Method and apparatus in communication node used for wireless communication

Abstract

The invention discloses a method and an apparatus in a communication node used for wireless communication. A communication node receives on a plurality of RS resource groups, any one of the plurality of RS resource groups being associated with a threshold; as a response that the performance of the first model is lower than or not higher than a first threshold, executing a first operation; the input of the first model comprises measurements for a first group of RS resources, the first threshold being the one threshold associated with the first group of RS resources; the first RS resource group is one RS resource group in the plurality of RS resource groups; the thresholds associated with at least two RS resource groups in the plurality of RS resource groups are different; the first operation comprises sending a first message; or the first operation comprises stopping applying at least the first RS resource group to the first model. The method provided by the invention is beneficial to realizing more efficient model performance monitoring.

Description

Technical Field

[0001] This application relates to transmission methods and apparatus in wireless communication systems, and more particularly to methods and apparatus for improving the performance of models. Background Technology

[0002] In NR (New Radio) Release 18, research on AI (Artificial Intelligence) / ML (Machine Learning) technologies was initiated to explore their impact on system performance and design. AI / ML technologies may also play a crucial role in future 6G communications. Compared to traditional processing methods, AI / ML is characterized by its training-based and deployment-required nature. According to the 3GPP (3rd Generation Partnership Project) standard TS38.300, AI / ML models and algorithms extend beyond the scope of 3GPP.

[0003] By performing performance monitoring of the model and reporting it to the network, the UE can assist the network in deciding on operations such as activation, deactivation, updating, and switching of AI / ML models / functionalities. Summary of the Invention

[0004] The applicant's research revealed that effectively monitoring model performance is a problem that needs to be solved. To address this problem, this application provides a solution. It should be noted that while many embodiments of this application are geared towards AI / ML, this application is also applicable to other scenarios, such as large models, LLM (Large Language Model), AIGC (Artificial Intelligence Generative Content), AGI (Artificial General Intelligence), GPT (Generative Pre-trained Transformer), ChatGPT, or NLP (Natural Language Processing). Although the specification of this application involves descriptions of some AI / ML models and algorithms, those skilled in the art will understand that these descriptions are not essential or irreplaceable for solutions related to wireless cellular communication. Furthermore, adopting a unified solution across different scenarios helps reduce hardware complexity and cost. Where there is no conflict, the embodiments and features in the first node of this application can be applied to the second node, and vice versa. Where there is no conflict, the embodiments and features in the embodiments of this application can be combined with each other arbitrarily.

[0005] As an example, the interpretation of terms in this application is based on the definitions in the 3GPP specification protocol TS38 series.

[0006] As an example, the interpretation of the terms in this application is based on the definitions in the 3GPP specification protocol TS28 series.

[0007] This application discloses a method used in a first node of wireless communication, characterized by comprising:

[0008] Received on multiple RS (Reference Signal) resource groups, each of which is associated with a threshold.

[0009] If the performance of the first model is lower than or equal to the first threshold response, the first operation is performed.

[0010] Wherein, the input of the first model includes a measurement for a first RS resource group, the first threshold is a threshold associated with the first RS resource group; the first RS resource group is one of the plurality of RS resource groups; the threshold associated with at least two of the plurality of RS resource groups is different;

[0011] Wherein, the first operation includes sending a first message; and / or, the first operation includes stopping the application of at least the first RS resource group to the first model.

[0012] For a given model, the network can configure a threshold for the UE to perform performance monitoring for that model. The method described above takes channel quality into account and uses a threshold associated with an RS resource group as the model's performance indicator. Furthermore, considering that a model might not fully realize its potential if used only for a single scenario or function, this method associates any one of multiple RS resource groups with a threshold, enabling flexible configuration of the performance monitoring threshold and thus achieving more efficient model performance monitoring.

[0013] As one embodiment, the first operation includes sending a first message.

[0014] The above method helps the network make decisions based on the first message.

[0015] As one embodiment, the first operation includes stopping the application of at least the first RS resource group to the first model.

[0016] The above methods help reduce signaling overhead.

[0017] The above methods help UEs make timely decisions about the model.

[0018] As one embodiment, the first operation includes sending a first message, and the first operation includes stopping the application of at least the first RS resource group to the first model.

[0019] The above method is helpful in informing network UEs about their decisions regarding the model.

[0020] According to one aspect of this application, it is characterized by comprising:

[0021] Receive the first signaling;

[0022] Wherein, the first signaling configures the associated threshold for each of the plurality of RS resource groups.

[0023] According to one aspect of this application, the first message indicates the first RS resource group.

[0024] The above method is beneficial for the network to adjust the first RS resource group.

[0025] The above method is beneficial for the network to make decisions on the model based on the first RS resource group.

[0026] According to one aspect of this application, the performance of the first model depends on the error between the output of the first model and the actual measurement result, which is obtained after the measurement for the first RS resource group.

[0027] The above method further defines how the performance of the first model is achieved. The performance of the first model is determined by the error between its output and the actual measurement results, resulting in a simple and efficient implementation.

[0028] According to one aspect of this application, the plurality of RS resource groups are associated with a plurality of cells respectively; wherein the plurality of cells include at least one serving cell.

[0029] The above method takes into account the application of the first model to multiple cells and the differences between different cells, further defining that multiple RS resource groups are associated with multiple cells respectively, and using independent performance monitoring thresholds on different cells, thereby improving the performance of the first model and / or reducing signaling overhead and / or system performance.

[0030] According to one aspect of this application, the plurality of RS resource groups are respectively associated with a plurality of functions; wherein any one of the plurality of functions adopts the first model.

[0031] The above method takes into account the application of the first model to multiple functions and the differences between different functions. It further defines multiple RS resource groups as associated with multiple functions respectively and uses independent performance monitoring thresholds on different functions, thereby improving the performance of the first model and / or reducing signaling overhead and / or system performance.

[0032] According to one aspect of this application, the first operation includes receiving a second signaling; wherein the second signaling triggers the first message.

[0033] The above method takes into account that unnecessary signaling will bring additional overhead and be detrimental to system performance. Triggering the first message through the second signaling is beneficial to network control, thereby avoiding the unnecessary sending of the first message and reducing signaling overhead.

[0034] According to one aspect of this application, a first event triggers the first message; wherein the first event depends on a second threshold, the second threshold being configurable.

[0035] The above method takes into account the possibility that the performance of the first model may deteriorate if the network cannot schedule the first message in a timely manner, or that the first message cannot be sent due to information loss caused by storage or other reasons. By triggering the first message through the first event, the network can obtain the first message in a timely manner, thereby optimizing the performance of the first model.

[0036] This application discloses a method used in a second node for wireless communication, characterized by comprising:

[0037] Receive the first message;

[0038] Wherein, the sender of the first message receives the message on multiple RS resource groups, each of which is associated with a threshold; as a response to the performance of the first model being lower or not higher than the first threshold, the sender of the first message performs a first operation; the input to the first model includes a measurement for a first RS resource group, the first threshold being the threshold associated with the first RS resource group; the first RS resource group is one of the multiple RS resource groups; at least two of the multiple RS resource groups are associated with different thresholds;

[0039] Wherein, the first operation includes sending the first message; and / or, the first operation includes sending the first message and stopping the application of at least the first RS resource group to the first model.

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

[0041] Transmitted on at least one of the plurality of RS resource groups.

[0042] According to one aspect of this application, it is characterized by comprising:

[0043] Send the first signaling;

[0044] Wherein, the first signaling configures the associated threshold for each of the plurality of RS resource groups.

[0045] According to one aspect of this application, the first message indicates the first RS resource group.

[0046] According to one aspect of this application, the performance of the first model depends on the error between the output of the first model and the actual measurement result, which is obtained after the measurement for the first RS resource group.

[0047] According to one aspect of this application, the plurality of RS resource groups are associated with a plurality of cells respectively; wherein the plurality of cells include at least one serving cell.

[0048] According to one aspect of this application, the plurality of RS resource groups are respectively associated with a plurality of functions; wherein any one of the plurality of functions adopts the first model.

[0049] According to one aspect of this application, the first operation includes receiving a second signaling; wherein the second signaling triggers the first message.

[0050] According to one aspect of this application, a first event triggers the first message; wherein the first event depends on a second threshold, the second threshold being configurable.

[0051] This application discloses a first node used for wireless communication, characterized in that it comprises:

[0052] A first receiver receives data on a plurality of RS resource groups, each of which is associated with a threshold.

[0053] The first processor, whose performance is lower than or equal to a first threshold response, executes the first operation.

[0054] Wherein, the input of the first model includes a measurement for a first RS resource group, the first threshold is a threshold associated with the first RS resource group; the first RS resource group is one of the plurality of RS resource groups; the threshold associated with at least two of the plurality of RS resource groups is different;

[0055] Wherein, the first operation includes sending a first message; and / or, the first operation includes stopping the application of at least the first RS resource group to the first model.

[0056] This application discloses a second node used for wireless communication, characterized in that it comprises:

[0057] The second receiver receives the first message;

[0058] Wherein, the sender of the first message receives the message on multiple RS resource groups, each of which is associated with a threshold; as a response to the performance of the first model being lower or not higher than the first threshold, the sender of the first message performs a first operation; the input to the first model includes a measurement for a first RS resource group, the first threshold being the threshold associated with the first RS resource group; the first RS resource group is one of the multiple RS resource groups; at least two of the multiple RS resource groups are associated with different thresholds;

[0059] Wherein, the first operation includes sending the first message; and / or, the first operation includes sending the first message and stopping the application of at least the first RS resource group to the first model. Attached Figure Description

[0060] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings:

[0061] Figure 1 A flowchart of a first operation according to an embodiment of this application is shown;

[0062] Figure 2 A schematic diagram of a network architecture according to an embodiment of this application is shown;

[0063] Figure 3 A schematic diagram of an embodiment of a wireless protocol architecture for the user plane and control plane according to an embodiment of this application is shown;

[0064] Figure 4 A schematic diagram of a first communication device and a second communication device according to an embodiment of this application is shown;

[0065] Figure 5 A flowchart illustrating a wireless signal transmission process according to an embodiment of this application is shown;

[0066] Figure 6 A flowchart illustrating a wireless signal transmission process according to another embodiment of this application is shown;

[0067] Figure 7 A flowchart illustrating a wireless signal transmission process according to yet another embodiment of this application is shown;

[0068] Figure 8 A schematic diagram showing a first message instructing a first RS resource group according to an embodiment of this application is shown;

[0069] Figure 9 A schematic diagram illustrating the performance of a first model according to an embodiment of this application is shown;

[0070] Figure 10 A schematic diagram illustrating multiple RS resource groups associated with multiple cells according to an embodiment of this application is shown;

[0071] Figure 11 A schematic diagram illustrating multiple RS resource groups associated with multiple functions according to an embodiment of this application is shown;

[0072] Figure 12 A structural block diagram of a processing apparatus for a first node according to an embodiment of this application is shown;

[0073] Figure 13 A structural block diagram of a processing apparatus for a second node according to an embodiment of this application is shown;

[0074] Figure 14 A schematic diagram of an AI / ML model according to an embodiment of this application is shown;

[0075] Figure 15 A schematic diagram illustrating the deployment of intelligent functions in a RAN domain according to an embodiment of this application is shown;

[0076] Figure 16 A schematic diagram illustrating the deployment of UE smart functions according to an embodiment of this application is shown;

[0077] Figure 17 A flowchart based on artificial intelligence or machine learning is shown according to an embodiment of this application. Detailed Implementation

[0078] The technical solution of this application will be further described in detail below with reference to the accompanying drawings. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of this application can be arbitrarily combined with each other.

[0079] Example 1

[0080] Example 1 illustrates a flowchart of a first operation according to an embodiment of this application, as shown in the attached diagram. Figure 1 As shown. (Attached) Figure 1 In the diagram, each box represents a step. It is particularly important to emphasize that the order of the boxes does not represent the chronological order of the steps they represent.

[0081] In Embodiment 1, the first node of this application receives data on a plurality of RS resource groups in step 101, wherein any RS resource group among the plurality of RS resource groups is associated with a threshold; in step 102, a first operation is performed as a response that the performance of a first model is lower or not higher than a first threshold; wherein the input of the first model includes a measurement for a first RS resource group, and the first threshold is the threshold associated with the first RS resource group; the first RS resource group is one of the plurality of RS resource groups; and the threshold associated with at least two of the plurality of RS resource groups is different.

[0082] As one embodiment, the first operation includes sending a first message.

[0083] As one embodiment, the first operation includes stopping the application of at least the first RS resource group to the first model.

[0084] As one embodiment, the first operation includes sending a first message, and the first operation includes stopping the application of at least the first RS resource group to the first model.

[0085] As a sub-implementation of the above embodiment, the sending of the first message occurs before the termination of the application of at least the first RS resource group to the first model.

[0086] As a sub-implementation of the above embodiments, the sending of the first message occurs after the termination of the application of at least the first RS resource group to the first model.

[0087] As an example, receiving on multiple RS resource groups means receiving the corresponding RS on each RS resource in each of the multiple RS resource groups.

[0088] As an example, receiving on multiple RS resource groups means receiving the corresponding RS on at least one RS resource in each of the multiple RS resource groups.

[0089] As an example, the plurality of RS resource groups belong to the same serving cell of the first node.

[0090] As an example, the same serving cell is PCell (Primary Cell).

[0091] As an example, the same serving cell is a PSCell (Primary SCG Cell).

[0092] As an example, the same serving cell is a SCell (Secondary Cell).

[0093] As one example, the plurality of RS resource groups belong to at least one serving cell of the same cell group of the first node.

[0094] As an example, the same cell group is an MCG (Master Cell Group).

[0095] As an example, the same cell group is an SCG (Secondary Cell Group).

