Terminal, wireless communication method, and base station

Abstract

A terminal according to one aspect of the present disclosure comprises: a reception unit that receives information related to a trigger condition for fallback of life cycle management applied to a specific use case of an artificial intelligence (AI) model and a specific condition for recovery from the fallback; and a control unit that controls a trigger of a fallback operation on the basis of the trigger condition and controls an operation after the fallback operation on the basis of the specific condition.

Description

Terminal, wireless communication method and base station

[0001] The present disclosure relates to a terminal, a wireless communication method, and a base station in a next-generation mobile communication system.

[0002] Long Term Evolution (LTE) has been specified for the Universal Mobile Telecommunications System (UMTS) network with the aim of achieving higher data rates and lower latency (Non-Patent Document 1). Also, LTE-Advanced (3GPP Rel. 10-14) has been specified with the aim of achieving higher capacity and more advanced features than LTE (Third Generation Partnership Project (3GPP (registered trademark)) Release (Rel.) 8, 9).

[0003] Successor systems to LTE (e.g., 5th generation mobile communication system (5G), 5G+ (plus), 6th generation mobile communication system (6G), New Radio (NR), 3GPP Rel. 15 or later, etc.) are also being considered.

[0004] 3GPP TS 36.300 V8.12.0 “Evolved Universal Terrestrial Radio Access (E-UTRA) and Evolved Universal Terrestrial Radio Access Network (E-UTRAN); Overall description; Stage 2 (Release 8)”, April 2010

[0005] Regarding future wireless communication technologies, the use of artificial intelligence (AI) technologies such as machine learning (ML) for network / device control and management is being considered.

[0006] Use cases for utilizing AI models include spatial domain downlink (DL) beam prediction, temporal DL beam prediction, positioning, etc. Such beam prediction methods may be referred to as AI-based beam prediction (beam reporting), AI-based positioning, AI-based beam management (BM), etc. Temporal DL beam prediction may be referred to as time-domain Channel State Information (CSI) prediction, for example.

[0007] Additionally, depending on the use case of utilizing AI models, life cycle management (LCM), which may also be called performance monitoring, is being considered.

[0008] However, AI / ML models may suffer from generalization issues when the environment changes, which may significantly affect wireless link quality in AI / ML use cases such as CSI feedback / beam management.

[0009] If the AI / ML does not function well, poor CSI / beam provided by the AI / ML may cause radio link failure / beam failure.

[0010] In existing schemes, fallback operations are proposed as part of LCM based on either NW-side or UE-side performance monitoring, which consists of, for example, metric calculation, metric / information reporting, fallback decision, etc.

[0011] However, in the existing schemes, the trigger conditions for the fallback operation are not sufficient, and there may be cases where, for example, a radio link failure occurs before the fallback operation is triggered and enabled.

[0012] Furthermore, the recovery of the radio link quality may then be further affected by a bad / invalid / failed model.

[0013] In this way, if the fallback operation in LCM is not properly controlled, appropriate overhead reduction, highly accurate channel estimation, and highly efficient resource utilization may not be achieved, which may hinder improvements in communication throughput and communication quality.

[0014] Therefore, one of the objects of the present disclosure is to provide a terminal, a wireless communication method, and a base station that can achieve suitable overhead reduction / channel estimation / resource utilization.

[0015] A terminal according to one aspect of the present disclosure includes a receiving unit that receives information regarding trigger conditions for a lifecycle management fallback that is applied to a specific use case of an artificial intelligence (AI) model and specific conditions for recovery from the fallback, and a control unit that controls the triggering of a fallback operation based on the trigger conditions and controls an operation after the fallback operation based on the specific conditions.

[0016] According to one aspect of the present disclosure, it is possible to achieve favorable overhead reduction / channel estimation / resource utilization.

[0017] FIG. 1 is a diagram illustrating an example of a framework for managing AI models. FIG. 2 is a diagram illustrating an example of specifying an AI model. FIG. 3 is a diagram illustrating an example of CSI feedback using an encoder / decoder. FIG. 4 is a diagram illustrating an example of a lifecycle management framework for performance monitoring at a UE according to an embodiment. FIG. 5 is a diagram illustrating an example of a lifecycle management framework for performance monitoring at a BS according to an embodiment. FIGs. 6A and 6B are diagrams illustrating an example of AI-based beam reporting. FIG. 7 is a diagram illustrating an example of performance monitoring of CSI compression at the UE side. FIGs. 8A and 8B are diagrams illustrating an example of model evaluation. FIG. 9 is a diagram illustrating another example of model evaluation. FIG. 10 is a diagram illustrating a UE-base station (gNB) process for LTM in Rel. 18. FIG. 11 is a diagram illustrating an example of a fallback according to an embodiment of the present disclosure. FIG. 12 is a diagram illustrating an example of a schematic configuration of a wireless communication system according to an embodiment. FIG. 13 is a diagram illustrating an example of a base station configuration according to an embodiment. FIG. 14 is a diagram illustrating an example of a user terminal configuration according to an embodiment. FIG. 15 is a diagram illustrating an example of hardware configurations of a base station and a user terminal according to an embodiment. FIG. 16 is a diagram illustrating an example of a vehicle according to an embodiment.

[0018] (Channel State Information (CSI) Measurement / Reporting) This section describes CSI measurement / reporting in existing NR standards (e.g., Rel. 15-17 NR). The UE generates (also referred to as determining, calculating, estimating, measuring, etc.) CSI based on a reference signal (RS) (or a resource for the RS), and transmits (also referred to as reporting, feedback, etc.) the generated CSI to a network (e.g., a base station). The CSI may be transmitted to the base station, for example, using an uplink control channel (e.g., a Physical Uplink Control Channel (PUCCH)) or an uplink shared channel (e.g., a Physical Uplink Shared Channel (PUSCH)).

[0019] In the present disclosure, the CSI may be a Channel Quality Indicator (CQI), a Precoding Matrix Indicator (PMI), a CSI-RS Resource Indicator (CRI), a SS / PBCH Block Resource Indicator (SSBRI), a Layer Indicator (LI), a Rank Indicator (RI), L1-RSRP (Layer 1 Reference Signal Received Power), L1-RSRQ (Reference Signal Received Quality), L1-SINR (Signal to Interference plus Noise Ratio), L1-SNR (Signal to Noise Ratio), information on a channel matrix (or channel coefficient), information on a precoding matrix (or precoding coefficient), a Beam / Transmission Configuration Indication (BCI), a Precoding Matrix Indicator (PMI), a CSI-RS Resource Indicator (CRI), a SS / PBCH Block Resource Indicator (SSBRI), a Layer Indicator (LI), a Rank Indicator (RI), a Layer 1 Reference Signal Received Power (L1-RSRP), a Reference Signal Received Quality (L1-RSRQ), a Signal to Interference plus Noise Ratio (L1-SINR), a Signal to Noise Ratio (L1-SNR), information on a channel matrix (or channel coefficient), information on a precoding matrix (or precoding coefficient), a Beam / Transmission Configuration Indication (BCI), a Precoding Matrix Indicator (PMI), a Beam / Transmission Configuration Indicator (BCI), a Precoding Matrix Indicator (PMI ... Precoding Matrix Indicator (PMI), a Precoding Matrix Indicator (PMI), a Precoding Matrix Indicator (PMI), a Precoding Matrix In The information may include at least one of information regarding the TCI state / spatial relation.

[0020] The RS used to generate the CSI may be, for example, at least one of a Channel State Information Reference Signal (CSI-RS), a Synchronization Signal / Physical Broadcast Channel (SS / PBCH) block, a Synchronization Signal (SS), a Demodulation Reference Signal (DMRS), etc.

[0021] In the present disclosure, RS, CSI-RS, non-zero power (NZP) CSI-RS, zero power (ZP) CSI-RS, CSI interference measurement (CSI-IM), CSI-SSB, and SSB may be interchangeable. Furthermore, CSI-RS may include other reference signals.

[0022] The UE may receive configuration information regarding CSI reporting (which may be referred to as a CSI report configuration, report setting, etc.) and control the CSI reporting based on the configuration information. The report configuration information may be, for example, a Radio Resource Control (RRC) information element (IE) "CSI-ReportConfig."

[0023] The CSI reporting configuration may include at least one of the following information: - Information about the CSI resources used for CSI measurements (resource configuration ID, e.g., "CSI-ResourceConfigId"), - Information about one or more quantities (CSI parameters) of CSI to report (report quantity information, e.g., "reportQuantity"), - Report type information indicating the time domain behavior of the reporting configuration (e.g., "reportConfigType").

[0024] In the present disclosure, a CSI resource may be interchangeably referred to as a time instance, a CSI-RS opportunity / CSI-IM opportunity / SSB opportunity, a CSI-RS resource opportunity / opportunities, a CSI opportunity, an opportunity, a CSI-RS resource / CSI-IM resource / SSB resource, a time resource, a frequency resource, an antenna port (e.g., a CSI-RS port), etc. The time unit of a CSI resource may be a slot, a symbol, etc.

[0025] The information about the CSI resource may include information about the CSI resource for channel measurement, information about the CSI resource for interference measurement (NZP-CSI-RS resource), information about the CSI-IM resource for interference measurement, and the like.

[0026] The reporting quantity information may specify any one or a combination of the above CSI parameters (eg, CRI, RI, PMI, CQI, LI, L1-RSRP, etc.).

[0027] The report type information may indicate a periodic CSI (P-CSI) report, an aperiodic CSI (A-CSI) report, or a semi-persistent CSI (SP-CSI) report.

[0028] The UE performs CSI-RS / SSB / CSI-IM measurements based on the CSI resource configuration corresponding to the CSI reporting configuration (the CSI resource configuration associated with the CSI-ResourceConfigId), and derives the CSI to report based on the measurement results.

[0029] The CSI resource configuration (e.g., CSI-ResourceConfig information element) may include a csi-RS-ResourceSetList field indicating more specific CSI-RS / SSB resources, resource type information (e.g., "resourceType") indicating the time domain behavior of the resource configuration, etc.

[0030] The resource type information may indicate a P-CSI resource, an A-CSI resource, or an SP-CSI resource.

[0031] (Application of Artificial Intelligence (AI) Technology to Wireless Communications) With regard to future wireless communications technologies, the use of AI technology such as machine learning (ML) for network / device control and management is being considered.

[0032] For example, it is being considered that terminals (user terminals, user equipment (UE)) / base stations (BSs) will utilize AI technology to improve Channel State Information (CSI) feedback (e.g., reduced overhead, improved accuracy, prediction), improve beam management (e.g., improved accuracy, prediction in the time / space domain), and improve position measurement (e.g., improved position estimation / prediction).

[0033] Based on the input information, the AI ​​model may output at least one information such as an estimate, a prediction, a selected action, a classification, etc. The UE / BS may input channel state information, reference signal measurements, etc. to the AI ​​model and output highly accurate channel state information / measurements / beam selection / location, future channel state information / radio link quality, etc.

[0034] In the present disclosure, AI may be interpreted as an object (also called a subject, object, data, function, program, etc.) that has (performs) at least one of the following characteristics: - Estimation based on observed or collected information; - Selection based on observed or collected information; - Prediction based on observed or collected information.

[0035] In the present disclosure, estimation, prediction, and inference may be used interchangeably. Also, in the present disclosure, estimate, predict, and infer may be used interchangeably.

[0036] In the present disclosure, an object may be, for example, an apparatus, device, etc., such as a UE or a BS. Also, in the present disclosure, an object may correspond to a program / model / entity that operates in the apparatus.

[0037] Also, in the present disclosure, an AI model may be interpreted as an object that has (performs) at least one of the following characteristics: - Generates an estimate by feeding information; - Predicts an estimate by feeding information; - Discovers features by feeding information; - Selects an action by feeding information.

[0038] Additionally, in this disclosure, an AI model may refer to a data-driven algorithm that applies AI techniques to generate a set of outputs based on a set of inputs.

[0039] In addition, in the present disclosure, the terms AI model, model, ML model, predictive analytics, predictive analysis model, tool, autoencoder, encoder, decoder, neural network model, AI algorithm, scheme, etc. may be interchangeable. The AI ​​model may be derived using at least one of regression analysis (e.g., linear regression analysis, multiple regression analysis, logistic regression analysis), support vector machine, random forest, neural network, deep learning, etc.

[0040] In this disclosure, the term "autoencoder" may be interchangeably referred to as any autoencoder, such as a stacked autoencoder, a convolutional autoencoder, etc. The encoder / decoder of this disclosure may employ a model such as a Residual Network (ResNet), a DenseNet, or a RefineNet.

[0041] Furthermore, in the present disclosure, the terms encoder, encoding, encode / encoded, modification / alteration / control by an encoder, compressing, compress / compressed, generating, generate / generated, etc. may be read interchangeably.

[0042] In addition, in the present disclosure, decoder, decoding, decode / decoded, modification / alteration / control by decoder, decompressing, decompress / decompressed, reconstructing, reconstruct / reconstructed, etc. may be read interchangeably.

[0043] In the present disclosure, a layer (of an AI model) may be interchangeably read as a layer (such as an input layer or an intermediate layer) used in the AI ​​model. The layer in the present disclosure may correspond to at least one of an input layer, an intermediate layer, an output layer, a batch normalization layer, a convolutional layer, an activation layer, a dense layer, a normalization layer, a pooling layer, an attention layer, a dropout layer, a fully connected layer, etc.

[0044] In this disclosure, methods for training an AI model may include supervised learning, unsupervised learning, reinforcement learning, federated learning, etc. Supervised learning may refer to the process of training a model from inputs and corresponding labels. Unsupervised learning may refer to the process of training a model without labeled data. Reinforcement learning may refer to the process of training a model from inputs (i.e., states) and feedback signals (i.e., rewards) resulting from the model's outputs (i.e., actions) in an environment with which the model interacts.

