non-negative integer; the first identifier is associated with a first index, and the first index is one of a plurality of indexes included in a first index set; the plurality of indexes correspond to a plurality of training types, or the plurality of indexes correspond to a plurality of inference types, or the plurality of indexes correspond to a plurality of time windows; the first index set is associated with a first sub-identifier set, and any sub-identifier included in the first sub-identifier set indicates a class of data; the plurality of classes of data indicated by the plurality of sub-identifiers included in the first sub-identifier set include at least one of channel quality information data, codebook data, and location assistance information data; the sub-identifier associated with the first identifier depends on the first index.

Images