[0096] As one embodiment, the cell associated with the first RS resource group is a serving cell of the first node; the cell associated with any RS resource group other than the first RS resource group among the plurality of RS resource groups is an additional cell.

[0097] As a sub-example of the above embodiment, the serving cell is PCell, and the additional cell is configured as AdditionalPCIIndex.

[0098] As a sub-example of the above embodiment, the serving cell is PSCell, and the additional cell is indicated by SSB-MTC-AdditionalPCI.

[0099] As an example, any one of the plurality of RS resource groups includes at least one RS resource for channel measurement.

[0100] As an example, any of the plurality of RS resource groups further includes at least one RS resource for interference measurement, such as a zero-power CSI (Channel State Information) RS resource.

[0101] As an example, the RS resources used for channel measurement are different in any two of the plurality of RS resource groups.

[0102] As one example, the channel measurement includes: channel estimation.

[0103] As one embodiment, the channel measurement includes: determining the channel quality through measurement.

[0104] As an example, the channel measurement includes RSRP (Reference Signal Received Power) measurement.

[0105] As an example, the channel measurement includes: RSRQ (Reference Signal Received Quality) measurement.

[0106] As an example, the channel measurement includes: SINR (Signal to Interference plus Noise Ratio) measurement.

[0107] As an example, the channel measurement includes: CQI (Channel Quality Indicator) measurement.

[0108] As an example, the channel measurement includes: PMI (Precoding Matrix Indicator) measurement.

[0109] As an example, the channel measurement includes: RI (Rank Indicator) measurement.

[0110] As an example, the channel measurement includes: CRI (CSI-RS Resource Indicator) measurement.

[0111] As an example, the channel measurement is a cell-level measurement result.

[0112] As an example, the channel measurement is for beam-level measurement results.

[0113] As an example, any one of the plurality of RS resource groups includes at least one RS resource for channel prediction.

[0114] As an example, any of the plurality of RS resource groups further includes at least one RS resource for interference prediction, such as a zero-power CSI-RS resource.

[0115] As an example, the RS resources used for channel prediction are different in any two of the plurality of RS resource groups.

[0116] As one embodiment, the channel prediction includes: determining the channel quality through prediction.

[0117] As an example, the channel prediction includes RSRP prediction.

[0118] As an example, the channel prediction includes: RSRQ prediction.

[0119] As an example, the channel prediction includes: SINR prediction.

[0120] As an example, the channel prediction includes: CQI prediction.

[0121] As an example, the channel prediction includes: PMI prediction.

[0122] As an example, the channel prediction includes: CRI prediction.

[0123] As an example, the channel prediction is a cell-level prediction result.

[0124] As an example, the channel prediction is for beam-level prediction results.

[0125] As an example, any one of the plurality of RS resource groups consists of RS resources of the same type.

[0126] As an example, the type of RS resources included in any of the plurality of RS resource groups is configurable.

[0127] As an example, any two RS resource groups among the plurality of RS resource groups consist of RS resources of the same type.

[0128] As an example, the types of RS resources included in any two of the plurality of RS resource groups are configurable.

[0129] As an example, candidates for the type of RS resource include SSB (Synchronization Signal Block) resources.

[0130] As an example, candidates for the type of RS resource include CSI-RS resources.

[0131] As an example, the candidate types of the RS resource include SSB resources and CSI-RS resources.

[0132] As an example, candidates for the type of RS resource include synchronization signal resources.

[0133] As an example, candidates for the type of RS resource include broadcast channel resources.

[0134] As an example, candidates for the type of RS resource include sounding RS resources.

[0135] As an example, candidates for the type of RS resource include Positioning RS resources.

[0136] As an example, candidates for the type of RS resource include training RS resources.

[0137] As an example, the training RS resources are for the first model.

[0138] As an example, the training RS resources are used to utilize the functionality of the first model.

[0139] As an example, candidates for the type of RS resource include aware RS resources.

[0140] As an example, the sensing signal uses a single-frequency wave.

[0141] As an example, the sensing signal uses a frequency-modulated wave.

[0142] As an example, the sensing signal employs OFDM.

[0143] As an example, the sensing signal is an ISAC (Integrated Sensing and Communication) signal.

[0144] As an example, the sensing signal is a radar signal.

[0145] As an example, the plurality of RS resource groups are indicated by at least one signaling signal.

[0146] As an example, the plurality of RS resource groups are configured by at least one signaling.

[0147] As an example, the plurality of RS resource groups are configured and activated by at least one signaling.

[0148] As an example, the at least one signaling indicates the index of each of the plurality of RS resource groups.

[0149] As an example, the at least one signaling indicates the index of each RS resource group in the plurality of RS resource groups and the index of each RS resource in each RS resource group.

[0150] As an example, the at least one signaling is higher-level signaling.

[0151] As one embodiment, the at least one signaling includes higher-layer signaling and lower-layer signaling.

[0152] As an example, the higher-layer signaling is the signaling of the RRC (Radio Resource Control) sublayer.

[0153] As an example, the higher-level signaling is an RRC message.

[0154] As an example, the higher-layer signaling is NAS (Non-access stratum) layer signaling.

[0155] As an example, the higher-level signaling is a NAS message.

[0156] As an example, the lower-layer signaling is the signaling of the MAC (Medium Access Control) sublayer.

[0157] As an example, the lower-layer signaling is physical layer signaling.

[0158] As an example, the lower-layer signaling is DCI (Downlink Control Information).

[0159] As an example, the first node determines the plurality of RS resource groups on its own.

[0160] As an example, the first node independently determines the index of the RS resource in each of the plurality of RS resource groups, and the at least one signaling indicates the index of each of the plurality of RS resource groups.

[0161] As an example, the first node independently determines the index of each RS resource group in the plurality of RS resource groups and the index of each RS resource in each RS resource group.

[0162] As an example, the first node independently determines the plurality of RS resource groups and each RS resource in each RS resource group from the plurality of RS resources.

[0163] As an example, the first node determines the plurality of RS resource groups from a plurality of RS resource groups.

[0164] As an example, the "self-determination" refers to self-selection.

[0165] As an example, the self-determination is based on measurement.

[0166] As an example, the self-determination is based on prediction.

[0167] As an example, the meaning of "self-determined" includes not being indicated by air interface signaling.

[0168] As an example, "self-determined" means that it is not determined by the network.

[0169] As an example, the plurality of RS resource groups are configured by at least one signaling and selected by the first node.

[0170] As an example, some of the RS resource groups among the plurality of RS resource groups are configured, and some of the RS resource groups among the plurality of RS resource groups are selected by the first node.

[0171] As an example, the first model is configured with the plurality of RS resource groups.

[0172] As an example, any one of the plurality of RS resource groups is configured with the first model.

[0173] As one embodiment, the first processor trains (including fine-tuning, reinforcement learning, etc.) the first model based on measurements on some or all of the multiple RS resource groups.

[0174] As an example, the first processor inputs measurements from some or all of the multiple RS resource groups into the first model and performs inference.

[0175] As an example, the multiple RS resource groups correspond to multiple BFR (Beam Failure Recovery) processes.

[0176] As an example, the multiple RS resource groups correspond to multiple RLM (Radio Link Monitoring) processes.

[0177] As an example, the multiple RS resource groups correspond to multiple TAGs (Timing Advance Groups).

[0178] As one example, the plurality of RS resource groups correspond to a plurality of candidate cells.

[0179] As one example, the plurality of RS resource groups correspond to a plurality of candidate configurations.

[0180] As an example, the plurality of RS resource groups correspond to a plurality of measurement configurations; each of the plurality of measurement configurations corresponds to a MeasId.

[0181] As an example, the plurality of RS resource groups correspond to a plurality of measurement objects; each of the plurality of measurement objects is configured with a MeasObjectId.

[0182] As an example, the plurality of RS resource groups correspond to a plurality of report configurations; each of the plurality of report configurations corresponds to a ReportConfigId.

[0183] As an example, the threshold associated with any one of the plurality of RS resource groups is configurable.

[0184] As an example, the threshold associated with any of the plurality of RS resource groups is indicated by higher-layer signaling.

[0185] As an example, the threshold associated with any of the plurality of RS resource groups is indicated by low-level signaling.

[0186] As an example, the threshold associated with any one of the plurality of RS resource groups is indicated by the output of the first model.

[0187] As an example, the threshold associated with at least one of the plurality of RS resource groups is determined by the first node itself.

[0188] As an example, the threshold associated with at least one of the plurality of RS resource groups is selected by the first node from a plurality of thresholds.

[0189] As a sub-implementation, the multiple thresholds are indicated by higher-layer signaling.

[0190] As a sub-implementation, the multiple thresholds are indicated by low-level signaling.

[0191] As a sub-implementation, the plurality of thresholds are predefined.

[0192] As a sub-example, the first node selects one threshold from the plurality of thresholds based on the number of RS resources in an RS resource group.

[0193] As an example, the more RS resources in an RS resource group, the higher the threshold associated with the RS resource group; the fewer RS ​​resources in an RS resource group, the lower the threshold associated with the RS resource group.

[0194] As an example, the fewer the number of RS resources in an RS resource group, the higher the threshold associated with the RS resource group; the more the number of RS resources in an RS resource group, the lower the threshold associated with the RS resource group.

[0195] As a sub-example, the first node selects one threshold from the plurality of thresholds based on the current performance of the first model.

[0196] As an example, the performance of the first model is measured by inference accuracy.

[0197] As an example, the performance metric of the first model is inference accuracy.

[0198] As an example, the performance of the first model is measured by an error function that includes the error between the output of the first model and the actual measurement result.

[0199] As an example, the performance metric of the first model is an error function.

[0200] As an example, the performance metric for the first model is an error.

[0201] As an example, the performance metric of the first model is a probability.

[0202] As an example, the performance of the first model is measured by system performance.

[0203] As an example, the performance metric of the first model is a system performance.

[0204] As an example, "lower than" or "not higher than" means "lower than".

[0205] As an example, "less than" means "greater than".

[0206] As an example, "less than" means not less than.

[0207] As an example, "lower than" or "not higher than" means "not higher than".

[0208] As an example, "not higher than" means "not greater than".

[0209] As an example, "not higher than" means "not less than".

[0210] As an example, the first model was used for CSI compression.

[0211] As an example, the first model is used for beam management.

[0212] As an example, the first model was used for positioning.

[0213] As an example, the first model is used for network energy saving.

[0214] As an example, the first model is used for mobility management.

[0215] As an example, the first model was used for RLF (Radio Link Failure) prediction.

[0216] As an example, the first model was used for HOF (Handover Failure) prediction.

[0217] As an example, the first model is used for switching predictions.

[0218] As an example, the first model is used to trigger event prediction.

[0219] As an example, "responding to the first model's performance being lower or not higher than a first threshold, performing the first operation" means: performing the first operation when at least the performance of the first model is lower or not higher than the first threshold.

[0220] As an example, "responding when the performance of the first model is lower or not higher than the first threshold, performing the first operation" means: performing the first operation when the performance of the first model is lower or not higher than the first threshold.

[0221] As an example, "responding when the performance of the first model is lower or not higher than the first threshold, performing the first operation" means: performing the first operation after the performance of the first model is lower or not higher than the first threshold.

[0222] As an example, "responding to the performance of the first model being lower or not higher than the first threshold, and performing the first operation" means that at least the performance of the first model being lower or not higher than the first threshold triggers the execution of the first operation.

[0223] As an example, "responding to the performance of the first model being lower or not higher than the first threshold, and performing the first operation" means that the performance of the first model being lower or not higher than the first threshold triggers the execution of the first operation.

[0224] As an example, "responding to the first model's performance being lower or not higher than a first threshold, and performing a first operation" means that the first model's performance being lower or not higher than the first threshold is used to confirm the execution of the first operation.

[0225] As an example, the measurement for the first RS resource group refers to the measurement result for each RS resource in the first RS resource group.

[0226] As an example, the measurement for the first RS resource group refers to the measurement result for at least one RS resource in the first RS resource group.

[0227] As an example, the measurement result is an L1 measurement result.

[0228] As an example, the measurement result is an L3 measurement result.

[0229] As an example, the measurement results are filtered.

[0230] As an example, the measurement results are unfiltered.

[0231] As an example, the measurement results are beam-level.

[0232] As an example, the measurement results are at the cell level.

[0233] As an example, the input to the first model also includes measurements for at least one RS resource group other than the first RS resource group among the plurality of RS resource groups.

[0234] As an example, the input to the first model does not include measurements for any RS resource group other than the first RS resource group among the plurality of RS resource groups.

[0235] As an example, the input of the first model refers to the input of the first model within a time interval.

[0236] As an example, the first RS resource group is any RS resource group among the plurality of RS resource groups that was used as input to the first model within a time interval.

[0237] As an example, the time interval is a finite period of time prior to the first operation.

[0238] As an example, during the time interval, the input of the first model includes measurements for the first RS resource group.

[0239] As an example, any one of the plurality of RS resource groups can be used as input to the first model.

[0240] As an example, any one of the plurality of RS resource groups is associated with the first model.

[0241] As an example, any one of the plurality of RS resource groups is configured with the identifier of the first model.

[0242] As an example, the first RS resource group is any one of the plurality of RS resource groups.

[0243] As an example, the threshold associated with any two of the plurality of RS resource groups is not the same.

[0244] As a sub-example of the above embodiment, each of the plurality of RS resource groups exclusively enjoys one of the threshold values.

[0245] As a sub-example of the above embodiment, any two RS resource groups among the plurality of RS resource groups are configured with the threshold by different domains.

[0246] As an example, at least two RS resource groups among the plurality of RS resource groups are associated with different thresholds, and there are two RS resource groups among the plurality of RS resource groups that are associated with the same threshold.

[0247] As a sub-example of the above embodiment, there are two RS resource groups among the plurality of RS resource groups that share the same threshold.