[0045] In the present disclosure, terms such as generate, calculate, derive, etc. may be interchangeable. In the present disclosure, terms such as implement, operate, operate, execute, etc. may be interchangeable. In the present disclosure, terms such as train, learn, update, retrain, etc. may be interchangeable. In the present disclosure, terms such as infer, after-training, live use, actual use, etc. may be interchangeable. In the present disclosure, signal may be interchangeable with signal / channel.

[0046] FIG. 1 is a diagram illustrating an example of a framework for managing AI models. In this example, each stage related to an AI model is shown as a block. This example is also referred to as AI model life cycle management (LCM).

[0047] The data collection stage corresponds to a stage of collecting data for generating / updating an AI model. The data collection stage may include data organization (e.g., determining which data to transfer for model training / model inference), data transfer (e.g., transferring data to an entity (e.g., UE, gNB) that performs model training / model inference), etc.

[0048] Note that data collection may refer to a process in which data is collected by a network node, a management entity, or a UE for the purpose of AI model training / data analysis / inference. In this disclosure, the terms "process" and "procedure" may be interchangeable. Also, in this disclosure, collection may refer to obtaining a data set (e.g., usable as input / output) for AI model training / inference based on measurements (e.g., channel measurements, beam measurements, radio link quality measurements, position estimation, etc.).

[0049] In the present disclosure, offline field data may be data collected from the field (real world) and used for offline training of an AI model. Also, in the present disclosure, online field data may be data collected from the field (real world) and used for online training of an AI model.

[0050] In the model training stage, model training is performed based on the data (training data) transferred from the collection stage. This stage may include data preparation (e.g., performing data preprocessing, cleaning, formatting, transformation, etc.), model training / validation, model testing (e.g., verifying whether the trained model meets a performance threshold), model exchange (e.g., transferring the model for distributed learning), and model deployment / update (deploying / updating the model to the entity that will perform model inference).

[0051] It should be noted that AI model training may refer to a process for training an AI model in a data-driven manner and obtaining a trained AI model for inference.

[0052] AI model validation may also refer to a sub-process of training that evaluates the quality of an AI model using a dataset different from the dataset used to train the model, which helps select model parameters that generalize beyond the dataset used to train the model.

[0053] AI model testing may also refer to a sub-process of training for evaluating the performance of the final AI model using a dataset different from that used for model training / validation. Note that, unlike validation, testing does not necessarily require subsequent model tuning.

[0054] In the model inference stage, model inference is performed based on the data (inference data) transferred from the collection stage. This stage may include data preparation (e.g., performing data preprocessing, cleaning, formatting, transformation, etc.), model inference, model monitoring (e.g., monitoring the performance of model inference), model performance feedback (feeding back model performance to the entity training the model), and output (providing model output to the actor).

[0055] Additionally, AI model inference may refer to the process of using a trained AI model to produce a set of outputs from a set of inputs.

[0056] Also, a UE side model may refer to an AI model whose inference is performed entirely in the UE, and a network side model may refer to an AI model whose inference is performed entirely in the network (e.g., gNB).

[0057] Also, a one-sided model may refer to a UE-side model or a network-side model. A two-sided model may refer to a pair of AI models in which joint inference is performed. Here, joint inference may include AI inference in which the inference is performed jointly across the UE and the network, e.g., a first part of the inference may be performed first by the UE and the remaining part by the gNB (or vice versa).

[0058] In addition, AI model monitoring may refer to a process for monitoring the inference performance of an AI model, and may be interchangeably read as model performance monitoring, performance monitoring, etc.

[0059] Note that model registration may refer to assigning a version identifier to a model and making the model executable (registering) the model by compiling it into the specific hardware used in the inference stage. Also, model deployment may refer to distributing (or activating in) a runtime image (or an image of an execution environment) of a fully developed and tested model to (or enabling in) a target (e.g., UE / gNB) where inference will be performed.

[0060] An actor stage may include action triggers (e.g., deciding whether to trigger an action on another entity), feedback (e.g., feeding back information needed for training data / inference data / performance feedback), etc.

[0061] For example, training of a model for mobility optimization may be performed in, for example, Operation, Administration and Maintenance (Management) (OAM) / gNodeB (gNB) in a network (NW). In the former case, interoperability, large-capacity storage, operator manageability, and model flexibility (feature engineering, etc.) are advantageous. In the latter case, the latency of model updates and the need for data exchange for model deployment are advantageous. Inference of the above model may be performed in, for example, a gNB.

[0062] The entity that performs training / inference may vary depending on the use case (i.e., the function of the AI ​​model), which may include beam management, beam prediction, autoencoder (or information compression), CSI feedback, positioning, etc.

[0063] For example, for AI-assisted beam management based on measurement reports, the OAM / gNB may perform model training and the gNB may perform model inference.

[0064] For AI-assisted UE-assisted positioning, a Location Management Function (LMF) may perform model training and the LMF may perform model inference.

[0065] For CSI feedback / channel estimation using an autoencoder, the OAM / gNB / UE may perform model training and the gNB / UE may perform model inference (jointly).

[0066] For AI-assisted beam management or AI-assisted UE-based positioning based on beam measurements, the OAM / gNB / UE may perform model training and the UE may perform model inference.

[0067] Note that model activation may mean activating an AI model for a specific function, model deactivation may mean disabling an AI model for a specific function, and model switching may mean deactivating a currently active AI model for a specific function and activating a different AI model.

[0068] Model transfer may also refer to distributing an AI model over the air interface. This distribution may include distributing parameters of a model structure already known at the receiving end, or a new model with parameters, or both. This distribution may include a complete model or a partial model. Model download may refer to transferring a model from the network to the UE. Model upload may refer to transferring a model from the UE to the network.

[0069] 2 is a diagram showing an example of specifying an AI model. In this example, a UE and a NW (e.g., a base station (BS)) can recognize models #1 and #2 (although they do not need to fully understand the details of the models). The UE may report, for example, the capabilities of model #1 and model #2 to the NW, and the NW may instruct the UE on the AI ​​model to use.

[0070] (AI-Based CSI Feedback) As a use case of utilizing an AI model, CSI compression using a two-sided AI model is being considered. Such a CSI compression method may be called AI-based CSI feedback and may be realized using, for example, an autoencoder.

[0071] 3 is a diagram illustrating an example of CSI feedback using an encoder / decoder. The UE inputs CSI to an encoder and transmits information (CSI feedback information) including encoded bits output from an antenna. The BS inputs the received CSI feedback information bits to a corresponding decoder to obtain the output CSI.

[0072] The input CSI may include, for example, information on channel coefficients (elements of a channel matrix) or information on precoding coefficients (elements of a precoding matrix). In other words, the CSI may correspond to information on the channel state in the space-frequency domain. Note that the input may include information other than CSI.

[0073] The CSI output from the decoder may be reconstructed CSI corresponding to the input to the encoder, or may be CSI different from the input to the encoder (e.g., information on precoding coefficients if the input information is information on channel coefficients).

[0074] The encoder / decoder may also include pre-processing for the input and post-processing for the output.

[0075] The encoded bits are more compressed than the input information before encoding, and this is expected to reduce the communication overhead required for CSI feedback.

[0076] (Lifecycle Management Framework for Performance Monitoring) In the following, each step in the lifecycle management framework for performance monitoring at the UE / BS will be described for AI-based CSI feedback.

[0077] FIG. 4 illustrates an example lifecycle management framework for performance monitoring in a UE according to an embodiment.

[0078] In the performance monitoring step, the UE monitors the performance of the model and the fallback scheme (non-AI based CSI feedback).

[0079] In the step of model evaluation at the UE, the UE evaluates the performance of the monitored / reported model and fallback scheme (non-AI based CSI feedback).

[0080] In the performance reporting step, the UE reports the monitored performance to the NW.

[0081] In the step of model evaluation in the NW, the NW evaluates the performance of the reported model and fallback scheme.

[0082] In the model request step, the UE sends a request to the NW regarding which model should be applied or whether a fallback scheme should be applied.

[0083] In the model activation / deactivation step, the UE may be instructed which scheme (model) to activate. The UE may activate a model or a fallback scheme.

[0084] Note that some of the steps shown in the figure (for example, steps indicated by dashed lines) may be performed as needed.

[0085] FIG. 5 illustrates an example lifecycle management framework for performance monitoring in a BS according to one embodiment.

[0086] In the step of reporting for performance monitoring, the UE reports information for performance monitoring in the NW (BS).

[0087] In the step of performance monitoring in the NW, the NW monitors the performance of the model and fallback scheme (non-AI based CSI feedback).

[0088] In the step of model evaluation in the NW, the NW evaluates the performance of the model and the fallback scheme.

[0089] In the model activation / deactivation step, the UE may be instructed which scheme (model) to activate. The UE may activate a model or a fallback scheme.

[0090] Note that some of the steps shown in the figure (for example, steps indicated by dashed lines) may be performed as needed.

[0091] (AI-Based Beam Reporting) As a use case of utilizing an AI model, spatial domain downlink (DL) beam prediction or temporal DL beam prediction using a one-sided AI model in a UE or a NW is being considered. Such a beam prediction method may be called AI-based beam prediction (beam reporting), AI-based beam management (BM), etc.

[0092] 6A and 6B are diagrams illustrating an example of AI-based beam reporting. Fig. 6A shows spatial domain DL beam prediction. The UE may measure a spatially sparse (or thick) beam, input the measurement results, etc., into an AI model, and output a predicted result of the beam quality of a spatially dense (or thin) beam.

[0093] 6B shows temporal DL beam prediction. The UE may measure a time series of beams, input the measurement results into an AI model, and output a prediction result of the beam quality of a future beam.

[0094] Note that the spatial domain DL beam prediction may be referred to as BM case 1, and the temporal DL beam prediction may be referred to as BM case 2. Furthermore, the temporal DL beam prediction may be referred to as, for example, time domain CSI prediction.

[0095] Furthermore, the beams / RS associated with the output (prediction result) of the AI ​​model may be referred to as set A. The beams / RS associated with the input of the AI ​​model may be referred to as set B.

[0096] Candidates for input to the AI ​​model for BM Case 1 / 2 include L1-RSRP (Layer 1 Reference Signal Received Power), assistance information (e.g., beam shape information, UE position / direction information, transmit beam usage information), channel impulse response (CIR) information, and corresponding DL transmit / receive beam IDs.

[0097] Possible outputs of the AI ​​model for BM Case 1 include the IDs of the top K (K is an integer) transmit / receive beams, the predicted L1-RSRP of these beams, the probability that each beam will be in the top K, and the angles of these beams.

[0098] In addition to the candidate outputs of the AI ​​model in BM Case 1, the candidate outputs of the AI ​​model in BM Case 2 include predicted beam obstructions.

[0099] (Performance monitoring of CSI compression at the UE side)

[0100] 7 is a diagram illustrating an example of performance monitoring of CSI compression at the UE side, in which if an encoder is available at the UE, the UE may monitor the expected performance.

[0101] The performance (expected performance) monitored in Figure 7 may be at least one of the following: (1) expected communication quality calculated based on the output of the AI ​​model, e.g., expected CQI that meets a certain block error probability under a specific resource allocation assumption, (2) expected performance of the reconstructed CSI compared to the target CSI (e.g., expected noise variance).

[0102] The CQI in (1) may be, for example, at least one of a wideband CQI, an average of subband CQIs, a weighted average of subband CQIs, a maximum / minimum of subband CQIs, etc. Furthermore, the specific resource allocation may correspond to a frequency / time resource allocation for receiving a certain channel / signal (e.g., PDSCH, PDCCH, corresponding DMRS), and the type of resource allocation (e.g., the expected number of symbols, the number of resource blocks, etc.) may be specified in a standard. Furthermore, the certain block error probability may be, for example, at least one of 0.1, 0.00001, etc.

[0103] As shown in Figure 7, it is assumed that the CSI output from the decoder is the reconstructed CSI corresponding to the input to the encoder. Note that the decoder in the UE is only provided for performance monitoring, and the CSI feedback sent by the UE is the output of the encoder. The UE does not have a decoder corresponding to the encoder.

[0104] The UE performs channel measurement based on the CSI-RS transmitted from the BS and obtains a channel matrix H. The UE estimates its performance based on H.

[0105] When the input of the encoder included in the UE is a precoding matrix W, the UE may obtain W by performing a specific process (e.g., Singular Value Decomposition (SVD)) on H. The UE estimates performance based on W.

[0106] Furthermore, when the input of the encoder of the UE is a precoding matrix p-W to which preprocessing (e.g., Inverse Discrete Fourier Transform (IDFT) and sampling) has been applied, the UE may obtain p-W by performing the preprocessing on the above-mentioned W. The UE may estimate performance based on p-W or may estimate performance based on W.

[0107] In addition, the UE may transmit a performance report to the BS as needed.

[0108] The UE may receive information on the expected performance of the AI ​​model corresponding to the AI ​​model of the encoder from the vendor's data server or NW. The information may be included in the AI ​​model information.

[0109] In the present disclosure, the data server may be interchangeably referred to as a repository, an uploader, a library, a cloud server, simply a server, etc. Furthermore, the data server in the present disclosure may be provided by any platform such as GitHub (registered trademark), or may be operated by any company / organization.

[0110] In this example, the UE performs channel measurement based on the CSI-RS transmitted from the BS and obtains H / W / p-W corresponding to the target CSI. The UE also calculates (estimates) expected performance based on the target CSI and the above-mentioned expected performance information. If performance monitoring is the only task, the UE does not need to operate the encoder.

[0111] <Model Evaluation> The UE may evaluate the performance of the CSI feedback method (such as the above-mentioned model performance, performance with non-AI-based CSI feedback, etc.) and decide at least one of which performance to report, which method to request, which method to activate, etc.

[0112] The UE may check (evaluate) whether at least one of the following conditions is met for one or more monitored performances: Condition 1: The monitored performance of the active / registered / configured model or non-AI based CSI feedback is smaller / larger than the monitored performance of one of the deactive models or non-AI based CSI feedback (e.g., another codebook type); Condition 2: The monitored performance of the registered / configured model is larger / smaller than the monitored performance of one of the non-AI based CSI feedback; Condition 3: The monitored performance of one monitored model (e.g., active model) or non-AI based CSI feedback is smaller than a threshold; Condition 4: The monitored performance of one monitored model (e.g., deactive model) or non-AI based CSI feedback is larger than a threshold; Condition 5: The monitored performance of one monitored model or non-AI based CSI feedback has changed more than Y times since the last performance report (transmission); Condition 6: The monitored performance of a monitored model or non-AI-based CSI feedback falls below the threshold a certain number of times over a certain period of time.