[0248] As a sub-example of the above embodiment, there are two RS resource groups among the plurality of RS resource groups that are configured with the same threshold by the same domain.

[0249] As an example, the first message is transmitted over the air interface.

[0250] As an example, the recipient of the first message is the second node in this application.

[0251] As an example, the recipient of the first message is the target node described in this application.

[0252] As an example, the recipient of the first message is a higher-level node among the target nodes in this application.

[0253] As an example, the recipient of the first message is a base station device in the target node described in this application.

[0254] In one embodiment, the recipient of the first message is the first node.

[0255] As one example, the recipient of the first message is a higher layer of the first node.

[0256] As an example, the first message includes at least one signaling from a higher layer.

[0257] As an example, the first message is at least one signaling from a higher layer.

[0258] As one example, the higher layer is the NAS (Non-Access Stratum) layer.

[0259] As an example, the first message includes a NAS message.

[0260] As an example, the first message includes an RRC container.

[0261] As one example, the higher layer is an RRC sublayer.

[0262] As an example, the first message is transmitted via an RB (Radio Bearer).

[0263] As an example, one of the RBs is an SRB (Signaling Radio Bearer).

[0264] As an example, one of the RBs is SRB2.

[0265] As an example, one of the RBs is SRB1 or SRB3.

[0266] As an example, the RB is neither a DRB nor an SRB.

[0267] As an example, the name of the RB includes RB and the name of the first radio bearer includes I, AI, ML, or LLM.

[0268] As an example, one of the RBs is a data and / or signaling that can be used to transmit training and / or inference.

[0269] As an example, one RB is dedicated to data and / or signaling for training and / or inference.

[0270] As a sub-implementation of the above embodiments, the above method uses a dedicated wireless bearer for AI / ML to avoid affecting communication data.

[0271] As a sub-implementation of the above embodiments, the RB is AI / ML dedicated.

[0272] As a sub-example of the above embodiments, the RB is dedicated to the AI / ML model.

[0273] As an example, the first message includes an RRC message.

[0274] As an example, the first message includes at least one RRC IE.

[0275] As an example, the first message includes at least one RRC field.

[0276] As one example, the first message includes a UE assistance information message.

[0277] As an example, the first message includes a UEAssistanceInformation message.

[0278] As an example, the first message includes a measurement report message.

[0279] As an example, the first message includes a MeasurementReport message.

[0280] As an example, the first message includes a failure information message.

[0281] As an example, the first message includes a ModelFailureInformation message.

[0282] As an example, the first message includes a UEInformationResponse message.

[0283] As one embodiment, the first message includes at least one signaling layer at a lower level.

[0284] As an example, the first message is at least one signaling layer at a lower level.

[0285] As an example, the lower layer is the MAC sublayer.

[0286] As a sub-example, the first message includes at least one MAC CE (Control Element).

[0287] As a sub-example, the first message includes at least one MAC CE and at least one MAC subheader.

[0288] As one example, the lower layer is the physical layer.

[0289] As a sub-implementation, the first message includes at least one UCI (Uplink Control Information).

[0290] As a sub-example, the first message includes at least one PUCCH (Physical Uplink Control Channel) transmission.

[0291] As a sub-example, the first message includes at least one PUSCH (Physical Uplink SharedChannel) transmission.

[0292] As an example, the first message includes at least one signaling from a higher layer and at least one signaling from a lower layer.

[0293] As an example, the higher layer is the NAS layer, and the lower layer is the MAC sublayer.

[0294] As one example, the higher layer is the NAS layer, and the lower layer is the physical layer.

[0295] As an example, the higher layer is the RRC sublayer, and the lower layer is the MAC sublayer.

[0296] As an example, the higher layer is the RRC sublayer, and the lower layer is the physical layer.

[0297] As an example, the higher layer is the protocol layer above the RRC sublayer, and the lower layer is the RRC sublayer.

[0298] Typically, since the first message is triggered by the performance of the first model being lower than or not higher than a first threshold, the first message at least indicates that the performance of the first model is lower than or not higher than the first threshold.

[0299] As an example, the first message includes a field indicating that the performance of the first model is lower than or not higher than the first threshold.

[0300] As an example, the field in the first message is set to indicate that the performance of the first model is lower than or not higher than the first threshold.

[0301] As an example, the value of the field in the first message is a specified candidate value among a plurality of candidate values, indicating that the performance of the first model is lower than or not higher than the first threshold.

[0302] As an example, any one of the plurality of candidate values ​​is a numerical value.

[0303] As an example, any one of the plurality of candidate values ​​is a string.

[0304] As an example, the plurality of candidate values ​​includes 1 and 0, and the specified candidate value is 1.

[0305] As an example, the plurality of candidate values ​​includes true and false, and the specified candidate value is true.

[0306] As one embodiment, the first message includes a first identifier, which indicates the first model.

[0307] As an example, the first identifier explicitly indicates the first model.

[0308] As an example, the first identifier implicitly indicates the first model.

[0309] As an example, the first identifier is associated with the first model.

[0310] As an example, the first identifier is configured to the first model.

[0311] As an example, the first identifier is an associated identifier (associated ID / associationID).

[0312] As an example, the first identifier is a configuration identifier.

[0313] As an example, the first identifier is a logical identifier.

[0314] As an example, the first identifier is a non-negative integer.

[0315] As an example, the first identifier is a positive integer.

[0316] As an example, the first message indicates the performance of the first model.

[0317] As an example, the first message indicates the first threshold.

[0318] As an example, the first message indicates at least a portion of the input to the first model.

[0319] As an example, the first message indicates a candidate RS resource group, which is an RS resource group other than the first RS resource group among the plurality of RS resource groups.

[0320] As an example, the first message indicates at least a portion of the output of the first model.

[0321] As an example, the first message indicates a first reason, which is one of a plurality of reasons; the first reason indicates that the performance of the first model is lower than or not higher than the first threshold.

[0322] As an example, any one of the plurality of reasons indicates the performance of the first model.

[0323] As an example, at least one of the plurality of reasons does not indicate the performance of the first model.

[0324] As an example, the plurality of reasons includes one reason indicating that the output of the first model exceeds a threshold.

[0325] As an example, the plurality of reasons includes one reason indicating that the storage amount of the first memory exceeds a threshold.

[0326] As an example, the first memory is a UE variable.

[0327] As one example, the first memory is a piece of hardware.

[0328] As one example, the first memory is a hard disk.

[0329] As one example, the first memory is a flash memory.

[0330] As one embodiment, the first memory is a storage medium.

[0331] As one example, the first memory is software.

[0332] As an example, the output of the first model is stored in the first memory.

[0333] As an example, the first model is stored in the first memory.

[0334] As an example, the input of the first model is stored in the first memory.

[0335] As an example, the parameters of the first model are stored in the first memory.

[0336] As an example, applying an RS resource group to the first model means using the actual measurement results of the RS resource group as input to the first model.

[0337] In the above method, if the application of an RS resource group to the first model is stopped, the actual measurement results of the RS resource group are not used as input to the first model.

[0338] In the above method, during the period from when an RS resource group is established until the first model is no longer applied, the actual measurement results of the RS resource group are not used as input to the first model.

[0339] As an example, applying an RS resource group to the first model means that the performance of the first model depends on the actual measurement results of the RS resource group.

[0340] In the above method, if an RS resource group is stopped from being applied to the first model, the performance of the first model does not depend on the actual measurement results of the RS resource group.

[0341] In the above method, during the period from when an RS resource group is established until the first model is no longer used, the performance of the first model does not depend on the actual measurement results of the RS resource group.

[0342] As an example, applying an RS resource group to the first model means: evaluating the performance of the first model using the actual measurement results of the RS resource group.

[0343] In the above method, if an RS resource group is stopped from being applied to the first model, the actual measurement results of the RS resource group are not used to evaluate the performance of the first model.

[0344] In the above method, during the period from when an RS resource group is used until the first model is stopped from being applied, the actual measurement results of the RS resource group are not used to evaluate the performance of the first model.

[0345] As an example, after stopping the application of an RS resource group to the first model, if the RS resource group has not been released and a signaling to start applying the RS resource group to the first model is received, the application of the RS resource group to the first model is restarted.

[0346] As an example, after stopping the application of an RS resource group to the first model, if the RS resource group is not released and a timer expires, the application of the RS resource group to the first model is restarted; wherein, the timer is started along with stopping the application of an RS resource group to the first model.

[0347] As an example, the first operation including stopping the application of at least the first RS resource group to the first model means that the first operation includes stopping the application of any one of the plurality of RS resource groups to the first model.

[0348] As an example, the first operation including stopping the application of at least the first RS resource group to the first model means that the first operation includes stopping the application of at least the first RS resource group to the first model, and starting the application of at least one RS resource group other than the at least first RS resource group to the first model.

[0349] As a sub-example of the above embodiments, at least one RS resource group other than the at least first RS resource group among the plurality of RS resource groups is indicated by the network.

[0350] As a sub-example of the above embodiment, at least one RS resource group other than the first RS resource group among the plurality of RS resource groups is determined by the first node itself.

[0351] As a sub-example of the above embodiment, at least one RS resource group other than the first RS resource group among the plurality of RS resource groups is randomly selected by the first node.

[0352] As a sub-example of the above embodiments, at least one RS resource group other than the at least first RS resource group among the plurality of RS resource groups is selected by the first node based on measurement.

[0353] As a sub-example of the above embodiment, at least one RS resource group other than the first RS resource group among the plurality of RS resource groups is selected by the first node based on the order of network configuration.

[0354] As an example, the first operation includes deactivating the first model, which includes stopping the application of at least the first RS resource group to the first model.

[0355] As one embodiment, the first operation includes releasing the first model, which includes stopping the application of at least the first RS resource group to the first model.

[0356] As one embodiment, the first operation includes stopping the first model, which includes stopping the application of at least the first RS resource group to the first model.

[0357] As one embodiment, the first operation includes deleting the first model, which includes stopping the application of at least the first RS resource group to the first model.

[0358] As one embodiment, the first operation includes updating the first model, which includes stopping the application of at least the first RS resource group to the first model.

[0359] As one embodiment, the first operation includes updating the RS resource group that evaluates the performance of the first model, the updating of the RS resource group that evaluates the performance of the first model includes stopping the application of at least the first RS resource group to the first model.

[0360] As one embodiment, the first operation includes sending a first message, and the first operation includes stopping the application of at least the first RS resource group to the first model.

[0361] Example 2

[0362] Example 2 illustrates a schematic diagram of a network architecture according to an embodiment of this application, as shown in the attached diagram. Figure 2 As shown.

[0363] Appendix Figure 2The network architecture 200 is described. The network architecture 200 is a 5G NR (New Radio) / LTE (Long-Term Evolution) / LTE-A (Long-Term Evolution Advanced) system, or a 5G+ network architecture, or a 6G network architecture, or a future evolution network architecture of 3GPP; the network architecture 200 may be referred to as 5GS (5G System) / EPS (Evolved Packet System), or 6GS (6G System); the network architecture 200 includes at least one of UE (User Equipment) 201, RAN (Radio Access Network) 202, core network 210, HSS (Home Subscriber Server) / UDM (Unified Data Management) 220, and Internet service 230. The network architecture 200 can interconnect with other access networks, but these entities / interfaces are not shown for simplicity. As shown, the network architecture 200 provides packet-switched services; however, those skilled in the art will readily understand that the various concepts presented throughout this application can be extended to networks providing circuit-switched services or other cellular networks. The RAN includes node 203 and other nodes 204. Node 203 provides user and control plane protocol termination toward UE 201. Node 203 can be connected to other nodes 204 via an Xn interface (e.g., backhaul) / X2 interface. Node 203 may also be referred to as a base station, base transceiver station, radio base station, radio transceiver, transceiver function, basic service set (BSS), extended service set (ESS), TRP (transmitter-receiver node), or some other suitable term. The core network 210 is a 5GC (5G Core Network) / EPC (Evolved Packet Core), or the core network 210 is a 6GC; node 203 provides UE 201 with an access point to the core network 210. Examples of UE201 include cellular phones, smartphones, Session Initiation Protocol (SIP) phones, laptops, personal digital assistants (PDAs), satellite radios, non-terrestrial base station communications, satellite mobile communications, global positioning systems, multimedia devices, video devices, digital audio players (e.g., MP3 players), cameras, game consoles, drones, aircraft, narrowband IoT devices, machine-type communication devices, land vehicles, automobiles, wearable devices, or any other similar functional devices.Those skilled in the art may also refer to UE201 as a mobile station, subscriber station, mobile unit, subscriber unit, radio unit, remote unit, mobile device, radio device, wireless communication device, remote device, mobile subscriber station, access terminal, mobile terminal, radio terminal, remote terminal, handheld device, user agent, mobile client, client, or any other suitable term. Node 203 is connected to the core network 210 via the 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 the control node that handles signaling between UE201 and the core network 210. In general, the MME / AMF / SMF211 provides bearer and connection management. All user IP (Internet Protocol) packets are transmitted through the S-GW / UPF212, which is itself connected to the P-GW / UPF213. The P-GW provides UE IP address allocation and other functions. The P-GW / UPF213 connects to Internet service 230. Internet service 230 includes operator-compliant Internet Protocol services, specifically including the Internet, intranet, IMS (IP Multimedia Subsystem), and packet switching services.

[0364] As one embodiment, the first node includes the UE201, and the second node belongs to the RAN202.

[0365] As one embodiment, the first node includes the UE201, the second node includes the node203, and the wireless link between the UE201 and the node203 includes a cellular network link.

[0366] As an example, the first node includes the UE201, and the second node is a higher-level node.

[0367] As one embodiment, the first node includes the UE201, the second node includes the node203, and the third node is a higher-level node; the wireless link between the UE201 and the node211 includes a cellular link, and the wireless link between the node211 and the higher-level node includes a backhaul link.