[0113] In the present disclosure, the monitored performance may be interchangeably read as the performance obtained by adding an offset X (X is, for example, a real number) to the monitored performance. The offset X may be determined based on a factor other than the pure performance (reproducible performance) (for example, performance that is not monitored / does not need to be monitored). Introducing the offset enables model evaluation that comprehensively takes into account the other factor.

[0114] Here, the unmonitored / unnecessary performance may refer to at least one of the overhead of CSI feedback, reliability (of the model / calculated value), complexity of the model, power consumption for the calculation, etc.

[0115] The values ​​of X, Y, thresholds, etc. (or information about the values) may be specified in advance in a standard, may be determined based on UE capabilities, may be notified to the UE from the NW, or may be included in AI model information (may be determined based on a model). Information about the values ​​of X, Y, thresholds, etc. may be specified / notified for each model / non-AI-based CSI feedback, may be specified / notified for each group of model / non-AI-based CSI feedback, or may be specified / notified for AI-based CSI feedback or non-AI-based CSI feedback.

[0116] Which (or which combination of) conditions 1-6 the UE checks may be specified / notified for each model / non-AI-based CSI feedback, for each group of model / non-AI-based CSI feedback, or for AI-based CSI feedback or non-AI-based CSI feedback.

[0117] 8A and 8B are diagrams illustrating an example of model evaluation. In this example, the performance to be evaluated is the maximum subband CQI, and the UE compares the CQIs of AI models #1 and #2 with the CQI of the extended type II codebook as non-AI-based CSI feedback. The UE may, for example, activate the scheme with the highest performance among them.

[0118] 8A shows an example where no offset is applied to each monitored performance. In this example, the UE determines that model #1 has the greatest monitored performance.

[0119] 8B shows an example in which an offset X (where X<0) is applied to the monitored performance of Model #2, and an offset X' (where X'>0) is applied to the monitored performance of the extended Type II codebook. In this example, the UE determines that the monitored performance of the extended Type II codebook is the greatest.

[0120] Condition 6 will be explained more specifically. Condition 6 may include, for example, the following steps: - when a first counter counts that the monitored performance is less than a first value a first number of times or more, starting a timer; - while the timer is running, when a second counter counts that the monitored performance is greater than a second value a second number of times or more, stopping the timer; - while the timer is running, if the monitored performance is less than the first value, resetting the second counter; - if the monitored performance is greater than the first value, resetting the first counter; - when the timer expires, evaluating the performance of the monitored model as low.

[0121] The first value is a first threshold value. out ), or a first offset from a baseline value for a particular model / non-AI based CSI feedback. out ) lower.

[0122] The second value is a second threshold in ) or a second offset from the baseline value for a particular model / non-AI based CSI feedback. in ) may be larger.

[0123] Note that resetting the counter may mean setting the counter to a specific value (for example, 0).

[0124] Here, values ​​(or information on the values) such as the first / second threshold, baseline value, first / second offset, first / second counter, counter granularity, and timer time length may be specified in advance in a standard, may be determined based on UE capabilities, may be notified from the NW to the UE, or may be included in AI model information (may be determined based on a model). Information on these values ​​may be specified / notified for each model / non-AI-based CSI feedback, may be specified / notified for each group of model / non-AI-based CSI feedback, or may be specified / notified for AI-based CSI feedback or non-AI-based CSI feedback.

[0125] 9 is a diagram showing another example of model evaluation. In this example, the monitored performance is initially good, but when a first counter counts a first number of times that the monitored performance is less than a first value, a timer is started. Subsequently, while the timer is running, a second counter counts a number of times that the monitored performance is greater than a second value, but the second counter never reaches a second number of times or more, causing the timer to expire and the performance of the model to be evaluated as poor.

[0126] The UE may evaluate the performance of one or more CSI feedback methods and select (determine) the top K (K is an integer) performances for reporting / model request / model activate / model deactivate.

[0127] The K performances may all be selected from the performance of AI-based CSI feedback, may all be selected from the performance of non-AI-based CSI feedback, or may be selected from the performance of AI-based CSI feedback and non-AI-based CSI feedback.

[0128] In other words, the UE may evaluate the performance of one or more CSI feedback methods, determine the top K (K is an integer) performances from the AI-based CSI feedback performances, and determine the top K′ (K′ is an integer) performances from the non-AI-based CSI feedback performances.

[0129] Values ​​such as K, K' (or information regarding these values) may be specified in advance in a standard, may be determined based on UE capabilities, may be notified to the UE from the NW, or may be information linked to a model (may be determined based on the model).

[0130] In the present disclosure, the UE may derive a performance based on one or more monitored performances and one or more unmonitored / unnecessary performances. Furthermore, in the present disclosure, the monitored performances may be averaged / weighted over a certain period when evaluated / compared. Information regarding the period, averaging / weighting method, etc. may be specified in advance in a standard, may be determined based on UE capabilities, may be notified to the UE from the NW, or may be information linked to a model (may be determined based on a model).

[0131] <Performance Report> [Report Timing] The UE may transmit a performance report based on information notified from the NW. For example, the UE may transmit a performance report in an uplink resource that is scheduled periodically / semi-persistently / aperiodically based on the RRC / MAC CE / DCI. In this case, the performance report may be included in the UCI. In this case, the UE may determine the reporting period / offset based on the RRC / MAC CE / DCI.

[0132] The UE may determine a trigger for the performance report and transmit the performance report when the trigger is triggered. For example, the UE may transmit the performance report when the above-mentioned conditions (e.g., at least one of conditions 1-6) are satisfied. In this case, the performance report may be included in the MAC CE (because it can be transmitted once the PUSCH is scheduled).

[0133] The UE may send a performance report when a new model is activated / registered / configured, or when a timer based on the configured / specified parameters (e.g., the timer shown in condition 6) expires.

[0134] [Contents of the Report] The performance report may include information indicating one or more of the monitored performances described above. The number of pieces of information indicating the monitored performances included in the performance report may be determined based on the above-described K( / K').

[0135] The performance report may include information indicating the model / non-AI-based CSI feedback corresponding to the reported performance (e.g., model ID, registered model ID, CSI report setting ID).

[0136] The UE may determine the performance to be reported based on at least one of the following: - Performance evaluated under the above-mentioned conditions (e.g., at least one of conditions 1-6); - Performance selected based on the above-mentioned K( / K'); - Performance determined to be reported based on notification from the NW.

[0137] The notification may be, for example, an activation command for a model / non-AI-based CSI feedback, or a notification containing information indicating what to report / monitor. The UE may report the performance of the activated / monitored model.

[0138] For example, if a UE has models #1 and #2, and only model #2 is activated, the UE may report the capabilities of model #2, but may not report the capabilities of model #1.

[0139] <Model Request> [Timing of Transmission of Model Request] The timing of transmission of a model request may be determined based on the above description of the timing of the performance report, where the performance report is replaced with a model request.

[0140] [Contents of Model Request] The model request may include information indicating the model / non-AI-based CSI feedback to be applied (e.g., model ID, registered model ID, CSI report setting ID).

[0141] The model / non-AI-based CSI feedback to be applied may correspond to the model / non-AI-based CSI feedback corresponding to the reported performance described in the third embodiment (note that the performance does not have to be reported). The model / non-AI-based CSI feedback to be applied may also be referred to as recommended model / non-AI-based CSI feedback.

[0142] For example, consider a case where a UE has Models #1 and #2, and Type II and Extended Type II are available as non-AI-based CSI feedback, and only Model #2 is activated. The UE may evaluate the performance (e.g., CQI) of these models / non-AI-based CSI feedback and select the non-AI-based CSI feedback as the recommended model / non-AI-based CSI feedback. The UE may report a model request indicating the non-AI-based CSI feedback to the NW.

[0143] It should be noted that the UE may transmit the information included in the performance report and the information included in the model request simultaneously (for example, using one UCI / MAC CE).

[0144] <Model Activation / Deactivation> The UE may perform model activation / deactivation based on information notified from the NW. The information may be called a model activation / deactivation command and may be transmitted using RRC / MAC CE / DCI.

[0145] The model activation / deactivation command may include information indicating the model / non-AI-based CSI feedback to be activated / deactivated (e.g., model ID, registered model ID, CSI report setting ID).

[0146] The model activation / deactivation command may correspond to information indicating whether the model request described in the fourth embodiment has been accepted (for example, it may be referred to as a model response). When the UE transmits a model request and receives a corresponding model response, if the model response indicates that the model request has been accepted, the UE may activate the model indicated by the model request, or if not, deactivate the model.

[0147] The UE may determine the model / non-AI-based CSI feedback to activate / deactivate by itself. For example, the UE may determine the model / non-AI-based CSI feedback to activate / deactivate when the above-mentioned conditions (e.g., at least one of conditions 1-6) are satisfied. The model / non-AI-based CSI feedback to activate / deactivate may correspond to the model / non-AI-based CSI feedback corresponding to the above-mentioned reported performance (note that the performance may not be reported).

[0148] The UE may report information about the model to be activated or information indicating that the active model is changed to the NW. In this case, model request / performance reporting is unnecessary, and therefore, a reduction in communication overhead can be expected.

[0149] It should be noted that the UE may expect only one model / non-AI based CSI feedback for a certain function to be active, or may expect multiple model / non-AI based CSI feedbacks to be active.

[0150] When a certain model is activated (active), the UE may apply the model to the calculation of CSI feedback information, and when a certain model is activated, the UE may not perform CSI calculation / CSI reporting based on the configured non-AI-based CSI feedback that is not for performance monitoring purposes.

[0151] If all models are not active, the UE may apply a non-AI-based CSI feedback scheme to calculate the CSI feedback information. The UE may be configured with information about the non-AI-based CSI feedback scheme to be applied if all models are not active (e.g., using a CSI Report Configuration information element in RRC).

[0152] Also, if no models are active, the UE may not provide CSI feedback.

[0153] [Model Application Time (Active / Deactive Time)] The UE may activate / deactivate a model during the model application time.

[0154] The model application time may correspond to the period from a start time (start point) to an end time (end point).

[0155] The start time may correspond to at least one of the following: - the last symbol for receiving a model activation / deactivation command or X time units after the last symbol; - the last symbol for transmitting HARQ information (e.g., HARQ-ACK) corresponding to the model activation / deactivation command or X time units after the last symbol.

[0156] In the present disclosure, the unit time may be interpreted as at least one of a symbol, a slot, a subslot, a subframe, a second (millisecond), and the like.

[0157] The end time may be at least one of the following: Y units of time after the corresponding start time; Y units of time after the corresponding model activation / deactivation command is applied / received; or until a new model activation / deactivation command is applied / received.

[0158] The values ​​of X, Y, etc. (or information about the values) may be specified in advance in a standard, may be determined based on UE capabilities, may be notified from the NW to the UE, or may be information linked to the model (may be determined based on the model). The model activation / deactivation command may include information about the model application time (for example, information indicating X / Y).

[0159] If the UE receives a new model activation / deactivation command during the model application time, the UE may restart the start time of the model application time (or update the start time to the above-mentioned start time based on the new model activation / deactivation command). The start time of the model application time may be restarted in at least one of the following cases: - When the model information specified by the new model activation / deactivation command is the same as the currently applied model, - When the new model activation / deactivation command does not include information about the model information / application time.

[0160] The UE may use different methods for determining (or updating) the start time / end time for activation and deactivation.

[0161] The UE may also use different methods for determining (or updating) the start time / end time for model activation / deactivation and for fallback scheme activation / deactivation.

[0162] [Determining the AI ​​model to be monitored] The UE may determine the AI ​​model to be monitored based on the above-mentioned description of model activation / deactivation, in which activate (activation) / deactivate (deactivation) is replaced with monitor activate (activation) / deactivate (deactivation).

[0163] The UE may monitor an AI model only if at least one of the following is met for the configured AI model: - receiving activation of the model; - deciding to activate the model; - during the model application period of the model; - until the performance of the model is determined (in other words, the performance has not yet been determined); - until the performance of the model is reported (in other words, the performance has not yet been reported); - for a certain period after receiving a monitor activation command (monitor triggering signal) for the model.

[0164] The certain period may correspond to a period obtained by replacing the model application period with a model monitor application period. In other words, model activation / deactivation and model monitor activation / deactivation may be controlled separately or simultaneously.

[0165] (Life Cycle Management (LCM)) In future wireless communication systems (for example, Rel. 18 and later), the introduction of multiple LCMs is being considered.

[0166] The plurality of LCMs may include a functionality-based LCM and a model-ID-based LCM. The functionality-based LCM may be referred to as a functionality-based LCM, and the model-ID-based LCM may be referred to as a model-ID-based LCM.

[0167] In functionality-based LCM, a network (e.g., a base station / network node) may instruct an operation related to the functionality of an AI / ML (e.g., at least one of activation, deactivation, fallback operation, and switch). Here, the fallback operation may be an operation based on information (input information) used when applying the corresponding AI function, or an operation based on information (input information) used when applying the corresponding AI function and a non-AI function.

[0168] The UE may perform model-level LCM (eg, model switching and / or model selection) among the indicated functionality.

[0169] Among other things, the functionality may be transparent as to which models are activated / deactivated.

[0170] UE Capability information reporting may be used to signal supported functionality.

[0171] In model-ID-based LCM, a network (e.g., a base station / network node) may instruct an operation (e.g., at least one of activation, deactivation, fallback operation, and switch) related to an individual AI / ML model by a model ID.

[0172] The UE may perform model-level LCM (e.g., at least one of model switching and model selection) based on instructions from the NW.

[0173] A model may be defined in the NW by a model identifier (ID).

[0174] (Cases Related to Model Delivery / Transfer) Regarding model delivery / transfer, when a model is trained by a network, three cases are defined in existing scenarios.

[0175] [Case y] First, the NW trains a model and distributes the model to the UE outside the 3GPP network through offline engineering by multiple vendors.

[0176] The UE then reports its support for the delivered model (this step may be called model identification).

[0177] [Case z2] First, the NW trains a model through offline engineering of multiple vendors and saves it in a proprietary format, which may mean a format defined for each vendor.

[0178] The UE then reports the use of the stored model in the 3GPP network (this step may be called model identification).

[0179] A model transport is then carried out.

[0180] The UE then reports support for the transferred model (this step may be called model identification).

[0181] [Case z4] First, the UE reports the supported model structure (this step may be called model identification).

[0182] The NW then forwards the model parameters of the supported model structures.

[0183] The UE then reports support for the transferred model (this step may be called model identification).

[0184] (Functionality Identification) As a procedure for identifying functionality, for example, the UE may report a specific condition in the UE capability (capability information). In this case, the NW may configure the corresponding functionality based on the reported condition.

[0185] Here, functionality may represent features / feature groups (FGs) available in the AI / ML that are enabled by a certain configuration, such as a set of RRC parameters / LPP parameters. The configuration may be supported based on conditions indicated by the UE capabilities. Furthermore, functionality may represent units that the NW can control on the UE side in the operation (activation / deactivation / switching) of LCM.

[0186] The operation of the LCM based on functionality may be controlled based on the configuration of features / feature groups available in the AI / ML described above, where signaling (signaling for activation / deactivation / switching) to support the operation of the LCM based on the functionality is considered.

[0187] The UE may also report applicable functionality updates, e.g., a mechanism for updating the applicable model after identifying the model needs to be considered.

[0188] (Model Identification) A model identified by a model ID may be associated with, for example, settings / conditions / additional conditions (specific scenarios, sites, data sets, etc.). A model may represent a unit that the NW can control on the UE side in the operation (activation / deactivation / switching) of LCM.

[0189] The operation of the Model ID-based LCM may be controlled based on the identified model, where the model may be associated with specific settings / conditions regarding UE capabilities of features / feature groups available in the AI / ML, and additional conditions determined / identified between the UE side and the NW side.

[0190] In addition, it is assumed that the identification process and control unit are different between the functionality-based LCM and the model ID-based LCM. Here, it is being considered to share the activation / deactivation / switching procedures between the functionality-based LCM and the model ID-based LCM.

[0191] The following types of model identification procedures are also considered: Type A: Model information and model IDs are associated without signaling. The UE reports supported model IDs to the NW. That is, the mapping between model IDs and model information is identified to the NW and UE without signaling. The NW and UE identify the corresponding model information based on the received model ID. Type B1: Model information is reported from the UE to the NW via the air interface (signaling). Model identification is initiated by the UE, and the NW assists (takes responsibility for) the remaining steps of model identification. During model identification, a model ID may be assigned to the model. Type B2: Model information is reported from the NW to the UE via the air interface (signaling). Model identification is initiated by the NW, and the UE responds to the remaining steps of model identification. During model identification, a model ID may be assigned to the model.

[0192] (Meta Information) The UE may receive at least one of the following as meta information.

[0193] Note that the term "meta information" in this disclosure is merely an example, and meta information may refer to at least one of specific setting information, scenario information, environmental information, support information, and model information. In this disclosure, meta information, setting information, scenario information, environmental information, support information, and model information may be read interchangeably.

[0194] The meta information used by the UE / NW may include at least one of information regarding NW configuration / deployment, information regarding the environment, information regarding the AL / ML model on the NW side, and information regarding the model required by the NW.

[0195] The information regarding network settings / deployment may include, for example, information regarding antenna settings.

[0196] The information regarding the antenna configuration may indicate, for example, at least one of the following: horizontal / vertical antenna element / panel number, port number, antenna spacing, antenna position, panel position, and transceiver unit (TxRU) mapping.

[0197] The information regarding network configuration / deployment may include, for example, information regarding beam configuration.

[0198] The information about the beam setting may include, for example, at least one of the beam width, the number of beams, and the beam direction.

[0199] The information regarding network configuration / deployment may include, for example, information regarding TRP.

[0200] The information regarding the beam setting may include, for example, at least one of the altitude of the TRP and the relative positions of the multi-TRP.

[0201] The information about the environment may include, for example, information about the deployment scenario.

[0202] The information about the deployment scenario may indicate, for example, at least one of Urban Macro (UMa), Urban Micro (Umi), and Indoor Hotspot (InH).

[0203] The information about the environment may include, for example, information about indoors or outdoors.

[0204] The information about indoor or outdoor may indicate, for example, an indoor / outdoor probability.

[0205] The information about the environment may for example be information about objects around the UE / base station.

[0206] The information about the objects around the UE / base station may, for example, indicate the location of the objects around the UE / base station.

[0207] The information about the environment may include, for example, the scenario setting format (meta information) described below.

[0208] Use cases using AI models may be associated with a scenario-setting format that consists of long-term features.

[0209] Note that the term "long-term features" may be interchangeably read as "short-term / mid-term / long-term features" or simply "features." Furthermore, the term "scenario setting format" may be interchangeably read as "meta information," "meta information format," "scenario and configuration format," "scenario configuration format," "scenario format," "setting format," "use case format," "environment format," "meta format," etc. Furthermore, the term "format" may be interchangeably read as "type," "mode," "data," "setting," etc.

[0210] The above features may include one or more combinations of the following elements: - Scenario / model (Urban Macro (UMa), Urban Micro (Umi), indoor, outdoor, indoor hotspot (InH), etc.); - Frequency / frequency range; - Numerology (or subcarrier spacing); - Distribution / set of general channel parameters (e.g., inter-site distances (ISD), gNB height, delay spread, angle spread, Doppler spread, etc.) in a single scenario / model; - UE distribution; - UE speed; - UE orbit; - Number of transmit beams / receive beams; - UE rotation pattern; - gNB / UE antenna configuration (e.g., transmit and receive antenna vectors); - Number of cells / sectors; - Bandwidth; - UE payload; - Channel quality (e.g., RSRP, SINR); - Beam configuration ID; - Physical Cell ID (PCI). Global Cell ID (GCI) Absolute Radio Frequency Channel Number (ARFCN) Line Of Site (LOS) / Non-Line Of Site (NLOS) probability.

[0211] A UE may be expected to be configured / registered with a model whose associated scenario configuration format matches the UE's configuration / status.

[0212] The UE may also be expected to activate a model whose associated scenario configuration format matches the UE's configuration / status.

[0213] The correspondence between a use case and a scenario configuration format may be specified in a standard, and information on the correspondence may be notified to the UE. Also, the features included in the scenario configuration format corresponding to the use case may be specified in a standard, and information on the features may be notified to the UE.

[0214] The information on the AL / ML model on the NW side may include, for example, information on paired models available on the NW side.

[0215] The information about the paired model available on the NW side may indicate, for example, a paired decoder for CSI compression.

[0216] The information about the AL / ML model on the NW side may include, for example, information about pre-processing / post-processing available on the NW side.

[0217] The information regarding pre-processing / post-processing available on the NW side may include, for example, at least one of quantization / dequantization processing, DFT transformation, IDFT transformation, FFT transformation, and IFFT transformation.

[0218] UE Assistance Information The UE may report assistance information / metadata (meta-information) for the AI / ML model.

[0219] The assistance information may include, for example, at least one of the following information (user status): overheating assistance information; DRX parameter preference; priority regarding maximum aggregate bandwidth; preference for maximum number of MIMO layers.

[0220] The AI / ML model may be an AI / ML model that is registered / configured / compiled / activated in the UE.

[0221] The aiding information / metadata for the AI / ML model may be transmitted together with or instead of the beam information, which may be, for example, information about the UE's antennas / beams.

[0222] The supporting information / metadata of the AI / ML model may be at least one of the information described below.

[0223] The supporting information / metadata for the AI / ML model may be the ID of the AI / ML model.

[0224] The ID of the AI / ML model may be a global / local AI / ML model ID.

[0225] The supporting information / metadata for the AI / ML model may be information regarding the applicable bandwidth corresponding to the AI / ML model ID.

[0226] The bandwidth may be indicated as the applicable minimum / maximum bandwidth.

[0227] The information about the bandwidth may include, for example, information indicating a band indicator (e.g., "freqBandIndicatorNR"). The information indicating the band indicator may be represented by a specific number of bits (e.g., 10 bits).

[0228] The information about the bandwidth may include, for example, information indicating the bandwidth of the RS associated with the corresponding AI / ML model (eg, "supportedBandwidth").

[0229] The information indicating the bandwidth of the RS associated with the corresponding AI / ML model may indicate the frequency for each frequency range (for example, FR1 / FR2 (FR2-1 / FR2-2) / FR3 / FR4 / FR5).

[0230] The supporting information / metadata of the AI / ML model may be information about the applicable area corresponding to the AI / ML model ID.

[0231] The information regarding the applicable area corresponding to the AI / ML model may include at least one of the following information (list of information): Area ID. Cell global ID (in NR). Physical cell ID (Identifier) ​​(in NR). ARFCN (Absolute Radio Frequency Channel Number). Evolved Cell Global ID (ECGI).

[0232] The area ID may include at least one of the global ID of the NR cell, the physical cell ID of the NR, and the ARFCN.

[0233] The support information / metadata for the AI / ML model may be antenna setting / beam information corresponding to the AI / ML model ID.

[0234] (KPI) Common Key Performance Indicators (KPIs) are being considered for monitoring the performance of AI models.

[0235] Below is an initial list of common KPIs for evaluating the performance impact of AI / ML models: Performance, Intermediate KPIs, Link-level and system-level performance, Generalization performance, Over-the-air overhead, Assistance information overhead, Data collection overhead, Model delivery / transfer overhead, Other AI / ML model related signaling overhead, Inference complexity, Model inference computational complexity: floating point operations (FLOPs) (note that the s is lowercase), Pre- and post-processing computational complexity, Model complexity (number of parameters / data size (e.g., Mbyte), etc.), Training complexity, LCM related complexity. complexity) and storage overhead, and latency (e.g., inference latency). Note that the above-mentioned KPIs are merely examples, and other KPIs (e.g., KPIs related to model training, use-case-specific KPIs considered for a given use case, etc.) may be added to the list. Among the above-mentioned KPIs, KPIs related to performance may be called performance KPIs.

[0236] Functionality Related Information A UE may report the functionality it supports (or is supported).

[0237] A functionality may be associated with, include, or belong to a particular piece of information.

[0238] In this disclosure, "be associated with," "include," "belong to," and "correspond to" may be used interchangeably.

[0239] The specific information may be at least one of the following: The specific information and the functionality-related information may be interchangeable.

[0240] The particular information may be information regarding applicable conditions / settings.

[0241] The specific information may be, for example, applicable NW configurations. For example, the applicable parameters related to the NW configurations may be at least one of parameters of system information and parameters of assistance information.

[0242] The specific information may for example be applicable UE configuration, for example applicable parameters of the UE configuration may be higher layer (e.g. RRC) parameters.

[0243] The specific information may be, for example, information about an applicable scenario, such as information indicating at least one of NLOS / LOS, UE distribution, SINR, RSRP, bandwidth, frequency, indoor / outdoor, and deployment scenario (e.g., at least one of Urban Macro (UMa), Urban Micro (Umi), and Indoor Hotspot (InH)).

[0244] The specific information may be, for example, information about an applicable arrangement, which may be information indicating at least one of an applicable antenna configuration (e.g., at least one of the number of antenna elements / panels in the horizontal / vertical directions, the number of ports, the antenna spacing, the antenna position, the panel position, and the transceiver unit (TxRU) mapping), a beam configuration (e.g., at least one of the beam width, the number of beams, and the beam direction), and TRP information (e.g., at least one of the altitude of the TRP and the relative positions of multiple TRPs).

[0245] The specific information may be, for example, information about an applicable site, or may be information indicating at least one of an area ID, an NR cell global ID, an NR physical cell ID, a specific frequency (e.g., ARFCN), and an ECGI.

[0246] The specific information may be, for example, information about an applicable paired model. The paired model may be, for example, two (two or more) network-side models (combinations of models). The specific information may be, for example, information indicating a paired model for CSI compression.

[0247] The specific information may be, for example, information about an applicable time. The information may be, for example, information indicating an applicable period / interval / duration. The period may be, for example, indicated using a specific time unit (e.g., slot / symbol / subslot / millisecond / second).

[0248] The applicable conditions / settings may be predefined in the specification or may be determined / identified using a specific ID / token. For example, the applicable settings / scenarios / conditions may be defined as specific parameters (e.g., test parameters) or may be represented using a specific ID / token.

[0249] The performance of a functionality (e.g., prediction accuracy, location error) may be assumed to be better (e.g., higher) than a particular threshold under applicable conditions / settings, which may be, for example, the conditions / settings of a particular test.

[0250] The thresholds / performance requirements may be pre-specified or may be determined / identified using a specific ID / token.

[0251] If the threshold / performance requirements are based on a specific ID / token, each vendor / operator can define and utilize the desired thresholds / performance.

[0252] (Prediction-Based RLM) The UE may perform RLM based on predicted values / measurements. Note that in the present disclosure, performing RLM based on predicted values / measurements may be interchangeable with at least one of the following: assessing or estimating radio link quality based on predicted values / measurements, comparing a value at a predicted time instance / measured time instance with a threshold in the radio link quality evaluation, and detecting RLF from a synchronized state (or in-sync (IS)) / unsynchronized state (or out-of-sync (OOS)) based on the predicted values / measurements.

[0253] In the present disclosure, a time instance that is a time offset away from the measurement timing (e.g., the time of the first / last measurement RS resource, the indication period of the measurement RS resource) may be referred to as a time instance with which a predicted value is associated, a predicted time instance, a predicted time instance, etc. Also, in the present disclosure, a time instance associated with a measurement may be referred to as a time instance with which a measurement value is associated, a measured time instance, a measurement time instance, etc.