[0368] As one embodiment, the first node includes the node 203, and the second node is a higher-level node; the wireless link between the node 211 and the higher-level node includes a backhaul link.

[0369] As an example, the higher-level node belongs to the core network 210.

[0370] As an example, the higher-level node is a NAS node.

[0371] As an example, the higher-level node is an OTT server.

[0372] As an example, the higher-level node is an OAM.

[0373] As one embodiment, the high-level node includes intelligent functions, which include at least one of training or inference.

[0374] As one example, the high-level node includes an intelligent module.

[0375] As an example, the higher-level node processes AI / ML models.

[0376] Example 3

[0377] Example 3 illustrates a schematic diagram of an embodiment of a wireless protocol architecture for a user plane and a control plane according to this application, as shown in the attached diagram. Figure 3 As shown. Figure 3 This is a schematic diagram illustrating an embodiment of a radio protocol architecture for the user plane 350 and the control plane 300. Figure 3The radio protocol architecture for control plane 300 is illustrated using three layers: Layer 1, Layer 2, and Layer 3. Layer 1 (L1 layer) is the lowest layer and implements various PHY (Physical Layer) signal processing functions. L1 layer will be referred to as PHY301 in this document. Layer 2 (L2 layer) 305 sits above PHY301 and includes the MAC (Medium Access Control) sublayer 302, the RLC (Radio Link Control) sublayer 303, and the PDCP (Packet Data Convergence Protocol) sublayer 304. The PDCP sublayer 304 provides multiplexing between different radio bearers and logical channels. It also provides security through encrypted data packets and cross-area mobility support. The RLC sublayer 303 provides segmentation and reassembly of upper-layer packets, retransmission of lost packets, and packet reordering to compensate for out-of-order reception due to HARQ (Hybrid Automatic Repeat Request). MAC sublayer 302 provides multiplexing between the logical and transport channels. MAC sublayer 302 is also responsible for allocating various radio resources (e.g., resource blocks) within a cell. MAC sublayer 302 is also responsible for HARQ operations. The RRC (Radio Resource Control) sublayer 306 in Layer 3 (L3) of the control plane 300 is responsible for acquiring radio resources (i.e., radio bearers) and using RRC signaling to configure the lower layers. The radio protocol architecture of the user plane 350 includes Layer 1 (L1) and Layer 2 (L2). The radio protocol architecture in the user plane 350 is substantially the same as the corresponding layers and sublayers in the control plane 300 for Physical Layer 351, PDCP sublayer 354 in L2 Layer 355, RLC sublayer 353 in L2 Layer 355, and MAC sublayer 352 in L2 Layer 355. However, PDCP sublayer 354 also provides header compression for upper layer packets to reduce radio transmission overhead. The L2 layer 355 in the user plane 350 also includes the SDAP (Service Data Adaptation Protocol) sublayer 356. The SDAP sublayer 356 is responsible for the mapping between QoS streams and data radio bearers (DRBs) to support service diversity.

[0378] As an example, Appendix Figure 3 The wireless protocol architecture described herein is applicable to the first node in this application.

[0379] As an example, Appendix Figure 3The wireless protocol architecture described herein is applicable to the second node in this application.

[0380] As an example, the RS corresponding to any RS resource in any RS resource group in the plurality of RS resource groups in this application is generated in the PHY301 or PHY351.

[0381] As an example, the first message in this application is generated on the protocol layer above the PDCP304 or PDCP354.

[0382] As an example, the first message in this application is generated on the protocol layer above the RRC306.

[0383] As an example, the first message in this application is generated at the NAS layer (attached). Figure 3 (Not shown).

[0384] As an example, the first message in this application is generated at the application layer (see attached). Figure 3 (Not shown).

[0385] As an example, the first message in this application is generated in the AI / ML layer (see attached). Figure 3 (Not shown).

[0386] As an example, the first message in this application is generated in the RRC306.

[0387] As an example, the first message in this application is generated by MAC302 or MAC352.

[0388] As an example, the first message in this application is generated by the PHY301 or PHY351.

[0389] As an example, the first signaling in this application is generated on the protocol layer above the PDCP304 or PDCP354.

[0390] As an example, the first signaling in this application is generated on the protocol layer above the RRC306.

[0391] As an example, the first signaling in this application is generated at the NAS layer (attached). Figure 3 (Not shown).

[0392] As an example, the first signaling in this application is generated at the application layer (attached). Figure 3 (Not shown).

[0393] As an example, the first signaling in this application is generated at the AI / ML layer (see attached). Figure 3(Not shown).

[0394] As an example, the first signaling in this application is generated in the RRC306.

[0395] As an example, the first signaling in this application is generated in MAC302 or MAC352.

[0396] As an example, the first signaling in this application is generated in the PHY301 or PHY351.

[0397] As an example, the second signaling in this application is generated on the protocol layer above the PDCP304 or PDCP354.

[0398] As an example, the second signaling in this application is generated on the protocol layer above the RRC306.

[0399] As an example, the second signaling in this application is generated at the NAS layer (see attached). Figure 3 (Not shown).

[0400] As an example, the second signaling in this application is generated at the application layer (see attached diagram). Figure 3 (Not shown).

[0401] As an example, the second signaling in this application is generated at the AI / ML layer (see attached). Figure 3 (Not shown).

[0402] As an example, the second signaling in this application is generated in the RRC306.

[0403] As an example, the second signaling in this application is generated in MAC302 or MAC352.

[0404] As an example, the second signaling in this application is generated in the PHY301 or PHY351.

[0405] As an example, the AI / ML layer is located on top of the RRC306.

[0406] As an example, the AI / ML layer is located on top of the SDAP356.

[0407] As an example, the AI / ML layer is used to transmit data of AI / ML functions or AI / ML models.

[0408] As one example, the AI / ML layer is used to transmit control signaling for AI / ML functions or AI / ML models.

[0409] As an example, this application does not limit the name of the AI / ML layer.

[0410] Example 4

[0411] Example 4 shows schematic diagrams of a first communication device and a second communication device according to this application, as shown in the appendix. Figure 4 As shown. Figure 4 This is a block diagram of a first communication device 450 and a second communication device 410 communicating with each other in the access network.

[0412] The first communication device 450 includes a controller / processor 459, a memory 460, a data source 467, a transmitting processor 468, a receiving processor 456, a multi-antenna transmitting processor 457, a multi-antenna receiving processor 458, a transmitter / receiver 454, and an antenna 452.

[0413] The second communication device 410 includes a controller / processor 475, a memory 476, a receiver processor 470, a transmitter processor 416, a multi-antenna receiver processor 472, a multi-antenna transmitter processor 471, a transmitter / receiver 418, and an antenna 420.

[0414] In the transmission from the second communication device 410 to the first communication device 450, at the second communication device 410, upper-layer data packets from the core network are provided to the controller / processor 475. The controller / processor 475 implements L2 layer functionality. In the transmission from the second communication device 410 to the first communication device 450, the controller / processor 475 provides header compression, encryption, packet segmentation and reordering, multiplexing between logical and transport channels, and radio resource allocation to the first communication device 450 based on various priority metrics. The controller / processor 475 is also responsible for retransmitting lost packets and signaling to the first communication device 450. The transmit processor 416 and the multi-antenna transmit processor 471 implement various signal processing functions for the L1 layer (i.e., the physical layer). Transmit processor 416 performs encoding and interleaving to facilitate forward error correction (FEC) at the second communication device 410, and mapping of signal clusters based on various modulation schemes (e.g., Binary Phase Shift Keying (BPSK), Quadrature Phase Shift Keying (QPSK), M-Phase Shift Keying (M-PSK), M-QAM). Multi-antenna transmit processor 471 performs digital spatial precoding on the encoded and modulated symbols, including codebook-based and non-codebook-based precoding, and beamforming processing, generating one or more spatial streams. Transmit processor 416 then maps each spatial stream to subcarriers, multiplexes it with a reference signal (e.g., a pilot) in the time and / or frequency domains, and subsequently uses inverse fast Fourier transform (IFFT) to generate a physical channel carrying the time-domain multicarrier symbol stream. Multi-antenna transmit processor 471 then performs transmit analog precoding / beamforming operations on the time-domain multicarrier symbol stream. Each transmitter 418 converts the baseband multicarrier symbol stream provided by the multi-antenna transmitter processor 471 into an radio frequency stream, which is then provided to different antennas 420.

[0415] In the transmission from the second communication device 410 to the first communication device 450, at the first communication device 450, each receiver 454 receives a signal through its corresponding antenna 452. Each receiver 454 recovers the information modulated onto the radio frequency carrier and converts the radio frequency stream into a baseband multicarrier symbol stream, which is then provided to the receiver processor 456. The receiver processor 456 and the multi-antenna receiver processor 458 implement various signal processing functions of the L1 layer. The multi-antenna receiver processor 458 performs receive analog precoding / beamforming operations on the baseband multicarrier symbol stream from the receiver 454. The receiver processor 456 uses a Fast Fourier Transform (FFT) to convert the baseband multicarrier symbol stream after the receive analog precoding / beamforming operations from the time domain to the frequency domain. In the frequency domain, the physical layer data signal and the reference signal are demultiplexed by the receiver processor 456, where the reference signal is used for channel estimation, and the data signal is recovered in the multi-antenna receiver processor 458 after multi-antenna detection to recover any spatial stream destined for the first communication device 450. Symbols on each spatial stream are demodulated and recovered in the receive processor 456, generating soft decisions. The receive processor 456 then decodes and deinterleaves the soft decisions to recover the upper-layer data and control signals transmitted by the second communication device 410 over 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 Layer 2. The controller / processor 459 may be associated with a memory 460 storing program code and data. The memory 460 may be referred to as computer-readable media. In the transmission from the second communication device 410 to the first communication device 450, the controller / processor 459 provides multiplexing, packet reassembly, decryption, header decompression, and control signal processing between the transport and logical channels to recover upper-layer data packets from the core network. The upper-layer data packets are then provided to all protocol layers above Layer 2. Various control signals may also be provided to Layer 3 for Layer 3 processing.

[0416] In the transmission from the first communication device 450 to the second communication device 410, at the first communication device 450, a data source 467 is used to provide upper-layer data packets to the controller / processor 459. The data source 467 represents all protocol layers above the L2 layer. Similar to the transmission functions at the second communication device 410 described in the transmission from the second communication device 410 to the first communication device 450, the controller / processor 459 implements header compression, encryption, packet segmentation and reordering, and multiplexing between logical and transport channels based on radio resource allocation, implementing L2 layer functions for the user plane and control plane. The controller / processor 459 is also responsible for retransmitting lost packets and signaling to the second communication device 410. Transmit processor 468 performs modulation mapping and channel coding processing, while multi-antenna transmit processor 457 performs digital multi-antenna spatial precoding, including codebook-based and non-codebook-based precoding, and beamforming processing. Subsequently, transmit processor 468 modulates the generated spatial stream into a multi-carrier / single-carrier symbol stream. After analog precoding / beamforming operations in multi-antenna transmit processor 457, the stream is provided to different antennas 452 via transmitter 454. Each transmitter 454 first converts the baseband symbol stream provided by multi-antenna transmit processor 457 into a radio frequency symbol stream before providing it to antenna 452.

[0417] In the transmission from the first communication device 450 to the second communication device 410, the function at the second communication device 410 is similar to the receiving function at the first communication device 450 described in the transmission from the second communication device 410 to the first communication device 450. Each receiver 418 receives radio frequency signals through its corresponding antenna 420, converts the received radio frequency signals into baseband signals, and provides the baseband signals to the multi-antenna receiving processor 472 and the receiving processor 470. The receiving processor 470 and the multi-antenna receiving processor 472 jointly implement the L1 layer functions. The controller / processor 475 implements the L2 layer functions. The controller / processor 475 may be associated with a memory 476 that stores program code and data. The memory 476 may be referred to as computer-readable media. In the transmission from the first communication device 450 to the second communication device 410, the controller / processor 475 provides multiplexing between the transmission and logical channels, packet reassembly, decryption, header decompression, and control signal processing to recover upper-layer data packets from the UE 450. Upper-layer packets from the controller / processor 475 can be provided to the core network.

[0418] As one embodiment, the first communication device 450 includes: at least one processor and at least one memory, the at least one memory including computer program code; the at least one memory and the computer program code are configured to be used with the at least one processor, and the first communication device 450 at least: receives on a plurality of RS resource groups, any one of the plurality of RS resource groups being associated with a threshold; and performs a first operation as a response that the performance of a first model is lower or not higher than the first threshold; wherein the input of the first model includes a measurement for a first RS resource group, the first threshold being the threshold associated with the first RS resource group; the first RS resource group is one of the plurality of RS resource groups; the threshold associated with at least two of the plurality of RS resource groups is different; wherein the first operation includes sending a first message; and / or, the first operation includes stopping the application of at least the first RS resource group to the first model.

[0419] As one embodiment, the first communication device 450 includes: a memory storing a computer-readable instruction program that, when executed by at least one processor, produces actions including: receiving on a plurality of RS resource groups, any one of the plurality of RS resource groups being associated with a threshold; and performing a first operation as a response that the performance of a first model is below or not above the first threshold; wherein the input to the first model includes a measurement for a first RS resource group, the first threshold being a threshold associated with the first RS resource group; the first RS resource group is one of the plurality of RS resource groups; the threshold associated with at least two of the plurality of RS resource groups is different; wherein the first operation includes sending a first message; and / or, the first operation includes stopping the application of at least the first RS resource group to the first model.