[0254] A UE may perform RLM based on predicted / measured values ​​if at least one of the following conditions is met: - the UE is configured / instructed to perform this RLM, - the UE reports capabilities (UE capabilities) related to the performance of this RLM, - a model identifier (ID) corresponding to the predicted RLM functionality (e.g., model ID of the model used for prediction) is activated for the UE, - the capability / accuracy of the prediction is higher / lower than a threshold, - before the relevant timer starts (e.g., only in cases where the OOS indication is based on prediction), - after the relevant timer starts (e.g., only in cases where the IS indication is based on prediction).

[0255] The parameters for determining the threshold may be predetermined in a standard, determined based on UE capabilities, configured / instructed to the UE by higher layer / physical layer signaling, or determined based on an associated model ID (e.g., model ID of a model used for prediction), prediction / model performance, etc. The associated timers will be described later, for example, in a fifth embodiment.

[0256] The UE may not be expected to perform RLM based on predicted values ​​only, in which case it may perform RLM based on actual measurements only, or on both predicted and actual measurements, where RLM based on predicted values ​​only may be assumed to be unreliable.

[0257] The UE may not be expected to perform RLM based only on actual measurements. In this case, it may perform RLM based only on predicted values ​​or both predicted and actual measurements. RLM based only on actual measurements may result in slower RLF detection.

[0258] The UE may not be expected to perform RLM based on both predicted and actual values, in which case it may perform RLM based on either predicted or actual values.

[0259] [Radio Link Quality Evaluation] First, radio link quality evaluation of the existing NR standard (for example, NR up to Rel. 17) will be described. In the existing NR standard, the physical layer in the UE evaluates the radio link quality evaluated over the previous time period once per indication period, based on a threshold (for example, Q ｏｕｔ , Q ｉｎ ) is evaluated against the threshold Q ｏｕｔ , Q ｉｎ The threshold Q may be determined based on a configuration for the UE, or if no configuration is made, a defined default value may be used. ｏｕｔ is defined as the level at which the downlink radio link is not reliably received, and corresponds to OOS BLER. ｉｎ is the downlink radio link ｏｕｔ is defined as the level at which the received signal is reliably received and is sufficiently high compared to the IS-BLER.

[0260] All the radio link qualities of the RS for RLM to be set are Q ｏｕｔ If the radio link quality of at least one of the configured RSs for RLM is Q or worse, Layer 1 of the UE sends an OOS indication to the upper layer. ｉｎ Better yet, Layer 1 of the UE sends an IS indication for this cell to higher layers.

[0261] Furthermore, Q ｏｕｔ , Q ｉｎ is derived based on hypothetical PDCCH transmission parameters, where the PDCCH transmission parameters include, for example, DCI format, aggregation level, bandwidth, subcarrier spacing, etc.

[0262] The radio link quality assessment associated with the predicted value and the radio link quality assessment associated with the measured value may be separate, and the physical layer in the UE may indicate separately the IS / OOS based on the radio link quality associated with the predicted value / predicted time instance and the IS / OOS based on the radio link quality associated with the measured value / measurement time instance.

[0263] The radio link quality assessment associated with the predicted value and the radio link quality assessment associated with the measured value may be joint (combined), and the physical layer in the UE may indicate IS / OOS based on the radio link quality associated with both the predicted value / predicted time instance and the measured value / measurement time instance.

[0264] The indication period corresponding to the radio link quality evaluation associated with the predicted value and the indication period corresponding to the radio link quality evaluation associated with the measured value may be the same or different. For example, the indication period corresponding to the radio link quality evaluation associated with the predicted value may be determined as the maximum period among the minimum period and the discontinuous reception (DRX) period of the RS resource monitored for calculating the predicted value.

[0265] The IS / OOS indication notified from the physical layer to a higher layer (for example, the RRC layer) may be notified together with information about the monitored RS resource (for example, the BWP index, etc.).

[0266] (Link Recovery Procedure Based on Prediction) In the present disclosure, radio link monitoring (RLM), link recovery procedure, and beam failure recovery (BFR) may be read interchangeably.

[0267] The UE may perform a link recovery procedure based on the predicted value / measurement. Note that in the present disclosure, performing a link recovery procedure based on the predicted value / measurement may be interchangeable with at least one of the following: assessing radio link quality based on the predicted value / measurement (e.g., determining whether to indicate a beam failure instance in the physical layer), comparing a value at a predicted time instance / measured time instance with a threshold in the radio link quality assessment, or triggering BFR based on a beam failure instance based on the predicted value / measurement.

[0268] In the present disclosure, a time instance that is a time offset away from the measurement timing (e.g., the time of the first / last measurement RS resource, the indication period of the measurement RS resource) may be referred to as a time instance with which a predicted value is associated, a predicted time instance, a predicted time instance, etc. Also, in the present disclosure, a time instance associated with a measurement may be referred to as a time instance with which a measurement value is associated, a measured time instance, a measurement time instance, etc.

[0269] A UE may perform a link recovery procedure based on predicted / measured values ​​if at least one of the following conditions is met: - the UE is configured / instructed to perform the link recovery procedure; - the UE reports capabilities (UE capabilities) related to the performance of the link recovery procedure; - a model identifier (ID) corresponding to the predicted link recovery procedure functionality (e.g., model ID of the model used for the prediction) is activated for the UE; - the capability / accuracy of the prediction is higher / lower than a threshold; - before the relevant timer starts; - after the relevant timer starts.

[0270] The parameters for determining the threshold may be predetermined in a standard, determined based on UE capabilities, configured / instructed to the UE by higher layer / physical layer signaling, or determined based on an associated model ID (e.g., model ID of a model used for prediction), prediction / model performance, etc. The associated timers will be described later, for example, in a sixth embodiment.

[0271] The UE may not be expected to perform link recovery procedures based solely on predicted values. In this case, it may perform link recovery procedures based solely on actual measurements, or both predicted and actual measurements. Link recovery procedures based solely on predicted values ​​may be assumed to be unreliable.

[0272] The UE may not be expected to perform link recovery procedures based solely on actual measurements. In this case, it may perform link recovery procedures based solely on predicted values ​​or both predicted and actual measurements. Link recovery procedures based solely on actual measurements may result in delayed BFR trigger / success.

[0273] The UE may not be expected to perform link recovery procedures based on both predicted and actual measurements, but may instead perform link recovery procedures based on either predicted or actual measurements.

[0274] [Beam Failure Instance] First, we will explain the beam failure instance determination of the existing NR standard (for example, NR up to Rel. 17). In the existing NR standard, the physical layer in the UE checks the beam failure instance once per indication period for the previous evaluation period (T Evaluate_BFD_XXX (XXX is SSB, CSI-RS, etc.)) is evaluated over a threshold (e.g., Q ｏｕｔ＿ＬＲ ) is evaluated against the threshold Q ｏｕｔ＿ＬＲ The threshold Q may be determined based on a configuration for the UE, or if no configuration is made, a defined default value may be used. ｏｕｔ＿ＬＲ is defined as the level at which the downlink radio link for a given resource configuration in the set of BFD-RSs is not reliably received, eg, corresponding to a 10% BLER of a hypothetical PDCCH transmission.

[0275] All wireless link qualities of the configured BFD-RS are Q ｏｕｔ＿ＬＲ In the worse case, Layer 1 of the UE sends a beam failure instance indication to higher layers.

[0276] Furthermore, Q ｏｕｔ＿ＬＲ is derived based on hypothetical PDCCH transmission parameters, where the PDCCH transmission parameters include, for example, DCI format, aggregation level, bandwidth, subcarrier spacing, etc.

[0277] (L1L2-triggered mobility (LTM) in Rel. 18) The UE establishes an RRC connection to the current serving cell (base station with PCI #1) and sends an L3 measurement report. Based on the L3 measurement report, the serving cell decides to perform LTM and prepares for LTM with one or more candidate cells. The UE and the serving cell then perform RRC reconfiguration. Note that the one or more candidate cells may include the target cell (base station with PCI #3).

[0278] In the present disclosure, cell switch and cell switching may be read interchangeably.

[0279] The UE, serving cell, target cell, and candidate cell perform DL synchronization. The UE performs and reports L1 (e.g., L1-RSRP / SINR) measurements of the serving / candidate / target cell. The UE, serving cell, target cell, and candidate cell perform UL synchronization.

[0280] The serving cell determines to switch the serving cell to the target cell (PCI #3) based on the L1 measurement report and sends a cell switch command to the UE. After receiving the cell switch command, the UE starts PDCCH monitoring for the target cell.

[0281] In the case of RACH-based LTM, the UE performs a RACH procedure to the target cell. In the case of RACH-less LTM, the UE sends an RRC reconfiguration complete message and transmits the first data to the target cell. This first data transmission is based on a dynamic grant or a configuration grant associated with the beam of the target cell. The target cell sends an ACK for this transmission to the UE.

[0282] Figure 10 is a diagram showing processing between a UE and a base station (gNB) in LTM in Rel. 18. Below, the processing of each step in Figure 10 will be described in detail.

[0283] 1: The UE sends a measurement report message to the gNB, which determines the LTM configuration and starts preparing one or more candidate cells.

[0284] 2: The gNB sends an RRC reconfiguration message to the UE including LTM candidate cell configurations for one or more candidate cells.

[0285] 3: The UE saves its LTM candidate cell configuration and sends an RRC reconfiguration complete message to the gNB.

[0286] 4a: The UE performs DL synchronization with one or more candidate cells before receiving a cell switch command. DL synchronization for candidate cells before the cell switch command may be supported based on at least SSB.

[0287] 4b: If requested by the network, the UE performs early TA acquisition with one or more candidate cells before receiving a cell switch command. This is triggered by a PDCCH order from the source cell via a CFRA. The UE then transmits a preamble to the indicated candidate cells. To minimize data interruption in the source cell due to Contention Free Random Access (CFRA) to the candidate cells, the UE does not receive a RAR intended for TA value acquisition. The TA value of the candidate cell is indicated in the cell switch command. The UE does not maintain a TA timer for the candidate cells and ensures the validity of the TA based on the network implementation.

[0288] 5: The UE performs L1 measurements configured on the candidate cells and sends an L1 measurement report to the gNB. L1 measurements are performed as long as the RRC reconfiguration in step 2 is applied.

[0289] 6: The gNB decides to perform a cell switch to the target cell and sends a MAC CE (Cell Switch Command) to trigger the cell switch. The MAC CE includes a candidate configuration for the index of the target cell. The UE switches to the target cell and applies the configuration indicated by the candidate configuration index.

[0290] 7: If the UE does not have a valid TA for the target cell, it performs a random access procedure to the target cell.

[0291] 8: The UE completes the LTM cell switch procedure by sending an RRC reconfiguration complete message. The UE performs the RA procedure in step 7, and considers the LTM execution to be completed successfully if the random access procedure is successfully completed. In RACH-less LTM, the UE considers the LTM execution to be completed successfully if the network determines that the first UL data has been successfully received. The UE determines the successful reception of the first UL data by receiving a PDCCH specifying the UE's C-RNTI in the target cell that schedules the next new transmission of the first UL data.

[0292] Note that the RRC reconfiguration complete message may always be sent in each LTM run, and steps 4-8 may be performed multiple times for subsequent LTMs using the candidate cell configuration provided in step 2.

[0293] (Analysis) However, AI / ML models may suffer from generalization issues when the environment changes, which may significantly affect wireless link quality in AI / ML use cases such as CSI feedback / beam management.

[0294] If the AI / ML does not function well, poor CSI / beam provided by the AI / ML may cause radio link failure / beam failure.

[0295] In existing schemes, fallback operations are proposed as part of LCM based on either NW-side or UE-side performance monitoring, which consists of, for example, metric calculation, metric / information reporting, fallback decision, etc.

[0296] However, in the existing schemes, the trigger conditions for the fallback operation are not sufficient, and there may be cases where, for example, a radio link failure occurs before the fallback operation is triggered and enabled.

[0297] Furthermore, the recovery of the radio link quality may then be further affected by a bad / invalid / failed model.

[0298] In this way, if the fallback operation in LCM is not properly controlled, appropriate overhead reduction, highly accurate channel estimation, and highly efficient resource utilization may not be achieved, which may hinder improvements in communication throughput and communication quality.

[0299] Therefore, the present inventors have devised a wireless communication method according to the present invention to solve these problems.

[0300] (Various Replacements, etc.) Hereinafter, embodiments according to the present disclosure will be described in detail with reference to the drawings. Wireless communication methods according to the embodiments may be applied independently or in combination.

[0301] In the present disclosure, "A / B" and "at least one of A and B" may be interpreted interchangeably. Also, in the present disclosure, "A / B / C" may mean "at least one of A, B, and C."

[0302] In the present disclosure, terms such as notify, activate, deactivate, indicate (or indicate), select, configure, update, and determine may be read interchangeably. In the present disclosure, terms such as support, control, controllable, operate, and operate may be read interchangeably.

[0303] In the present disclosure, Radio Resource Control (RRC), RRC parameters, RRC messages, higher layer parameters, fields, information elements (IEs), settings, etc. may be interchangeable. In the present disclosure, Medium Access Control (MAC) control elements (CEs), update commands, activation / deactivation commands, etc. may be interchangeable.

[0304] In the present disclosure, the higher layer signaling may be, for example, any one of Radio Resource Control (RRC) signaling, Medium Access Control (MAC) signaling, broadcast information, other messages (e.g., messages from the core network such as positioning protocol (e.g., NR Positioning Protocol A (NRPPa) / LTE Positioning Protocol (LPP)) messages), or a combination thereof.

[0305] In the present disclosure, MAC signaling may use, for example, a MAC Control Element (MAC CE), a MAC Protocol Data Unit (PDU), etc. Broadcast information may be, for example, a Master Information Block (MIB), a System Information Block (SIB), Remaining Minimum System Information (RMSI), Other System Information (OSI), etc.

[0306] In the present disclosure, physical layer signaling may be, for example, Downlink Control Information (DCI), Uplink Control Information (UCI), and the like.