[0420] As one embodiment, the second communication device 410 includes: at least one processor and at least one memory, the at least one memory including computer program code; the at least one memory and the computer program code are configured to be used with the at least one processor. The second communication device 410 at least: receives a first message; wherein the sender of the first message receives it on a plurality of RS resource groups, any one of the plurality of RS resource groups being associated with a threshold; as a response that the performance of a first model is below or not above the first threshold, the sender of the first message performs a first operation; the input to the first model includes a measurement for a first RS resource group, the first threshold being the threshold associated with the first RS resource group; the first RS resource group is one of the plurality of RS resource groups; the threshold associated with at least two of the plurality of RS resource groups is different; wherein the first operation includes sending the first message; and / or, the first operation includes sending the first message and stopping the application of at least the first RS resource group to the first model.

[0421] As one embodiment, the second communication device 410 includes: a memory storing a computer-readable instruction program that, when executed by at least one processor, produces an action including: receiving a first message; wherein the sender of the first message receives it on a plurality of RS resource groups, any one of the plurality of RS resource groups being associated with a threshold; as a response that the performance of a first model is below or not above the first threshold, the sender of the first message performs a first operation; the input to the first model includes a measurement for a first RS resource group, the first threshold being a threshold associated with the first RS resource group; the first RS resource group is one of the plurality of RS resource groups; the threshold associated with at least two of the plurality of RS resource groups is different; wherein the first operation includes sending the first message; and / or, the first operation includes sending the first message and stopping the application of at least the first RS resource group to the first model.

[0422] As an example, at least one of the antenna 452, the receiver 454, the receiving processor 456, and the controller / processor 459 is used to receive RS on at least one RS resource in at least one of the plurality of RS resource groups.

[0423] As an example, at least one of the antenna 420, the transmitter 418, the transmission processor 416, and the controller / processor 475 is used to transmit RS on at least one RS resource in at least one of the plurality of RS resource groups.

[0424] As an example, at least one of the antenna 452, the transmitter 454, the transmission processor 468, and the controller / processor 459 is used to transmit the first message.

[0425] As an example, at least one of the antenna 420, the receiver 418, the receiving processor 470, and the controller / processor 475 is used to receive the first message.

[0426] As one embodiment, at least one of the antenna 452, the receiver 454, the receiving processor 456, and the controller / processor 459 is used to receive the first signaling.

[0427] As one embodiment, at least one of the antenna 420, the transmitter 418, the transmission processor 416, and the controller / processor 475 is used to transmit the first signaling.

[0428] As an example, the first communication device 450 corresponds to the first node in this application.

[0429] As an example, the first node in this application includes the first communication device 450.

[0430] As an example, the second communication device 410 corresponds to the second node in this application.

[0431] As an example, the second node in this application includes the second communication device 410.

[0432] As an example, the first communication device 450 is a user equipment.

[0433] As an example, the first communication device 450 is a base station device.

[0434] As an example, the first communication device 450 is a relay device.

[0435] As one embodiment, the second communication device 410 is a user equipment.

[0436] As one embodiment, the second communication device 410 is a base station device.

[0437] As one embodiment, the second communication device 410 is a high-level node.

[0438] As one embodiment, the second communication device 410 is a relay device.

[0439] Example 5

[0440] Example 5 illustrates a wireless signal transmission flowchart according to an embodiment of this application, as shown in the attached diagram. Figure 5 As shown. It should be noted that the order in this example does not limit the order of signal transmission and implementation in this application.

[0441] for First node U01 In step S5101, a first signaling is received, wherein the first signaling configures an associated threshold for each of the plurality of RS resource groups; in step S5102, it is received on the plurality of RS resource groups, wherein any RS resource group among the plurality of RS resource groups is associated with a threshold; in step S5103, the performance of the first model is lower than or not higher than the first threshold; in step S5104, as a response to the performance of the first model being lower than or not higher than the first threshold, a first operation is performed, the first operation including sending a first message; in step S5105, as a response to the performance of the first model being lower than or not higher than the first threshold, a first operation is performed, the first operation including stopping the application of at least the first RS resource group to the first model.

[0442] for Target node N02 In step S5201, the first signaling is sent; in step S5202, it is sent on multiple RS resource groups; in step S5203, the first message is received.

[0443] In Example 5, the input to the first model includes a measurement for a first RS resource group, and the first threshold is a threshold associated with the first RS resource group; the first RS resource group is one of the plurality of RS resource groups; and the threshold associated with at least two of the plurality of RS resource groups is different.

[0444] As an example, the target node N02 is the second node in this application.

[0445] As one embodiment, the target node N02 includes the second node in this application.

[0446] As an example, the target node N02 is at least one RAN node.

[0447] As an example, the target node N02 is at least one higher-level node.

[0448] As an example, the target node N02 includes at least one RAN node and at least one higher-level node.

[0449] As an example, the at least one RAN node is a base station device.

[0450] As one example, the at least one RAN node is a plurality of base station devices.

[0451] As an example, the at least one high-level node is a high-level node.

[0452] As one embodiment, the at least one high-level node is a plurality of high-level nodes.

[0453] As an example, the at least one RAN node in the target node N02 includes a TRP.

[0454] As an example, the at least one RAN node in the target node N02 includes at least one DU.

[0455] As an example, the at least one RAN node in the target node N02 includes at least one CU.

[0456] As an example, the at least one RAN node in the target node N02 includes at least one NB.

[0457] As an example, the at least one RAN node in the target node N02 includes at least one gNB.

[0458] As an example, the at least one RAN node in the target node N02 includes at least one 6GNB.

[0459] As an example, at least one RAN node in the target node N02 transmits on multiple RS resource groups.

[0460] As an example, each of the at least one RAN nodes in the target node N02 transmits on at least one RS resource group among a plurality of RS resource groups.

[0461] As an example, at least one RAN node in the target node N02 sends the first signaling.

[0462] As an example, at least one higher-level node in the target node N02 sends the first signaling.

[0463] As an example, the dashed box F5.1 is optional.

[0464] As an example, the dashed box F5.1 does not exist.

[0465] As an example, the dashed box F5.1 is present.

[0466] As an example, for each of the plurality of RS resource groups, the first signaling indicates a candidate threshold from L candidate thresholds as the associated threshold.

[0467] As an example, more than one RS resource group among the plurality of RS resource groups is configured with the same candidate threshold.

[0468] As an example, the number of RS resource groups in the plurality of RS resource groups is greater than L.

[0469] As one embodiment, the first signaling configures the plurality of RS resource groups.

[0470] As an example, the threshold associated with any one of the plurality of RS resource groups is configurable.

[0471] As an example, any RS resource group among the plurality of RS resource groups is associated with the same candidate for the threshold.

[0472] As an example, at least two of the plurality of RS resource groups are associated with different candidates for the threshold.

[0473] As an example, the first signaling is transmitted over the air interface.

[0474] As an example, the first signaling is higher-level signaling.

[0475] As an example, the first signaling includes a NAS message.

[0476] As one embodiment, the first signaling includes an RRC container.

[0477] As an example, the first signaling is transmitted via an RB (Radio Bearer).

[0478] As one embodiment, the first signaling and the first message are transmitted through the same RB.

[0479] As one embodiment, the first signaling and the first message are transmitted through different RBs.

[0480] As an example, the first signaling includes an RRC message.

[0481] As one embodiment, the first signaling includes an RRC connection reconfiguration message.

[0482] As an example, the RRC connection reconfiguration message is an RRCReconfiguration message.

[0483] As an example, the name of the RRC connection reconfiguration message includes "Reconfiguration".

[0484] As one embodiment, the first signaling includes an RRC connection recovery message.

[0485] As an example, the RRC connection reconfiguration message is an RRCResume message.

[0486] As an example, the name of the RRC connection reconfiguration message includes Resume.

[0487] As one embodiment, the first signaling includes at least one RRC IE.

[0488] As an example, the first signaling includes at least one RRC field.

[0489] As one example, the first message includes low-level signaling.

[0490] As an example, the higher-layer signaling in the first signaling configures a candidate threshold associated with the one threshold for each of the plurality of RS resource groups; the lower-layer signaling in the first signaling activates the associated one threshold for at least one of the plurality of RS resource groups.

[0491] As an example, the first signaling indicates the threshold associated with each of the plurality of RS resource groups.

[0492] As an example, the first signaling indicates a target threshold and at least one offset, wherein the threshold associated with each of the plurality of RS resource groups depends on at least one of the target threshold and the at least one offset.

[0493] As an example, the threshold associated with one of the plurality of RS resource groups is equal to the sum of the target threshold and one of the at least one offset.

[0494] As an example, the threshold associated with one of the plurality of RS resource groups is the target threshold.

[0495] As an example, the dashed box F5.2 is optional.

[0496] As an example, the dashed box F5.2 does not exist.

[0497] As an example, the dashed box F5.2 is present.

[0498] As an example, the first message terminates at the target node N02.

[0499] As an example, the first message terminates in one of the RAN nodes in the target node N02.

[0500] As an example, the first message terminates at a higher-level node in the target node N02.

[0501] As a sub-implementation of the above embodiment, the first operation includes sending an RRC container; the RRC container includes the first message; as a response from a RAN node in the target node N02 to receiving the RRC container, the RAN node in the target node N02 forwards the first message to a higher-layer node in the target node N02.

[0502] As an example, step S5105 is optional.

[0503] As an example, step S5105 is present.

[0504] As an example, step S5105 is not present.

[0505] As an example, the dashed box F5.2 exists, and the step S5105 exists.

[0506] As an example, the dashed box F5.2 exists, but step S5105 does not exist.

[0507] As an example, the dashed box F5.2 does not exist, but step S5105 does exist.

[0508] Example 6

[0509] Example 6 illustrates a wireless signal transmission flowchart according to another embodiment of this application, as shown in the attached diagram. Figure 6 As shown. It should be noted that the order in this example does not limit the order of signal transmission and implementation in this application.

[0510] for First node U01In step S6101, as a response where the performance of the first model is lower than or not higher than a first threshold, a first operation is performed, the first operation including sending a first message; in step S6102, the first message is received.

[0511] As one embodiment, the first message is sent at the first protocol layer of the first node U01; the first message is received at the second protocol layer of the first node U01.

[0512] As an example, the first message terminates at the first node U01.

[0513] As an example, the first node U01 sends and receives the first signaling.

[0514] As an example, Appendix Figure 5 The dashed box F5.1 in the text can be replaced by steps S6101 and S6102.

[0515] As one embodiment, the first operation includes sending a first message, and the first operation does not include stopping the application of at least the first RS resource group to the first model.

[0516] As one embodiment, the first operation includes sending a first message and stopping the application of at least the first RS resource group to the first model.

[0517] As a sub-implementation of the above embodiments, the sending of the first message and the stopping of applying at least the first RS resource group to the first model are executed simultaneously.

[0518] As a sub-implementation of the above embodiments, in response to the performance of the first model being lower or not higher than a first threshold, a first message is sent; in response to the first message being received, the application of at least the first RS resource group to the first model is stopped.

[0519] As a sub-implementation of the above embodiments, receiving the first message triggers the stop application of at least the first RS resource group to the first model.

[0520] As one embodiment, the first protocol layer of the first node U01 sends the first message; the second protocol layer of the first node U01 receives the first message.

[0521] As one embodiment, the first protocol layer is below the second protocol layer.

[0522] As one embodiment, the first protocol layer is on top of the second protocol layer.

[0523] As an example, the first protocol layer is the AI / ML layer.

[0524] As an example, the first model runs on the first protocol layer.

[0525] As an example, the first model is deployed on the first protocol layer.

[0526] As one embodiment, the second protocol layer is the RRC sublayer.

[0527] As one example, the second protocol layer is the MAC sublayer.

[0528] As one embodiment, the second protocol layer is the physical layer.

[0529] Example 7

[0530] Example 7 illustrates a wireless signal transmission flowchart according to yet another embodiment of this application. (See attached diagram) Figure 7 Step S7101 in the appendix can be replaced Figure 5 Step S5103 in the appendix Figure 7 Step S7104 in the appendix can be replaced Figure 5 For step S5104, please refer to the appendix for other steps. Figure 5 I won't go into details here.

[0531] for First node U01 In step S7101, the performance of the first model is lower than or not higher than a first threshold; in step S7102, a second signaling is received; wherein the second signaling triggers the first message; in step S7103, a first event occurs; the first event triggers the first message; wherein the first event depends on a second threshold, which is configurable; in step S7104, as a response to the performance of the first model being lower than or not higher than the first threshold, a first operation is performed; wherein the first operation includes sending a first message.

[0532] for Target node N02 In step S7201, the second signaling is sent; in step S7202, the first message is received.

[0533] As an example, at least one RAN node in the target node N02 sends the second signaling and receives the first message.

[0534] As an example, at least one higher-level node in the target node N02 sends the second signaling and receives the first message.

[0535] As an example, the first node U01 receives another RRC container; the other RRC container includes the first signaling.

[0536] As a sub-example of the above embodiment, at least one higher-level node in the target node N02 sends the first message to at least one RAN node in the target node N02; the at least one RAN node in the target node N02 encapsulates the first message in another RRC container and sends it to the first node U01.

[0537] As one embodiment, the second signaling is generated in a higher-level node of the target node N02.

[0538] As one embodiment, the second signaling is generated in a RAN node of the target node N02.

[0539] As one example, the second signaling requests the first message.

[0540] As one embodiment, the second signaling indicates the first message.

[0541] As one example, the second signaling and the first message belong to the same protocol layer.

[0542] As one example, the second signaling and the first message belong to different protocol layers.

[0543] As one example, the second signaling is higher-level signaling.

[0544] As one example, the second signaling includes a NAS message.

[0545] As one embodiment, the second signaling includes an RRC container.

[0546] As an example, the second signaling is transmitted via an SRB.

[0547] As one example, the second signaling and the associated first message are transmitted through the same SRB.

[0548] As one example, the second signaling and the associated first message are transmitted through different SRBs.

[0549] As one example, the second signaling includes an RRC message.

[0550] As one embodiment, the second signaling includes at least one RRC IE.

[0551] As one embodiment, the second signaling includes at least one RRC field.