[0307] In the present disclosure, the terms index, identifier (ID), indicator, resource ID, etc. may be interchangeable. In the present disclosure, the terms sequence, list, set, group, cluster, subset, etc. may be interchangeable.

[0308] In the present disclosure, the terms panel, UE panel, panel group, beam, beam group, precoder, Uplink (UL) transmitting entity, Transmission / Reception Point (TRP), base station, Spatial Relation Information (SRI), spatial relation, SRS Resource Indicator (SRI), Control Resource Set (CORESET), Physical Downlink Shared Channel (PDSCH), Codeword (CW), Transport Block (TB), Reference Signal (RS), antenna port (e.g., Demodulation Reference Signal (DMRS) port), antenna port group (e.g., DMRS port group), group (e.g., spatial relation group, Code Division Multiplexing (CDM) group, reference signal group, CORESET group, Physical Uplink Control Channel (PUCCH) group, PUCCH resource group), resource (e.g., reference signal resource, SRS resource), resource set (e.g., reference signal resource set), CORESET pool, downlink Transmission Configuration Indication state (TCI state) (DL TCI state), uplink TCI state (UL TCI state), unified TCI state, common TCI state, Quasi-Co-Location (QCL), QCL assumption, etc. may be read as interchangeable.

[0309] In the present disclosure, CSI-RS, non-zero power (NZP) CSI-RS, zero power (ZP) CSI-RS, and CSI interference measurement (CSI-IM) may be interchangeable. Furthermore, CSI-RS may include other reference signals.

[0310] In this disclosure, a measured / reported RS may refer to an RS that is measured / reported for a CSI report.

[0311] In the present disclosure, timing, time, duration, slot, subslot, symbol, subframe, etc. may be read interchangeably.

[0312] In the present disclosure, the terms direction, axis, dimension, domain, polarization, polarization component, etc. may be read interchangeably.

[0313] In the present disclosure, estimation, prediction, and inference may be used interchangeably. Also, in the present disclosure, estimate, predict, and infer may be used interchangeably.

[0314] In the present disclosure, the terms autoencoder, encoder, decoder, etc. may be replaced with at least one of a model, an ML model, a neural network model, an AI model, an AI algorithm, etc. Furthermore, the term autoencoder may be replaced with any autoencoder, such as a stacked autoencoder or a convolutional autoencoder. The encoder / decoder of the present disclosure may employ a model such as a Residual Network (ResNet), a DenseNet, or a RefineNet.

[0315] In the present disclosure, the terms bit, bit string, bit sequence, sequence, value, information, value obtained from a bit, information obtained from a bit, etc. may be read interchangeably.

[0316] In the present disclosure, a layer (for an encoder) may be interchangeably read as a layer (such as an input layer or an intermediate layer) used in an AI model. The layer in the present disclosure may correspond to at least one of an input layer, an intermediate layer, an output layer, a batch normalization layer, a convolutional layer, an activation layer, a dense layer, a normalization layer, a pooling layer, an attention layer, a dropout layer, a fully connected layer, etc.

[0317] In the present disclosure, RSRP may be interchangeably read as any parameter related to received power / received quality, etc. (e.g., RSRQ, SINR, CSI), etc.

[0318] In the present disclosure, the RS may be, for example, a CSI-RS, an SS / PBCH block (SS block (SSB)), etc. Also, the RS index may be a CSI-RS Resource Indicator (CRI), an SS / PBCH Block Indicator (SSBRI), etc.

[0319] In the present disclosure, channel measurement / estimation may be performed using at least one of, for example, a Channel State Information Reference Signal (CSI-RS), a Synchronization Signal (SS), a Synchronization Signal / Physical Broadcast Channel (SS / PBCH) block, a Demodulation Reference Signal (DMRS), a Sounding Reference Signal (SRS), and the like.

[0320] In the present disclosure, the terms "receive beam assumption," "number of receive beams," "index of receive beam," "receive beam selection," "receive beam setting," and "receive beam instruction" may be interchangeable. In the present disclosure, the terms "receive beam," "transmit beam," "DL receive beam," "DL transmit beam," and "pair of transmit beam and receive beam" may be interchangeable. In the present disclosure, the terms "transmit / receive beam" may be interchangeable with the terms "transmit / receive beam for beam prediction" and "transmit / receive beam for CSI measurement / reporting for beam prediction."

[0321] In this disclosure, functionality may refer to the use of a model or the physical meaning of the model's input / output. Multiple models may have the same functionality. Monitoring (checking performance), activation, deactivation, switching, fallback, and updating may be instructed (controlled) based on the functionality (e.g., for each function).

[0322] A model ID may also refer to an identifier for a model (or a set of models). Multiple models may be assigned the same model ID in an actual deployment. In this case, these models may actually be different models (e.g., have different numbers of layers) but may be treated as the same model.

[0323] In this disclosure, the use cases may include AI / ML for at least one of enhanced CSI feedback, beam management, and enhanced positioning, and may also include other new use cases for AI / ML.

[0324] In the present disclosure, the model ID may be interchangeably read as a meta-information (or a set of meta-information) ID. The meta-information (or meta-information ID) may be associated with information about the applicability of the model / functionality, the environment, the UE / gNB settings, etc.

[0325] In the present disclosure, functionality may be simply read as "function."

[0326] In the present disclosure, functionality, function, functionality ID, model, and model ID may be read interchangeably.

[0327] In the present disclosure, update, report, and send may be read interchangeably.

[0328] In the present disclosure, meta information, assistance information, sensing information, KPI, performance KPI, UE status, and status may be read interchangeably.

[0329] In the present disclosure, monitor and evaluation may be read interchangeably.

[0330] In the present disclosure, the terms "decision," "judgement," and "application of a particular action" may be read interchangeably.

[0331] In the present disclosure, the terms entity, specific entity, UE, NW, gNB, and LMF may be read interchangeably.

[0332] In the present disclosure, NW, LMF, gNB, and BS may be read interchangeably.

[0333] In the present disclosure, the UE-side model and the UE may be read interchangeably.

[0334] In the present disclosure, the model, UE-side model, logical model, and physical model may be interchangeable.

[0335] In this disclosure, a model / functionality may refer to a data-driven algorithm that applies AI / ML techniques to generate a set of outputs based on a set of inputs.

[0336] In the present disclosure, the terms performance index and monitoring index may be read interchangeably.

[0337] In the present disclosure, the terms association, correspondence, and mapping may be read interchangeably.

[0338] In the present disclosure, monitor results, monitored results, post-monitoring results, and monitoring results may be read interchangeably.

[0339] In the present disclosure, the monitoring results may include inference results, and information regarding at least one of a performance index, and the content of an event occurrence based on the performance index / whether or not an event has occurred.

[0340] In the present disclosure, for UE-side performance monitoring of models / functionality, the UE may report the following information as monitoring results: Performance indicators corresponding to the monitored models / functionality; Occurrence of an event in the calculation of the performance indicators corresponding to the monitored models / functionality (e.g., the value of a positive indicator is greater than / less than a threshold for a certain duration).

[0341] In this disclosure, AI / ML-based CSI reporting may refer to a CSI reporting associated with a model ID or specific functionality, where the specific functionality may be, for example, at least one of predicted CSI, compressed CSI, advanced CSI, and type[x] CSI.

[0342] In the present disclosure, AI / ML-based CSI reporting and CSI reporting may be read interchangeably.

[0343] In the present disclosure, a proxy model may refer to a model that is used only for performance monitoring and has no other uses other than performance monitoring, or may refer to a model that estimates side information (CQI / RI, etc.) of the recovered CSI.

[0344] In this disclosure, AI / ML functionality, AI / ML CSI functionality may refer to functionality commanded by the NW or reported by the UE, such as at least one of predicted CSI, compressed CSI, advanced CSI, and type [x] CSI.

[0345] In the present disclosure, AI / ML functionality, AI / ML CSI functionality, and functionality may be read interchangeably.

[0346] In this disclosure, an AI / ML model, an AI / ML CSI model may refer to a model / entity that is identified by a specific ID and performs a specific function (functionality).

[0347] In the present disclosure, report quantity, information regarding the report quantity, and report quantity information may be read interchangeably.

[0348] In the present disclosure, reporting settings, CSI reporting settings, settings related to performance monitoring of CSI reports, monitor reporting settings, and reporting settings for monitoring may be read interchangeably.

[0349] In the present disclosure, reports, CSI reports, measurement result reports, monitoring reports, and monitor result reports may be read interchangeably.

[0350] In the present disclosure, CSI-RS and PDSCH / DMRS may be read as interchangeable.

[0351] In this disclosure, type X monitoring (results) may refer to precoded RS resource-based monitoring (results), and type Y monitoring (results) may refer to PDSCH / DMRS-based monitoring (results).

[0352] In this disclosure, RS Type B may refer to an RS (signal / channel) associated with a measurement report / monitoring result report, and RS Type A may refer to an RS (signal / channel) associated with a model ID or a CSI report associated with a specific functionality / feature.

[0353] In the present disclosure, the terms performance metric, metric for monitoring reporting, and KPI may be read interchangeably.

[0354] (Wireless communication method) In view of the above-described problems, an embodiment of the present disclosure describes a proactive operation (fallback / model switching) of a UE based on a specific condition. Fig. 11 is a diagram illustrating an example of a fallback according to an embodiment of the present disclosure.

[0355] The specific condition may mean a condition / additional condition that is set / indicated, and the condition / additional condition may be a radio link failure / beam failure detection indication triggered by the UE, or an LTM indication triggered by a MAC CE (cell switch command).

[0356] The UE can achieve reuse (recovery) of a particular AI / ML model from proactive fallback if conditions / additional conditions are met.

[0357] The UE may control the execution of proactive action (fallback / model switching) for example according to at least one of the following options:

[0358] (Option 1) Occurrence / trigger of a specific event. The specific event can be, for example, an RLF / BF detected by the UE, or reception of a cell switching command for LTM.

[0359] (Option 2) - Receiving instructions / instruction information regarding conditions / additional conditions on the network side.

[0360] (Option 3) - Awareness on the UE side that conditions / additional conditions have changed.

[0361] In the present disclosure, proactive action, fallback action, and proactive fallback action may be read interchangeably.

[0362] In the present disclosure, examples of fallback operations include, but are not limited to, deactivating functionality / models, switching models, and applying existing operations (schemes of existing specifications). The existing operations may mean operations that do not use AI / ML models.

[0363] In the present disclosure, examples of post-fallback actions include, but are not limited to, reporting, reactivation, etc. These post-fallback actions may be treated as part of the fallback action (may be interpreted as a replacement).

[0364] Specifically, the embodiments of the present disclosure can be broadly categorized as follows: First embodiment: Fallback operation in LCM. Second embodiment: UE operation after fallback. Third embodiment: Example of CSI feedback. Fourth embodiment: Example of beam prediction. Each embodiment will be described below based on these.

[0365] In the present disclosure, each embodiment / option may be applied alone or in combination with other embodiments / options.

[0366] First Embodiment The first embodiment relates to a fallback operation in an LCM.

[0367] [Aspect 1-1] Aspect 1-1 relates to the trigger conditions for the fallback operation.

[0368] The UE may deactivate one / multiple / all configured functionalities / models based on specific conditions. The specific conditions may be at least one of the following conditions / additional conditions / events. The specific conditions may be set / indicated by higher layer signaling / physical layer signaling. The specific conditions (conditions 1 to 9) are listed below.

[0369] <Condition 1> When the UE performs performance monitoring and determines that the performance of the corresponding functionality / model is lower than a specified / configured threshold. Alternatively, when the UE monitors that the performance of the corresponding functionality / model is lower than a threshold. Condition 1 may be the same condition as in the existing scheme.

[0370] <Condition 2> - The UE receives Nx consecutive out-of-sync indications from lower layers or beam failure instances, where Nx may be a number configured / indicated / specified for the UE.

[0371] <Condition 3> When the UE starts a predetermined timer to deal with a problem due to a radio link failure, for example, when the UE starts a timer (T310) after receiving N310 consecutive out-of-sync indications (see FIG. 11).

[0372] <Modification of Condition 3> When the UE starts a predetermined timer to handle a problem with a radio link failure using an activated functionality / model, the UE may extend / reduce the timer period based on the result of performance monitoring for the functionality / model. Alternatively, the UE may increase / decrease the number of indications for triggering the timer (Nx as described above).

[0373] For example, if the UE monitors that the performance of the corresponding functionality / model is lower than a threshold, the UE may reduce the duration of the timer and perform a fallback operation.

[0374] Here, if the radio link failure is still not resolved (recovered) even after the UE performs the fallback operation, the UE may declare a radio link failure (RLF) after the timer expires and start the radio link failure recovery procedure early.

[0375] Note that before declaring RLF, the UE detects a problem for radio link failure and starts a timer, and declares RLF when the timer expires. The fallback operation may be triggered after detecting radio link failure but before declaring RLF.

[0376] <Condition 4> - If the UE receives a beam failure instance within the configured / instructed time window, in this case the number of beam failure instances may be less than the normal number (beamFailureInstanceMaxCount) to trigger the fallback operation earlier.

[0377] <Condition 5> - When the UE declares a radio link failure (RLF) / beam failure (BF).

[0378] <Condition 6> - When the UE is configured with a TCI state for a candidate cell for LTM, or when the UE receives a cell switch command via higher layer signaling / physical layer signaling, where the cell switch command may be a cell switch command for LTM.

[0379] <Variations of Condition 6> - If there is no pre-configuration according to specific application conditions (cell ID, etc.) during measurement of the TCI status for a candidate cell for LTM or after receiving a cell switch command, the UE may switch functionality / model to a pre-configured functionality / model or may perform a fallback operation.

[0380] In this case, since the target cell (candidate cell) may be in a different environment or use a different model, the UE may fall back to an existing beam management scheme to achieve better measurement and reporting of the candidate beam.