[0552] As one embodiment, the second signaling includes a UEInformationRequest message.

[0553] As one embodiment, the name of the second signaling includes Request or request; the name of the first message includes Response or response.

[0554] As an example, the name of the second signaling includes a field whose name includes "req".

[0555] As one example, the second signaling is low-level signaling.

[0556] As an example, the second signaling is a MAC CE.

[0557] As an example, the second signaling is a MAC subheader.

[0558] As an example, the second signaling is a DCI (Downlink Control Information).

[0559] As an example, the second signaling is a PDCCH (Physical Downlink Control Channel) transmission.

[0560] As an example, the second signaling is a PDSCH (Physical Downlink SharedChannel) transmission.

[0561] As one embodiment, the first operation includes storing relevant information about the first model in a first memory; wherein the relevant information about the first model includes information from the first message.

[0562] The above method, by storing relevant information of the first model, facilitates subsequent network scheduling of the relevant information of the first model.

[0563] As an example, the first operation includes storing relevant information of the first model in a first memory, and sending a first message indicating that the relevant information of the first model is stored in the first memory; wherein the relevant information of the first model includes the information in the first message.

[0564] The above method is helpful in assisting the network to schedule the relevant information of the first model.

[0565] As an example, the first information occupies one field.

[0566] As an example, the value of the first information is set to true.

[0567] As an example, the first information is one bit.

[0568] As an example, the first information is a code point.

[0569] As one embodiment, the first event includes: the storage amount of the first memory exceeds a threshold.

[0570] As an example, the first event includes: the relevant information of the first model being stored in the first memory for a period of time exceeding a threshold.

[0571] As an example, the first event includes: the relevant information of the first model being stored in the first memory for a period of time exceeding a threshold; the threshold being less than another threshold; and, assuming that the relevant information of the first model has been stored in the first memory for a period of time exceeding the other threshold, the relevant information of the first model being deleted from the first memory.

[0572] Example 8

[0573] Example 8 illustrates a schematic diagram of a first message indicating a first RS resource group according to an embodiment of this application.

[0574] In embodiment 8, the first operation includes sending a first message; the first message indicates the first RS resource group.

[0575] As an example, the first message indicates the index of the first RS resource group.

[0576] As an example, the first message indicates the index of at least one RS resource in the first RS resource group.

[0577] As an example, the first message indicates the actual measurement result obtained after the measurement for the first RS resource group.

[0578] As an example, the first message indicates the output of the first model and the actual measurement result, which is obtained after the measurement for the first RS resource group.

[0579] As an example, the first message indicates the error between the output of the first model and the actual measurement result, which is obtained after the measurement for the first RS resource group.

[0580] As one embodiment, the first message includes a first identifier, which indicates the first model.

[0581] As an example, the first message indicates the performance of the first model.

[0582] As an example, the first message indicates the first threshold.

[0583] As an example, the first message indicates at least a portion of the input to the first model.

[0584] As an example, the first message indicates at least a portion of the output of the first model.

[0585] As an example, the first message indicates a first reason, which is one of a plurality of reasons; the first reason indicates that the performance of the first model is lower than or not higher than the first threshold.

[0586] Example 9

[0587] Example 9 illustrates a schematic diagram of the performance of a first model according to an embodiment of this application. In the appendix... Figure 9 In the diagram, the horizontal axis represents time. At time t1, the output of the first model is obtained; at time t2, the actual measurement result is obtained; time t1 is earlier than time t2.

[0588] In Example 9, the performance of the first model depends on the error between the output of the first model and the actual measurement result, which is obtained after the measurement for the first RS resource group.

[0589] As an example, the input to the first model includes measurements for the first RS resource group; the actual measurement result is obtained by measuring at least one RS resource; the performance of the first model depends on the error between the output of the first model and the actual measurement result, which is obtained after the measurement for the first RS resource group.

[0590] As a sub-implementation of the above embodiments, the at least one RS resource is the output of the first model.

[0591] As a sub-implementation of the above embodiments, the at least one RS resource is the RS resource corresponding to the inference result of the first model.

[0592] As a sub-implementation of the above embodiments, the at least one RS resource is the first RS resource group.

[0593] As a sub-implementation of the above embodiments, the at least one RS resource is not the first RS resource group.

[0594] As a sub-implementation of the above embodiments, any one of the at least one RS resources is an RS resource in the first RS resource group.

[0595] As a sub-implementation of the above embodiments, the at least one RS resource includes at least one RS resource that does not belong to the first RS resource group.

[0596] As a sub-implementation of the above embodiments, the at least one RS resource includes at least one RS resource that does not belong to the first RS resource group.

[0597] As a sub-implementation of the above embodiment, the first RS resource group is a set B, and the at least one RS resource is a set A.

[0598] As an example, the performance of the first model depends on the error between a single output of the first model and a single actual measurement result.

[0599] As an example, the performance of the first model depends on the error between multiple outputs of the first model and multiple actual measurement results.

[0600] As an example, the performance of the first model depends on the error between a single output of the first model and multiple actual measurement results.

[0601] As an example, the output of the first model is used for monitoring, and the actual measurement results are used for comparison.

[0602] As an example, the output of the first model is obtained before the measurement for the first RS resource group.

[0603] As an example, the output of the first model includes the inference measurement results for the first RS resource group.

[0604] As an example, the output of the first model includes system performance obtained based on the inference measurement results for the first RS resource group.

[0605] As an example, the output of the first model includes the at least one RS resource.

[0606] As an example, the output of the first model includes the predicted measurement results of the at least one RS resource.

[0607] As an example, the output of the first model includes the prediction results of the at least one RS resource.

[0608] As an example, the actual measurement results are based on measurements for at least one RS resource.

[0609] As an example, the actual measurement result is RRSP.

[0610] As an example, the actual measurement result is RRSQ.

[0611] As an example, the actual measurement result is SINR.

[0612] As an example, the actual measurement results are beam-level.

[0613] As an example, the actual measurement results are at the cell level.

[0614] As an example, the actual measurement result is L1.

[0615] As an example, the actual measurement result is L3.

[0616] As an example, the unit of the actual measurement result is dB.

[0617] As an example, the unit of the actual measurement result is dBm.

[0618] As an example, the performance of the first model is the error between the output of the first model and the actual measurement result.

[0619] As an example, the performance of the first model is the value of an error function; wherein the error function includes the output of the first model and the actual measurement result.

[0620] As an example, the error is based on statistics.

[0621] As an example, the error is a one-time event.

[0622] As an example, the error is MSE (Mean Square Error).

[0623] As an example, the error is MMSE (Minimum Mean Square Error).

[0624] As an example, the error is a difference.

[0625] As an example, the error is the absolute value of the difference.

[0626] As an example, the error is the standard deviation.

[0627] As an example, the error is an absolute error.

[0628] As an example, the error is accuracy.

[0629] As an example, the error is the covariance.

[0630] As an example, the error is a probability.

[0631] As an example, the error is a constant.

[0632] As an example, the error is a dimensionless quantity.

[0633] As an example, the unit of the error is dB.

[0634] As an example, the unit of the error is dBm.

[0635] As an example, the performance of the first model being lower than or not higher than a first threshold means that the error between the output of the first model and the actual measurement result is greater than or not less than the threshold.

[0636] As a sub-example of the above embodiments, the actual measurement result is RSRP, RSRQ, or SINR.

[0637] As an example, the performance of the first model being lower than or not higher than a first threshold means that the error between the output of the first model and the actual measurement result is less than or not greater than the threshold.

[0638] As a sub-example of the above embodiments, the actual measurement result is BLER.

[0639] As an example, the performance of the first model being lower than or not higher than a first threshold means that the error between the output of the first model and the actual measurement result is greater than or not less than the threshold.

[0640] As an example, the performance of the first model being lower than or not higher than a first threshold means that the error between the output of the first model and the actual measurement result is less than or not greater than the threshold.

[0641] As an example, the performance metric of the first model includes the error between the output of the first model and the actual measurement result.

[0642] As an example, the performance of the first model is measured by the error between the output of the first model and the actual measurement result.

[0643] Example 10

[0644] Example 10 illustrates a schematic diagram showing multiple RS resource groups associated with multiple cells according to an embodiment of this application, as shown in the attached diagram. Figure 10 As shown. In the appendix Figure 10 In the diagram, box 1001 represents multiple RS resource groups, box 1002 represents multiple cells, RS resource group #1 is associated with cell #1, RS resource group #2 is associated with cell #2, and so on.

[0645] In Example 10, the plurality of RS resource groups are associated with a plurality of cells; wherein the plurality of cells include at least one serving cell.

[0646] As an example, the plurality of RS resource groups is two RS resource groups.

[0647] As an example, the plurality of RS resource groups is more than two RS resource groups.

[0648] As an example, the cell associated with the first RS resource group is a serving cell of the first node; the cell associated with any RS resource group other than the first RS resource group among the plurality of RS resource groups is a neighboring cell.

[0649] As an example, the cell associated with the first RS resource group is a serving cell of the first node; the cell associated with any RS resource group other than the first RS resource group among the plurality of RS resource groups is a candidate cell.

[0650] As a sub-example of the above embodiments, the serving cell is PCell, and the candidate cell is an LTM (L1 / L2 Triggered Mobility) candidate cell.

[0651] As a sub-example of the above embodiments, the serving cell is PCell, and the candidate cell is a CHO (Conditional Handover) candidate cell.

[0652] As a sub-example of the above embodiments, the serving cell is a PSCell, and the candidate cell is an LTM candidate cell.

[0653] As a sub-example of the above embodiment, the serving cell is a PSCell, and the candidate cell is a CPC (Conditional PSCell Change) candidate cell.

[0654] As an example, the cell associated with any one of the plurality of RS resource groups is a serving cell.

[0655] As a sub-example of the above embodiments, the serving cell is any serving cell in the MCG.

[0656] As a sub-example of the above embodiments, the serving cell is any serving cell in the SCG.

[0657] As an example, the first message may assist the network in reconfiguring, releasing, or deactivating the cell associated with the first RS resource group.

[0658] Example 11

[0659] Example 11 illustrates a schematic diagram showing multiple RS resource groups associated with multiple functions according to an embodiment of this application, as shown in the attached diagram. Figure 11 As shown. In the appendix Figure 11 In the diagram, box 1101 represents multiple RS resource groups, box 1102 represents multiple functions, RS resource group #1 is associated with function #1, RS resource group #2 is associated with function #2, and so on.

[0660] In Example 11, the plurality of RS resource groups are associated with a plurality of functions respectively; wherein any one of the plurality of functions adopts the first model.

[0661] As an example, the function refers to a use case.

[0662] As an example, the term "function" refers to "functionality".

[0663] As an example, the function refers to AI / ML functionality.

[0664] As an example, one of the multiple functions may employ multiple models, and the first model is one of the multiple models.

[0665] As an example, any one of the plurality of functions may use only the first model.

[0666] As an example, the plurality of RS resource groups is two RS resource groups.

[0667] As an example, the plurality of RS resource groups is more than two RS resource groups.

[0668] As an example, one of the multiple functions is used for CSI compression.

[0669] As an example, one of the multiple functions is used for beam management.

[0670] As an example, one of the multiple functions is used for positioning.

[0671] As an example, one of the multiple functions is used for mobility management.

[0672] As an example, one of the multiple functions is used for RLF (Radio Link Failure) prediction.

[0673] As an example, one of the multiple functions is used for HOF (Handover Failure) prediction.

[0674] As an example, one of the multiple functions is used to switch predictions.

[0675] As an example, one of the multiple functions is used to trigger event prediction.

[0676] As an example, one of the multiple functions is used for network energy saving.

[0677] As an example, "any one of the plurality of functions adopts the first model" means that any one of the plurality of functions is configured with the first model.

[0678] As an example, the first node receives multiple RRC signaling messages, each of which configures parameters for the multiple functions. Each of the multiple RRC signaling messages includes an RRC information block, which indicates the first model.

[0679] As an example, "any one of the plurality of functions adopts the first model" means that the first model is configured with any one of the plurality of functions.

[0680] As an example, the first node receives an RRC signaling message, which configures the parameters of the first model. The RRC signaling message includes an RRC information block, which indicates the plurality of functions.

[0681] As an example, the first message indicates at least one of the plurality of functions.

[0682] As an example, the first message indicates the function among the plurality of functions associated with the first RS resource group.

[0683] As an example, the first message may assist the network in activating, reconfiguring, releasing, or suspending the functions associated with the first RS resource group.

[0684] Example 12

[0685] Example 12 illustrates a structural block diagram of a processing apparatus for a first node according to an embodiment of this application; as shown in the appendix. Figure 12 As shown. In the appendix Figure 12 In the first node, the processing device 1200 includes a first receiver 1201 and a first processor 1202.

[0686] A first receiver 1201 receives on a plurality of RS resource groups, each of which is associated with a threshold.

[0687] The first processor 1202, as a response where the performance of the first model is lower or not higher than a first threshold, executes the first operation;

[0688] In Example 12, the input to the first model includes a measurement for a first RS resource group, and the first threshold is a threshold associated with the first RS resource group; the first RS resource group is one of the plurality of RS resource groups; the threshold associated with at least two of the plurality of RS resource groups is different;

[0689] Wherein, the first operation includes sending a first message; and / or, the first operation includes stopping the application of at least the first RS resource group to the first model.

[0690] As one embodiment, the first receiver 1201 receives a first signaling; wherein the first signaling configures an associated threshold for each of the plurality of RS resource groups.

[0691] As an example, the first message indicates the first RS resource group.

[0692] As an example, the performance of the first model depends on the error between the output of the first model and the actual measurement result, which is obtained after the measurement for the first RS resource group.

[0693] As one embodiment, the plurality of RS resource groups are associated with a plurality of cells, wherein the plurality of cells include at least one serving cell.