[0381] Furthermore, the above-mentioned preconfigured functionality / model may mean functionality / model configured for each candidate cell. Furthermore, "preconfigured" may mean that a specific parameter is signaled before RRC reconfiguration, or that a specific parameter is pre-implemented in the UE (product) (i.e., no signaling is required). In other words, "preconfigured parameters" may mean parameters signaled before RRC reconfiguration, or specific parameters pre-implemented in the UE (product). The specific parameters may be parameters set / instructed by, for example, LTM-related signaling (cell switch command).

[0382] <Condition 7> When the UE detects an application condition of a functionality / model or a change in the application condition. The application condition may be, for example, a condition related to a cell ID, a scenario, or a channel.

[0383] <Condition 8> The UE detects that a metric value (performance indicator) associated with the corresponding functionality / model is lower than a predefined / configured threshold, and a certain duration may be configured for detecting that the metric value is lower than the threshold.

[0384] <Condition 9> - A condition depending on the UE implementation may be pre-configured. In this case, the UE may be instructed to activate / trigger the fallback operation without satisfying the condition notified / configured by the NW (without a condition).

[0385] [Aspect 1-2] Aspect 1-2 relates to UE operation after a fallback operation is triggered.

[0386] As described above, after deactivating a configured functionality / model based on a specific condition, the UE may switch the functionality / model to (another) configured functionality / model, or may fall back to a configured / instructed / specified scheme depending on the configured functionality / model.

[0387] For example, the UE may be configured with a functionality / model with specific application / additional conditions for each functionality, and may fall back to the configured functionality / model.

[0388] Alternatively, the UE may be configured with parameters corresponding to a scheme of a legacy specification, in which case the UE may fall back to the scheme of the legacy specification.

[0389] For example, in the case of AI / ML-based CSI, the UE may be configured with an existing CSI such as Type 1 / Type 2 / Extended Type 1 / Extended Type 2 in the CSI reporting configuration, and may be configured with a parameter set for CSI reporting. In this case, the UE may report the configured CSI after a proactive fallback operation.

[0390] <Variations> The fallback behavior may be to apply a predefined / configured model ID that has the same functionality as the deactivated functionality.

[0391] [Aspect 1-3] Aspect 1-3 relates to reporting of fallback operations.

[0392] After deactivating a functionality / model, the UE may report the deactivation (content of fallback behavior), which may be performed implicitly / explicitly using higher layer signaling / physical layer signaling.

[0393] For example, if the UE performs the reporting implicitly, the UE may change the content / format of the proactive fallback action or may apply a special reporting content combination to indicate the occurrence of the fallback action.

[0394] For example, when the UE explicitly performs the report, the UE may send signaling (report) to the NW (e.g., base station) indicating that the corresponding functionality / model has been deactivated.

[0395] [Modification] The UE may report changes in the conditions / additional conditions or the occurrence of the above-mentioned events (such as triggering a fallback) through higher layer signaling / physical layer signaling.

[0396] After reporting or receiving a command to switch / fallback the model, the UE may switch to the configured functionality / model or fallback to the existing scheme.

[0397] According to this embodiment, the UE can control the triggering of the fallback operation in the LCM based on certain conditions.

[0398] Second Embodiment The second embodiment relates to UE behavior after fallback in LCM. More specifically, the second embodiment describes UE behavior when reactivating functionality / model when the triggering situation (deactivation of functionality / model) for proactive behavior (fallback / model switch) has ended. This UE behavior may be referred to as UE behavior for recovery from fallback (see, for example, FIG. 11 ).

[0399] After the fallback operation, the UE may perform at least one of the following operations in Aspects 2-1 to 2-2. As will be described in detail later, the UE may determine whether to reactivate functionality / models based on specific conditions.

[0400] [Aspect 2-1] The UE may activate functionality / model based on a received configuration / command. After model switching (switching to a configured default model) or fallback, the UE may control UE operation (reactivation operation) according to at least one of the following conditions 1 to 5. The UE operation (reactivation operation) may mean reactivating functionality / model. Conditions 1 to 5 may also be referred to as conditions for reactivation. According to the following conditions 1 to 5, signaling overhead can be reduced.

[0401] <Condition 1> The UE receives Ny consecutive "in-sync" indications from lower layers, where Ny may be a number configured / indicated / specified for the UE.

[0402] <Condition 2> The UE stops a predetermined timer for dealing with problems due to radio link failure. For example, the UE stops the timer (T310) after receiving an "in-sync" indication during N311 consecutive synchronizations (see FIG. 11).

[0403] <Condition 3> - When the UE successfully executes LTM (cell switching) based on a cell switching command after a fallback operation involving handover, for example, when the UE executes the first DL reception / UL transmission in the target cell.

[0404] <Variations of Condition 3> The UE may (re)activate functionality / models according to applicable conditions (e.g., cell ID) and pre-configuration (e.g., a list of cell ID and model ID pairs, i.e., a list in which cell IDs and model IDs are associated).

[0405] <Condition 4> - When the UE performs a connection re-establishment procedure after an RLF / BF, or when the UE performs DL reception / UL transmission for the first time after an RLF / BF.

[0406] <Modification 1 of Condition 4> Furthermore, when the UE re-establishes a radio link (connection) within the same cell.

[0407] <Variation 2 of Condition 4> When the UE re-establishes a radio link (connection) in a different cell, the UE may (re)activate functionality / model according to applicable conditions (e.g., cell ID) and pre-configuration (e.g., a list of cell ID and model ID pairs, i.e., a list in which cell IDs and model IDs are associated).

[0408] <Condition 5> - When the UE detects an application condition of a functionality / model or restoration / restore of the application condition.

[0409] [Aspect 2-2] The UE may control UE operation (reactivation operation) in accordance with at least one of the following conditions 1 to 4. The UE operation may mean not reactivating (not recovering) a functionality / model. In other words, the UE may not perform reactivation if at least one of the following conditions 1 to 4 applies. The following conditions 1 to 4 may be referred to as conditions for no-reactivation (no-reactivation-conditions).

[0410] <Condition 1> - When a predetermined timer for handling problems related to a radio link failure has ended / expired (see, for example, FIG. 11).

[0411] <Condition 2> - When the UE fails to transmit or receive a specific signal (DL reception / UL transmission in the target cell) while performing cell switching based on a cell switching command.

[0412] <Condition 3> - When the UE receives a specific command related to the corresponding functionality / model, the specific command may be, for example, signaling (which may be called higher layer signaling / physical layer signaling, e.g., a command for LCM) for activating / deactivating / falling back / switching to a specific model.

[0413] <Condition 4> - When the UE receives reconfiguration signaling (RRC reconfiguration) related to the corresponding functionality / model.

[0414] [Variation] The UE may report the status regarding the conditions for reactivation through higher layer signaling / physical layer signaling.

[0415] The UE may reactivate the corresponding functionality / model after reporting or after receiving a command to reactivate.

[0416] According to this embodiment, the UE can appropriately control the UE behavior after fallback in LCM based on certain conditions.

[0417] In the third embodiment, an example in which the first or second embodiment is applied to CSI feedback will be described. That is, a UE may apply the first or second embodiment described above to CSI feedback using the AI / ML technique (AI-based CSI feedback).

[0418] [Action 1] After the UE deactivates the functionality / model associated with CSI feedback, the UE may switch to the configured functionality / model associated with CSI feedback or may fall back to the existing CSI feedback scheme.

[0419] The UE may control the above-mentioned model switching / fallback based on specific higher layer / physical layer signaling configuration / instructions.

[0420] The existing CSI feedback may be, for example, Type 1 / 2 CSI, enhanced Type 2 CSI, predicted Type 2 CSI, port selection Type 2 CSI, enhanced port selection Type 2 CSI, etc.

[0421] The UE may be configured with a parameter / combination of parameters corresponding to an existing CSI feedback type (CSI type), or may use the configured parameter / combination of parameters as a default setting.

[0422] [Operation 2] After deactivating a functionality / model, the UE may report the deactivation (content or occurrence of fallback operation). The UE may perform this reporting implicitly / explicitly using higher layer signaling / physical layer signaling.

[0423] For example, if the UE performs the reporting implicitly, the UE may change the content / format of the proactive fallback action or may apply a special reporting content combination to indicate the occurrence of the fallback action.

[0424] For example, when the UE explicitly performs the report, the UE may send signaling (report) to the NW (e.g., base station) indicating that the corresponding functionality / model has been deactivated.

[0425] The UE may report existing CSI or CSI generated by functionality / model (AI-based CSI) through higher layer signaling / physical layer signaling.

[0426] Which CSI the UE reports (existing CSI or AI-based CSI) may be determined depending on the blind decoding of the base station (gNB).

[0427] When reporting existing CSI (performing existing CSI reporting), the UE may use a specified combination of existing content (such as a specified RI / CQI combination).

[0428] Furthermore, when reporting legacy CSI (performing legacy CSI reporting), the UE may use configured resources that are different from the resources used for AI-based CSI, for example, by explicit signaling indicating that functionality / models are deactivated.

[0429] The UE may report the type of CSI (existing CSI / AI-based CSI) through higher layer signaling / physical layer signaling.

[0430] According to this embodiment, the UE can appropriately control the fallback in CSI feedback.

[0431] <Fourth Embodiment> In the fourth embodiment, an example will be described in which the first or second embodiment is applied to beam prediction. That is, a UE may apply the first or second embodiment described above in beam prediction using AI / ML technology (AI-based beam prediction). Beam prediction may be interchangeably read as beam management.

[0432] [Action 1] After the UE deactivates the functionality / model associated with beam prediction, the UE may switch to the configured functionality / model associated with beam prediction or may fall back to the existing beam prediction scheme.

[0433] The UE may control the above-mentioned model switching / fallback based on specific higher layer / physical layer signaling configuration / instructions.

[0434] [Operation 2] The UE may use higher layer / physical layer signaling to implicitly report measurement results instead of prediction results.

[0435] Alternatively, the UE may use higher layer / physical layer signaling to explicitly report whether the value being reported is a predicted value or a measured value.

[0436] According to this embodiment, the UE can appropriately control the fallback in beam prediction.

[0437] <Supplementary Notes> [Supplementary Note 1: AI Model Information] In the present disclosure, AI model information may refer to information including at least one of the following: - Input / output information of the AI ​​model. - Pre-processing / post-processing information for the input / output of the AI ​​model. - Parameter information of the AI ​​model. - Training information for the AI ​​model. - Inference information for the AI ​​model. - Performance information regarding the AI ​​model.

[0438] Here, the input / output information of the AI ​​model may include information on at least one of the following: - Contents of the input / output data (e.g., RSRP, SINR, amplitude / phase information in the channel matrix (or precoding matrix), information on the angle of arrival (Angle of Arrival (AoA)), information on the angle of departure (Angle of Departure (AoD)), location information); - Auxiliary information of the data (which may be called meta-information); - Type of the input / output data (e.g., immutable value, floating-point number); - Bit width of the input / output data (e.g., 64 bits for each input value); - Quantization interval (quantization step size) of the input / output data (e.g., 1 dBm for L1-RSRP); - Range that the input / output data can take (e.g., [0, 1]).

[0439] In the present disclosure, the information on AoA may include information on at least one of an azimuth angle of arrival and a zenith angle of arrival (ZoA). The information on AoD may include information on at least one of an azimuth angle of departure and a zenith angle of departure (ZoD).

[0440] In the present disclosure, location information may be location information related to a UE / NW. The location information may include at least one of information (e.g., latitude, longitude, altitude) obtained using a positioning system (e.g., a satellite positioning system (Global Navigation Satellite System (GNSS), Global Positioning System (GPS), etc.)), information about a BS neighboring (or serving) the UE (e.g., a BS / cell identifier (ID), a BS-UE distance, a direction / angle of the BS (UE) as seen from the UE (BS), coordinates of the BS (UE) as seen from the UE (BS) (e.g., X / Y / Z axis coordinates), etc.), a specific address of the UE (e.g., an Internet Protocol (IP) address), etc. The location information of the UE is not limited to information based on the position of the BS, and may be information based on a specific point.

[0441] The location information may include information about its implementation (e.g., location / position / orientation of antennas, location / orientation of antenna panels, number of antennas, number of antenna panels, etc.).

[0442] The location information may include mobility information, which may include information indicating at least one of information indicating a mobility type, a moving speed of the UE, an acceleration of the UE, and a moving direction of the UE.

[0443] Here, the mobility type may correspond to at least one of a fixed location UE, a movable / moving UE, a no mobility UE, a low mobility UE, a middle mobility UE, a high mobility UE, a cell-edge UE, a not-cell-edge UE, etc.

[0444] In the present disclosure, environmental information (for data) may be information about the environment in which the data is acquired / used, and may correspond to, for example, frequency information (such as a band ID), environmental type information (information indicating at least one of indoor, outdoor, Urban Macro (UMa), Urban Micro (Umi), etc.), information indicating Line Of Site (LOS) / Non-Line Of Site (NLOS), etc.

[0445] Here, LOS may mean that the UE and the BS are in an environment where they can see each other (or there is no obstruction), and NLOS may mean that the UE and the BS are not in an environment where they can see each other (or there is an obstruction). The information indicating LOS / NLOS may indicate a soft value (e.g., the probability of LOS / NLOS) or a hard value (e.g., either LOS or NLOS).

[0446] In the present disclosure, meta-information may mean, for example, information regarding input / output information suitable for an AI model, information regarding acquired / acquirable data, etc. Specifically, meta-information may include information regarding beams of RS (e.g., CSI-RS / SRS / SSB, etc.) (e.g., the pointing angle of each beam, the 3 dB beam width, the shape of the pointed beam, the number of beams), layout information of the gNB / UE antenna, frequency information, environmental information, meta-information ID, etc. Note that meta-information may be used as input / output of the AI ​​model.

[0447] The pre-processing / post-processing information for the input / output of the AI ​​model may include information on at least one of the following: Whether to apply normalization (e.g., Z-score normalization (standardization), min-max normalization); Parameters for normalization (e.g., mean / variance for Z-score normalization, min / max for min-max normalization); Whether to apply a specific numerical conversion method (e.g., one hot encoding, label encoding, etc.); Selection rules for whether to use as training data.