[0694] As an example, the plurality of RS resource groups are associated with a plurality of functions respectively; wherein any one of the plurality of functions adopts the first model.

[0695] As one embodiment, the first operation includes receiving a second signaling; wherein the second signaling triggers the first message.

[0696] As one example, a first event triggers the first message; wherein the first event depends on a second threshold, which is configurable.

[0697] As one embodiment, the first processor 1201 includes a first transmitter.

[0698] As one embodiment, the first processor 1201 includes a first receiver 1201.

[0699] As one embodiment, the first processor 1201 includes a first receiver 1201 and a first transmitter.

[0700] As one embodiment, the first receiver 1201 includes the appendix to this application. Figure 4 The antenna 452, receiver 454, multi-antenna receiver processor 458, receiver processor 456, controller / processor 459, memory 460, or data source 467 are at least one of these.

[0701] As one embodiment, the first receiver 1201 includes the appendix to this application. Figure 4 At least antenna 452 and receiver 454 are included.

[0702] As one embodiment, the first transmitter includes the appendix to this application. Figure 4The antenna 452 or transmitter 454 or multi-antenna transmitter processor 457 or transmitter processor 468 or controller / processor 459 or memory 460 or data source 467 is at least one of them.

[0703] As one embodiment, the first transmitter includes the appendix to this application. Figure 4 At least antenna 452 and transmitter 454 are included.

[0704] Example 13

[0705] Example 13 illustrates a structural block diagram of a processing apparatus for a second node according to an embodiment of this application; as shown in the appendix. Figure 13 As shown. In the appendix Figure 13 In the second node, the processing device 1300 includes a second receiver 1302.

[0706] The second receiver 1302 receives the first message;

[0707] In Example 13, the sender of the first message receives the message on multiple RS resource groups, each of which is associated with a threshold. In response to a first model's performance being below or not above the first threshold, the sender of the first message performs a first operation. The input to the first model includes a measurement for a first RS resource group, and the first threshold is the threshold associated with the first RS resource group. The first RS resource group is one of the multiple RS resource groups. At least two of the multiple RS resource groups are associated with different thresholds.

[0708] Wherein, the first operation includes sending the first message; and / or, the first operation includes sending the first message and stopping the application of at least the first RS resource group to the first model.

[0709] As an example, the second node is a high-level node.

[0710] As an example, the second node is a RAN node.

[0711] As one example, the second node is a base station device.

[0712] As an example, the plurality of RS resource groups are sent by a single RAN node.

[0713] As an example, the plurality of RS resource groups are sent by a plurality of RAN nodes respectively; the second node is not any of the plurality of RAN nodes.

[0714] As one embodiment, the plurality of RS resource groups are sent by a plurality of RAN nodes respectively; the second node is one of the plurality of RAN nodes.

[0715] As one embodiment, the second transmitter 1301 transmits on at least one of the plurality of RS resource groups.

[0716] As one embodiment, the second transmitter 1301 transmits on each of the plurality of RS resource groups.

[0717] As one embodiment, the second transmitter 1301 transmits on at least one of the plurality of RS resource groups.

[0718] As one embodiment, the second transmitter 1301 sends a first signaling; wherein the first signaling configures an associated threshold for each of the plurality of RS resource groups.

[0719] As an example, the first message indicates the first RS resource group.

[0720] As an example, the performance of the first model depends on the error between the output of the first model and the actual measurement result, which is obtained after the measurement for the first RS resource group.

[0721] As one embodiment, the plurality of RS resource groups are associated with a plurality of cells, wherein the plurality of cells include at least one serving cell.

[0722] As an example, the plurality of RS resource groups are associated with a plurality of functions respectively; wherein any one of the plurality of functions adopts the first model.

[0723] As one embodiment, the first operation includes receiving a second signaling; wherein the second signaling triggers the first message.

[0724] As one example, a first event triggers the first message; wherein the first event depends on a second threshold, which is configurable.

[0725] As one embodiment, the second transmitter 1301 includes the appendix to this application. Figure 4 The antenna 420, transmitter 418, multi-antenna transmitter processor 471, transmitter processor 416, controller / processor 475, or memory 476 are at least one of them.

[0726] As one embodiment, the second transmitter 1301 includes the appendix to this application. Figure 4At least antenna 420 and transmitter 418 are included.

[0727] As one embodiment, the second receiver 1302 includes the appendix to this application. Figure 4 The antenna 420, receiver 418, multi-antenna receiver processor 472, receiver processor 470, controller / processor 475, or memory 476 are at least one of them.

[0728] As one embodiment, the second receiver 1302 includes the appendix to this application. Figure 4 At least antenna 420 and receiver 418 are included.

[0729] Example 14

[0730] Example 14 illustrates a schematic diagram of an AI / ML model according to an embodiment of this application, as shown in the attached diagram. Figure 14 As shown. (Attached) Figure 14 It includes Module 1, Module 2, Module 3, Module 4, and Module 5.

[0731] In Example 14, in the appendix Figure 14 In the AI / ML model shown, the first module sends a first dataset to the second module, the first module sends a second dataset to the third module, the first module sends a third dataset to the fifth module, the fifth module sends a first type of parameter set to the second module, the fifth module sends a second type of parameter set to the third module, the fifth module sends a third type of parameter set to the fourth module, the second module sends a fourth type of parameter set to the fourth module, and the fourth module sends a fifth type of parameter set to the third module.

[0732] As an example, the first module, the second module, the third module, the fourth module, and the fifth module in an AI / ML model all belong to the first node.

[0733] The above method avoids air interface signaling interaction and shortens transmission latency.

[0734] As an example, any one of the first module, second module, third module, fourth module, and fifth module in an AI / ML model does not belong to the first node.

[0735] The above method reduces the hardware complexity of the first node.

[0736] As an example, at least one of the first module, the second module, the third module, the fourth module, and the fifth module in an AI / ML model belongs to the first node; and at least one of the first module, the second module, the third module, the fourth module, and the fifth module belongs to a network node.

[0737] The above method balances the hardware complexity and transmission latency of the first node.

[0738] As an example, the first module is used for data collection.

[0739] As an example, the first module is responsible for data collection.

[0740] As an example, the first module has a data collection function.

[0741] As one example, the second module has a training function.

[0742] As one example, the training function is used for AI / ML model training.

[0743] As an example, the training function is responsible for training the AI / ML model.

[0744] As an example, the training function includes AI / ML model training capabilities.

[0745] As an example, the training function performs AI / ML model training.

[0746] As an example, the second module performs validation.

[0747] As an example, the second module performs testing.

[0748] As an example, the second module generates AI / ML model performance metrics.

[0749] As one example, the second module is responsible for data preparation.

[0750] As one embodiment, the data preparation includes at least one of data pre-processing, cleaning, formatting, or transformation.

[0751] As an example, the third module has reasoning capabilities.

[0752] As an example, the inference function is used for inference.

[0753] As an example, the reasoning function is responsible for reasoning.

[0754] As one example, the fourth module is used for AI / ML model storage.

[0755] As an example, the fourth module has AI / ML model storage functionality.

[0756] As an example, the fourth module is responsible for storing the trained AI / ML model.

[0757] As an example, the fourth module is responsible for storing trained AI / ML models that can be used to perform inference processing.

[0758] As one example, the fifth module is used for management.

[0759] As an example, the fifth module is responsible for management.

[0760] As one example, the fifth module has management functions.

[0761] As an example, the fifth module manages the AI / ML model.

[0762] As an example, the first dataset is training data.

[0763] As an example, the first dataset is the input to the second module.

[0764] As an example, the first dataset includes at least a portion of the data on the first logical channel.

[0765] As an example, the second dataset is inference data.

[0766] As an example, the second dataset is the input to the third module.

[0767] As one embodiment, the second dataset includes at least a portion of the data on the first logical channel.

[0768] As an example, the third dataset is monitoring data.

[0769] As an example, the third dataset is the input to the fifth module.

[0770] As an example, the third dataset includes at least a portion of the data on the first logical channel.

[0771] As an example, the first type of parameter group includes monitoring output.

[0772] As one example, the second type of parameter group includes management instructions.

[0773] As an example, the second type of parameter group is used for fine-tuning operations of the inference function.

[0774] As an example, the second type of parameter group includes the identifier of the AI / ML model.

[0775] As an example, the second group of parameters is used to select the AI / ML model.

[0776] As an example, the second type of parameter group is used to switch between AI / ML models.

[0777] As an example, the second type of parameter group is used to activate / deactivate the AI / ML model.

[0778] As an example, the second type of parameter group is used to fall back the AI / ML model.

[0779] As an example, the third group of parameters includes AI / ML model transfer requests.

[0780] As an example, the third type of parameter group includes AI / ML model delivery requests.

[0781] As an example, the fourth parameter group includes trained AI / ML models.

[0782] As an example, the fourth group of parameters includes the updated AI / ML model.

[0783] As an example, the fourth group of parameters indicates the identifier of the AI / ML model.

[0784] As an example, the fifth parameter group includes AI / ML model transfer.

[0785] As an example, the fifth parameter group includes AI / ML model delivery.

[0786] As an example, the fifth parameter group indicates the identifier of the AI / ML model.

[0787] As an example, the first type of output does not exist.

[0788] As an example, the first type of output exists.

[0789] As one embodiment, the first type of output includes at least a portion of the data on the first logical channel.

[0790] As an example, the second module sends the first type of output to the fifth module.

[0791] As an example, the first type of output includes monitoring output.

[0792] As an example, the second type of output does not exist.

[0793] As an example, the second type of output exists.

[0794] As one embodiment, the second type of output includes at least a portion of the data on the first logical channel.

[0795] As an example, the third module sends the second type of output to the fifth module.

[0796] As an example, the second type of output includes inference output.

[0797] As an example, the second type of output is used by the fifth module to monitor the performance of the AI / ML model.

[0798] As an example, the data on the first logical channel includes at least a portion of the first dataset in the AI / ML model.

[0799] As an example, the first dataset in the AI / ML model is configured by the network.

[0800] As an example, the first dataset in the AI / ML model is determined by the first node.

[0801] As an example, the first dataset in the AI / ML model includes the stored data of the first node; the stored data may come from the network, the logs of the first node, or other RAN nodes.

[0802] As an example, the first dataset in the AI / ML model includes measurement information of the first node; the measurement information may be the movement state of the first node, such as movement speed, or the number of cells switched within a given time interval; the measurement information may also be measurement results for a reference signal, such as cell-level measurement results, or beam-level measurement results, or time-domain measurement results, or frequency-domain measurement results, or spatial-domain measurement results, or a combination thereof.

[0803] As an example, the data on the first logical channel includes at least a portion of the second dataset in the AI / ML model.

[0804] As an example, the second dataset in the AI / ML model is configured by the network.

[0805] As an example, the second dataset in the AI / ML model is determined by the first node.

[0806] As an example, the second dataset in the AI / ML model includes the stored data of the first node; the stored data may come from the network, the logs of the first node, or other RAN nodes.

[0807] As an example, the second dataset in the AI / ML model includes measurement information of the first node; the measurement information may be the movement state of the first node, such as movement speed, or the number of cells switched within a given time interval; the measurement information may also be measurement results for a reference signal, such as cell-level measurement results, or beam-level measurement results, or time-domain measurement results, or frequency-domain measurement results, or spatial-domain measurement results, or a combination thereof.

[0808] As an example, the data on the first logical channel includes at least a portion of the third dataset in the AI / ML model.

[0809] As an example, the third dataset in the AI / ML model is configured by the network.

[0810] As an example, the third dataset in the AI / ML model is determined by the first node.

[0811] As an example, the third dataset in the AI / ML model includes the stored data of the first node; the stored data may come from the network, the logs of the first node, or other RAN nodes.

[0812] As an example, the third dataset in the AI / ML model includes measurement information of the first node; the measurement information may be the movement state of the first node, such as movement speed, or the number of cells switched within a given time interval; the measurement information may also be measurement results for a reference signal, such as cell-level measurement results, or beam-level measurement results, or time-domain measurement results, or frequency-domain measurement results, or spatial-domain measurement results, or a combination thereof.

[0813] As an example, Example 14 is merely to illustrate that this application can be used in AI / ML models. This example does not limit the application of this application to non-AI / ML operations, nor does it limit the application of this application to other types of AI / ML models to obtain and attach... Figure 14 The AI / ML model shown has comparable performance.

[0814] Example 15

[0815] Example 15 illustrates a schematic diagram of intelligent function deployment in a RAN (Radio Access Network) domain according to an embodiment of this application; as shown in the appendix. Figure 15 As shown. In Example 15, the gNB can be replaced with, for example, an eNB, or a network device such as a 6G base station.

[0816] Intelligent functions in the RAN domain include training (also known as ML training, AI training, or AI / ML training), testing (also known as ML testing, AI testing, or AI / ML testing), and inference (also known as ML inference, AI inference, or AI / ML inference), among others. Training, testing, and inference functions can be deployed independently or co-located. Deployment of intelligent functions can be achieved through software, such as downloading and / or running executable files; or through a combination of software and hardware, such as accelerating specific computing units through hardware to improve processing speed or save power.

[0817] Training functions 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, training functions for MDA (Management Data Analytics) can be deployed in MDAF (MDA Function); training functions for network data analytics can be deployed in NWDAF (Network Data Analytics Function), meaning the training function is MTLF (Model Training Logical Function).

[0818] Similarly, inference functions can be deployed in cross-domain management systems or domain-specific management systems; for example, the inference function is an MDAF, or the inference function is an AnLF (Analytics logical function) located in an NWDAF.

[0819] Similarly, testing functionality can also be deployed in cross-domain management systems or domain-specific management systems.

[0820] In embodiment 15, the training function 1702 of the RAN domain is located in the management function 1703 of the RAN domain; while the inference function is located in the base station, that is, inference function 1704 is located in gNB 1705, and inference function 1706 is located in gNB 1707. Figure 15 The ellipsis in the text indicates other gNBs that include other reasoning functions and are not shown.

[0821] Appendix Figure 15 In this context, the management of inference functions for multiple base stations is handled by the RAN domain management function 1703, which interacts with the RAN domain MnS (Management Service) consumer / cross-domain management 1701 (as shown in the attached diagram). Figure 15 (As shown by the dashed arrow 1708 in the image).

[0822] Optionally, the management of inference functions can also be completed by the base station itself, that is, each base station can independently interact with the RAN domain MnS consumer / cross-domain management 1701.

[0823] It should be noted that Embodiment 15 is merely a non-limiting implementation; optionally, the RAN domain training function may also be deployed at the base station; or optionally, some base stations may deploy both inference function and RAN domain training function, while some base stations may only deploy inference function.

[0824] As an example, one of the gNBs (or base stations) in Example 15 is the second node of this application.

[0825] As an example, the appendix described in this application Figure 2 The node 203 in the middle includes the attached Figure 15 In the RAN domain MnS consumer / cross-domain management 1701.

[0826] As an example, the appendix described in this application Figure 2 The node 203 in the middle includes the attached Figure 15 The training function 1702 in the middle.

[0827] As an example, the appendix described in this application Figure 2 The node 203 in the middle includes the attached Figure 15 Management functions in 1703.

[0828] As an example, the appendix described in this application Figure 2 The node 203 in the middle includes the attached Figure 15 The reasoning function in 1705.

[0829] As an example, the appendix described in this application Figure 2 The node 211 in the middle includes the attached Figure 15 In the RAN domain MnS consumer / cross-domain management 1701.

[0830] As an example, the appendix Figure 15 The training function 1702 in the middle performs training based on the data received on the first logical channel.

[0831] As an example, the appendix Figure 15 The management function 1703 manages the data received on the first logical channel.

[0832] As an example, the appendix Figure 15 The inference function in the middle performs inference based on the data received on the first logical channel.

[0833] As an example, the appendix Figure 15 The RAN domain MnS consumer / cross-domain management 1701 performs training and / or inference based on the data received on the first logical channel.

[0834] Example 16

[0835] Example 16 illustrates a schematic diagram of UE smart function deployment according to an embodiment of this application; as shown in the appendix. Figure 16 As shown. (Attached) Figure 16 The training function 1805 for the RAN domain is optional.

[0836] The UE intelligent function 1804 is deployed in the first node of this application. The UE intelligent function 1804 includes an inference function 1806. The inference function 1806 uses an AI / ML model (also known as an AI model, or an ML model, or an AI / ML model) for inference. An AI / ML model is typically trained before being used for AI / ML inference.

[0837] As an example, the UE intelligent function 1804 includes a RAN domain training function 1805, which runs training data through an AI / ML model to obtain a relevant loss and adjusts the parameters of the AI / ML model based on the calculated loss; the training includes at least one of ML initial training, ML re-training, and reinforcement learning.

[0838] 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 capabilities of the UE side.

[0839] Optionally, the UE intelligent function 1804 also includes a CN domain training function. Figure 16 (Not included in the text).

[0840] Optionally, the UE intelligent function 1804 also includes an intelligent deployment function. Figure 16 It does not include the means to load AI / ML models and data.

[0841] As an example, the first node indicates whether it supports training functions (RAN domain or CN domain) through capability reporting. The capability reporting is RRC signaling or NAS (Non-Access Stratum) signaling.

[0842] As an example, the AI / ML model and related metadata are loaded by the first node from a network device or a remote server.

[0843] Optionally, the UE intelligent function 1804 is an MnS (Management Service) producer that provides data to the CN domain MnF (Management Function) 1801, and / or the RAN domain MnF 1802, and / or the cross-domain management system 1803 for management or analysis (as shown by double arrow 1807).

[0844] Optionally, the UE intelligent function 1804 is an MnS consumer that loads data from the CN domain MnF1801, and / or the RAN domain MnF1802, and / or the cross-domain management system 1803 for AI / ML-related management, such as managing data requests, AI / ML model activation, and / or AI / ML model training (as shown by double arrow 1807).

[0845] As an example, the AI / ML model is based on a neural network.

[0846] As an example, the AI / ML model is based on CNN (Conventional Neural Networks).

[0847] As an example, the AI / ML model is based on the Transformer architecture.

[0848] As an example, the appendix described in this application Figure 4 The first communication device 450 in the middle includes an attachment Figure 16 The reasoning function 1806 mentioned above.

[0849] As an example, the appendix described in this application Figure 10 The first processor 1003 in the process includes an appendix Figure 16 The reasoning function 1806 mentioned above.

[0850] As an example, the appendix described in this application Figure 12 The third module includes appendices. Figure 16 The reasoning function 1806 mentioned above.

[0851] As an example, the first node in this application includes an appendix. Figure 16 The reasoning function 1806 mentioned above.

[0852] As an example, the second node in this application includes an appendix. Figure 16 The MnF1802 mentioned above.

[0853] As an example, the second node in this application includes an appendix. Figure 16 The RAN field MnF1802 mentioned in the text.

[0854] As an example, the appendix described in this application Figure 2 The UE201 mentioned above includes an appendix. Figure 16 The reasoning function 1806 mentioned above.

[0855] As an example, the appendix described in this application Figure 2 The UE241 mentioned above includes an appendix. Figure 16 The reasoning function 1806 mentioned above.

[0856] As an example, the appendix described in this application Figure 2 The node 203 in the middle includes an appendix Figure 16 The MnF1801 mentioned above.

[0857] As an example, the appendix described in this application Figure 2 The node 203 in the middle includes an appendix Figure 16 The CN field MnF1801 mentioned in the document.

[0858] As an example, the appendix described in this application Figure 2 The node 203 in the middle includes an appendix Figure 16 The cross-domain management system 1803 mentioned above.

[0859] As an example, the appendix described in this application Figure 2 The node 211 in the middle includes attached Figure 16 The MnF1801 mentioned above.

[0860] As an example, the appendix described in this application Figure 2 The node 211 in the middle includes attached Figure 16 The CN field MnF1801 mentioned in the document.

[0861] As an example, the appendix described in this application Figure 2 The node 211 in the middle includes attached Figure 16 The cross-domain management system 1803 mentioned above.

[0862] As one embodiment, the data on the first logical channel is provided by an appender. Figure 16 The UE intelligent function 1804 mentioned above is generated.

[0863] As an example, the data on the first logical channel is for... Figure 16 The UE intelligent function 1804 mentioned above.

[0864] As an example, the data on the first logical channel is attached. Figure 16 The output of the reasoning function 1806 in the above.

[0865] As an example, the appendix Figure 16 The CN domain MnF1801, and / or the RAN domain MnF1802, and / or the cross-domain management system 1803 perform training and / or inference based on the data received on the first logical channel.

[0866] Example 17

[0867] Example 17 illustrates a flowchart based on artificial intelligence or machine learning according to an embodiment of this application; as attached. Figure 17 As shown. (Attached) Figure 17 This includes a third, fourth, fifth, sixth, and seventh operation. In Example 17, the third and fourth operations belong to the first stage, the fifth operation belongs to the second stage, the sixth operation belongs to the third stage, and the seventh operation belongs to the fourth stage. (See Appendix...) Figure 17 In the diagram, the lines with arrows indicate the sequence of processes.

[0868] As an example, the third operation includes AI / ML training, the fourth operation includes AI / ML testing, the fifth operation includes AI / ML emulation, the sixth operation includes AI / ML entity loading, and the seventh operation includes AI / ML inference.

[0869] As an example, the first stage includes a training phase, the second stage includes an emulation phase, the third stage includes a deployment phase, and the fourth stage includes an inference phase.

[0870] As an example, the first stage includes AI / ML model training.

[0871] As an example, the first stage includes AI / ML model training and AI / ML testing.

[0872] As an example, the AI / ML model training includes initial training and re-training of one or a group of AI / ML entities.

[0873] As an example, the training of the AI / ML model depends on training data.

[0874] As an example, the AI / ML model training includes AI / ML entity validation.

[0875] As an example, the AI / ML entity verification is used to evaluate the performance of the AI / ML entity.

[0876] As an example, the AI / ML entity verification relies on verification data.

[0877] As an example, if the AI / ML entity verification results do not meet expectations, the AI / ML model will be retrained.

[0878] As an example, the AI / ML testing includes testing the validated AI / ML entities to estimate the performance of the trained AI / ML model.

[0879] As an example, if the AI / ML test results meet expectations, the AI / ML entity proceeds to the next stage; otherwise, the AI / ML model will be retrained.

[0880] As an example, the AI / ML test relies on test data.

[0881] As one embodiment, the second stage includes AI / ML simulation, which performs AI / ML entity reasoning in a simulation environment.

[0882] As an example, the AI / ML simulation estimates the performance of AI / ML entity reasoning in a simulation environment before using AI / ML entities.

[0883] As one embodiment, the second stage is optional.

[0884] As an example, the third stage includes AI / ML entity loading, which is to obtain trained AI / ML entities to obtain the desired AI / ML inference capabilities.

[0885] As an example, the third stage is optional.

[0886] As an example, the third stage is no longer needed when the training and inference functions are co-located.

[0887] As an example, the fourth stage includes AI / ML inference.

[0888] Those skilled in the art will understand that all or part of the steps in the above methods can be implemented by a program instructing related hardware, and the program can be stored in a computer-readable storage medium, such as a read-only memory, hard disk, or optical disk. Optionally, all or part of the steps in the above embodiments can also be implemented using one or more integrated circuits. Accordingly, each module unit in the above embodiments can be implemented in hardware or in the form of software functional modules. This application is not limited to any specific combination of software and hardware. The user equipment, terminal, and UE in this application include, but are not limited to, drones, communication modules on drones, remote-controlled aircraft, aircraft, small aircraft, mobile phones, tablets, laptops, vehicle-mounted communication devices, wireless sensors, internet cards, IoT terminals, RFID terminals, NB-IoT terminals, MTC (Machine Type Communication) terminals, eMTC (enhanced MTC) terminals, data cards, internet cards, vehicle-mounted communication devices, low-cost mobile phones, low-cost tablets, and other wireless communication devices. The base station or system equipment in this application includes, but is not limited to, macrocell base stations, microcell base stations, home base stations, relay base stations, gNB (NR Node B), TRP (Transmitter Receiver Point), and other wireless communication equipment.

[0889] The above description is merely a preferred embodiment of this application and is not intended to limit the scope of protection of this application. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A first node used for wireless communication, characterized in that, include: A first receiver receives data on a plurality of RS resource groups, each of which is associated with a threshold. The first processor, whose performance is lower than or equal to a first threshold response, executes the first operation. Wherein, the input of the first model includes a measurement for a first RS resource group, the first threshold is a threshold associated with the first RS resource group; the first RS resource group is one of the plurality of RS resource groups; the threshold associated with at least two of the plurality of RS resource groups is different; The first operation includes sending a first message; or the first operation includes stopping the application of at least the first RS resource group to the first model.

2. The first node according to claim 1, characterized in that, include: The first receiver receives the first signaling; Wherein, the first signaling configures the associated threshold for each of the plurality of RS resource groups.

3. The first node according to claim 1 or 2, characterized in that, The first message indicates the first RS resource group.

4. The first node according to any one of claims 1 to 3, characterized in that, The performance of the first model depends on the error between the output of the first model and the actual measurement result, which is obtained after the measurement for the first RS resource group.

5. The first node according to any one of claims 1 to 4, characterized in that, The plurality of RS resource groups are associated with a plurality of cells, wherein the plurality of cells include at least one serving cell.

6. The first node according to any one of claims 1 to 5, characterized in that, The plurality of RS resource groups are associated with a plurality of functions respectively; wherein any one of the plurality of functions adopts the first model.

7. The first node according to any one of claims 1 to 6, characterized in that, The first operation includes receiving a second signaling; wherein the second signaling triggers the first message.

8. The first node according to any one of claims 1 to 7, characterized in that, The first event triggers the first message; wherein the first event depends on a second threshold, which is configurable.

9. A method used in a first node of wireless communication, characterized in that, include: Received on multiple RS resource groups, each of which is associated with a threshold; If the performance of the first model is lower than or equal to the first threshold response, the first operation is performed. Wherein, the input of the first model includes a measurement for a first RS resource group, the first threshold is a threshold associated with the first RS resource group; the first RS resource group is one of the plurality of RS resource groups; the threshold associated with at least two of the plurality of RS resource groups is different; The first operation includes sending a first message; or the first operation includes stopping the application of at least the first RS resource group to the first model.

10. A second node used for wireless communication, characterized in that, include: The second receiver receives the first message; Wherein, the sender of the first message receives the message on multiple RS resource groups, each of which is associated with a threshold; as a response to the performance of the first model being lower or not higher than the first threshold, the sender of the first message performs a first operation; the input to the first model includes a measurement for a first RS resource group, the first threshold being the threshold associated with the first RS resource group; the first RS resource group is one of the multiple RS resource groups; at least two of the multiple RS resource groups are associated with different thresholds; The first operation includes sending the first message; or the first operation includes sending the first message and stopping the application of at least the first RS resource group to the first model.

11. A method used in a second node for wireless communication, characterized in that, include: Receive the first message; Wherein, the sender of the first message receives the message on multiple RS resource groups, each of which is associated with a threshold; as a response to the performance of the first model being lower or not higher than the first threshold, the sender of the first message performs a first operation; the input to the first model includes a measurement for a first RS resource group, the first threshold being the threshold associated with the first RS resource group; the first RS resource group is one of the multiple RS resource groups; at least two of the multiple RS resource groups are associated with different thresholds; The first operation includes sending the first message; or the first operation includes sending the first message and stopping the application of at least the first RS resource group to the first model.

Images