[0448] For example, Z-score normalization (x) is performed as a preprocessing step for input information x. ｎｅｗ = (x - μ) / σ, where μ is the mean of x and σ is the standard deviation) ｎｅｗ may be input to the AI ​​model, and the output y ｏｕｔ may be subjected to post-processing to obtain the final output y.

[0449] The information on the parameters of the AI ​​model may include information on at least one of the following: - Information on weights in the AI ​​model (e.g., neuron coefficients (connection coefficients)); - Structure of the AI ​​model; - Type of the AI ​​model as a model component (e.g., Residual Network (ResNet), DenseNet, RefineNet, Transformer model, CRBlock, Recurrent Neural Network (RNN), Long Short-Term Memory (LSTM), Gated Recurrent Unit (GRU)); - Function of the AI ​​model as a model component (e.g., decoder, encoder).

[0450] Note that the weight information in the AI ​​model may include information on at least one of the following: - Bit width (size) of the weight information; - Quantization interval of the weight information; - Granularity of the weight information; - Range that the weight information can take; - Weight parameters in the AI ​​model; - Information on the difference from the AI ​​model before update (if updating); - Weight initialization method (e.g., zero initialization, random initialization (based on normal distribution / uniform distribution / truncated normal distribution), Xavier initialization (for sigmoid function), He initialization (for rectified linear units (ReLU))).

[0451] The structure of the AI ​​model may also include information about at least one of the following: the number of layers, the type of layer (e.g., convolutional layer, activation layer, dense layer, normalization layer, pooling layer, attention layer), layer information, time series specific parameters (e.g., bidirectionality, time step), parameters for training (e.g., type of function (e.g., L2 regularization, dropout function, etc.), where (e.g., after which layer) to place this function).

[0452] The layer information may include information about at least one of the following: Number of neurons in each layer; Kernel size; Stride for pooling / convolutional layers; Pooling method (MaxPooling, AveragePooling, etc.); Residual block information; Number of heads; Normalization method (Batch normalization, instance normalization, layer normalization, etc.); Activation function (Sigmoid, tanh function, ReLU, leaky ReLU information, Maxout, Softmax).

[0453] An AI model may be included as a component of another AI model, for example, an AI model that includes model component #1, ResNet, model component #2, a Transformer model, a dense layer, and a normalization layer in that order.

[0454] The training information for the AI ​​model may include information about at least one of the following: - Information for the optimization algorithm (e.g., type of optimization (Stochastic Gradient Descent (SGD)), AdaGrad, Adam, etc.), parameters of the optimization (learning rate, momentum information, etc.); - Information on the loss function (e.g., information on metrics of the loss function (Mean Absolute Error (MAE)), Mean Square Error (MSE), Cross Entropy Loss, NLL Loss, Kullback-Leibler (KL) Divergence, etc.)); - Parameters to be frozen for training (e.g., layers, weights); - Parameters to be updated (e.g., layers, weights); - Parameters to be (used as) initial parameters for training (e.g., layers, weights); - Method of training / updating the AI ​​model (e.g., (recommended) number of epochs, batch size, number of data to use for training).

[0455] The inference information for the AI ​​model may include information regarding decision tree branch pruning, parameter quantization, and functions of the AI ​​model, etc. Here, the functions of the AI ​​model may correspond to at least one of, for example, time domain beam prediction, spatial domain beam prediction, an autoencoder for CSI feedback, and an autoencoder for beam management.

[0456] An autoencoder for CSI feedback may be used as follows: The UE inputs the CSI / channel matrix / precoding matrix into the AI ​​model of the encoder and transmits the encoded bits as CSI feedback (CSI report). The BS inputs the received encoded bits into the AI ​​model of the decoder to reconstruct the CSI / channel matrix / precoding matrix, which is the output.

[0457] In spatial domain beam prediction, the UE / BS may input measurement results (beam quality, e.g., RSRP) based on sparse (or thick) beams into an AI model and output dense (or thin) beam quality.

[0458] In time domain beam prediction, the UE / BS may input time series (past, present, etc.) measurement results (beam quality, e.g., RSRP) into an AI model and output future beam quality.

[0459] The performance information regarding the AI ​​model may include information regarding the expected value of a loss function defined for the AI ​​model.

[0460] The AI ​​model information in the present disclosure may include information regarding the application range (applicable range) of the AI ​​model. The application range may be indicated by a physical cell ID, a serving cell index, etc. The information regarding the application range may be included in the above-mentioned environment information.

[0461] AI model information regarding a specific AI model may be predetermined in a standard or may be notified to a UE from a network (NW). An AI model defined in a standard may be referred to as a reference AI model. AI model information regarding a reference AI model may be referred to as reference AI model information.

[0462] Note that the AI ​​model information in the present disclosure may include an index for identifying the AI ​​model (which may be referred to as, for example, an AI model index, an AI model ID, a model ID, etc.). The AI ​​model information in the present disclosure may include an AI model index in addition to / instead of the input / output information of the AI ​​model described above. The association between the AI ​​model index and the AI ​​model information (for example, input / output information of the AI ​​model) may be predetermined in a standard, or may be notified to the UE from the NW.

[0463] The AI ​​model information in the present disclosure may be associated with an AI model and may be referred to as AI model relevant information, simply relevant information, etc. The AI ​​model relevant information does not need to explicitly include information for identifying the AI ​​model. The AI ​​model relevant information may be information that includes only meta information, for example.

[0464] In the present disclosure, the model ID may be interchangeably read as an ID (model set ID) corresponding to a set of AI models. Furthermore, in the present disclosure, the model ID may be interchangeably read as a meta information ID. The meta information (or the meta information ID) may be associated with information about a beam (beam setting) as described above. For example, the meta information (or the meta information ID) may be used by the UE to select an AI model taking into account which beam the BS is using, or may be used to notify the BS of which beam to use to apply the AI ​​model deployed by the UE. Furthermore, in the present disclosure, the meta information ID may be interchangeably read as an ID (meta information set ID) corresponding to a set of meta information.

[0465] [Supplementary Note 2: Notification of Information to UE] In the above-described embodiments, any information may be notified (from the NW) to the UE (in other words, reception of any information from the BS by the UE) using physical layer signaling (e.g., DCI), higher layer signaling (e.g., RRC signaling, MAC CE), a specific signal / channel (e.g., PDCCH, PDSCH, reference signal), or a combination thereof.

[0466] When the notification is performed by a MAC CE, the MAC CE may be identified by including a new Logical Channel ID (LCID) in the MAC subheader, which is not defined in existing standards.

[0467] When the notification is made by DCI, the notification may be made by a specific field of the DCI, a Radio Network Temporary Identifier (RNTI) used to scramble Cyclic Redundancy Check (CRC) bits assigned to the DCI, the format of the DCI, etc.

[0468] Furthermore, notification of any information to the UE in the above embodiments may be performed periodically, semi-persistently, or aperiodically.

[0469] [Supplementary Note 3: Notification of Information from UE] In the above-described embodiments, notification of any information from the UE (to the NW) (in other words, transmission / report of any information from the UE to the BS) may be performed using physical layer signaling (e.g., UCI), higher layer signaling (e.g., RRC signaling, MAC CE), a specific signal / channel (e.g., PUCCH, PUSCH, reference signal), or a combination thereof.

[0470] When the notification is performed by a MAC CE, the MAC CE may be identified by including a new LCID, which is not defined in existing standards, in the MAC subheader.

[0471] If the notification is made by UCI, the notification may be transmitted using PUCCH or PUSCH.

[0472] Furthermore, any information in the above-described embodiments may be notified from the UE periodically, semi-persistently, or aperiodically.

[0473] [Supplementary Note 4] In this disclosure, functionality may refer to a set of parameters (e.g., a set of parameters for CSI prediction / beam prediction / CSI compression) that can be supported based on conditions indicated by UE capabilities.

[0474] The UE may report parameter values ​​related to the functionality / model as conditions to the NW via higher layer signaling (e.g., RRC, MAC CE) / physical layer signaling (e.g., DCI). For example, the UE may report the conditions using UE capability / feature / feature group reporting.

[0475] The UE may report or instruct the parameter values ​​related to the functionality / model as additional conditions to the NW using higher layer signaling / physical layer signaling or methods other than signaling via the air interface of the NW (e.g., operator's configurations, pre-configured messages, etc.).

[0476] The UE may report or be instructed on information / instructions about the parameters corresponding to these additional conditions (e.g., parameter names) as information / instructions about the additional conditions using higher layer signaling / physical layer signaling or methods other than signaling via the air interface of the network (e.g., operator's configurations, pre-configured messages, etc.).

[0477] For example, the UE may report a device ID, a vendor ID, etc. as additional information. The UE may also be notified of a cell ID as an additional condition. The UE may also report or be instructed to report information such as a cell ID / UE ID instead of a parameter name.

[0478] Methods other than signaling via the air interface of the network may be methods related to pre-configuration of the UE (for example, configuration by the UE vendor) or operator configuration provided by the network operator.

[0479] [Supplementary Note 5] In the present disclosure, AI / ML-based CSI reporting (AI-based CSI reporting) may refer to CSI reporting associated with a model ID or specific functionality (e.g., predicted CSI, compressed CSI, advanced CSI, type x CSI, etc.).

[0480] In the present disclosure, AI / ML functionality, AI / ML functionality for CSI, and functionality for CSI may refer to functionality directed by the NW or functionality reported by the UE (e.g., predicted CSI, compressed CSI, advanced CSI, type x CSI, etc.).

[0481] In this disclosure, AI / ML model, AI / ML model for CSI, and model for CSI may refer to models / entities that are identified by ID or functionality and perform the specific functions described above.

[0482] In the present disclosure, AI / ML functionality, AI / ML functionality for beam management, and functionality for beam management may refer to functionality instructed by the NW or functionality reported by the UE (e.g., predicted L1-RSRP / SSB index / CSI-RS ID reporting associated with configured reference signals, etc.).

[0483] In this disclosure, AI / ML model, AI / ML model for beam management, and model for beam management may refer to models / entities that are identified by ID or functionality and perform the specific functions described above.

[0484] [Application of Each Embodiment] At least one of the above-described embodiments may be applied when a specific condition is met. The specific condition may be defined in a standard or may be notified to a UE / BS using higher layer signaling / physical layer signaling.

[0485] At least one of the above embodiments may be applied only to UEs that have reported or support specific UE capabilities, for example (by way of example only): Supporting specific processing / operations / control / information for at least one of the above embodiments; Supporting AI / ML-based CSI / beam prediction; Supporting LTM; Supporting fallback in AI / ML-based performance monitoring (LCM).

[0486] The particular UE capability may indicate support for particular processes / operations / controls / information for at least one of the above embodiments / options / choices.

[0487] Furthermore, the above-mentioned specific UE capability may be a capability that is applied across all frequencies (commonly regardless of frequency), or may be a capability for each frequency (e.g., one or a combination of a cell, a band, a band combination, a BWP, a component carrier, etc.), or may be a capability for each frequency range (e.g., Frequency Range 1 (FR1), FR2, FR3, FR4, FR5, FR2-1, FR2-2), or may be a capability for each subcarrier spacing (SubCarrier Spacing (SCS)), or may be a capability for each Feature Set (FS) or Feature Set Per Component-carrier (FSPC).

[0488] Furthermore, the specific UE capability may be a capability that is applied to all duplexing methods (commonly regardless of the duplexing method), or may be a capability for each duplexing method (e.g., Time Division Duplex (TDD) or Frequency Division Duplex (FDD)).

[0489] Furthermore, at least one of the above-described embodiments may be applied when the UE configures / activates / triggers specific information related to the above-described embodiments (or performs the operations of the above-described embodiments) through higher layer signaling / physical layer signaling. For example, the specific information may be information indicating that LCM based on a model / functionality ID is enabled, any RRC parameter for a specific release (e.g., Rel. 18 / 19), etc.

[0490] If the UE does not support at least one of the specific UE capabilities or is not configured with the specific information, the UE may apply, for example, the behavior of Rel. 15 / 16 / 17.

[0491] (Supplementary Notes) The following inventions are supplementary notes regarding an embodiment (first / third / fourth embodiment) of the present disclosure. [Supplementary Note 1] A terminal having: a receiving unit that receives information regarding a specific condition for lifecycle management fallback applied to a specific use case of an artificial intelligence (AI) model; and a control unit that controls triggering of a fallback operation based on the specific condition. [Supplementary Note 2] The terminal described in Supplementary Note 1, wherein the specific condition is reception of an instruction regarding at least one of radio link failure, beam failure detection, and cell switching triggered by the terminal. [Supplementary Note 3] The terminal described in Supplementary Note 1 or Supplementary Note 2, wherein the control unit executes, as the fallback operation, at least one of deactivating functionality or a model, model switching, and an operation that does not utilize the AI ​​model. [Supplementary Note 4] The terminal described in any of Supplements 1 to 3, wherein the specific use case is one of AI-based channel state information (CSI) feedback and beam management.

[0492] (Supplementary Notes) The following inventions are supplementary notes regarding an embodiment (second, third, or fourth embodiment) of the present disclosure. [Supplementary Note 1] A terminal having: a receiving unit that receives information regarding a trigger condition for a fallback of lifecycle management that is applied to a specific use case of an artificial intelligence (AI) model and a specific condition for recovery from the fallback; and a control unit that controls triggering of a fallback operation based on the trigger condition and controls operation after the fallback operation based on the specific condition. [Supplementary Note 2] The terminal described in Supplementary Note 1, wherein the specific condition is receiving a predetermined number of in-sync instructions from a lower layer. [Supplementary Note 3] The terminal described in Supplementary Note 1 or Supplementary Note 2, wherein the control unit determines whether to perform reactivation of a functionality or a model based on the specific condition. [Supplementary Note 4] The terminal described in any of Supplements 1 to 3, wherein the specific use case is one of AI-based channel state information (CSI) feedback and beam management.

[0493] (Wireless Communication System) The configuration of a wireless communication system according to an embodiment of the present disclosure will be described below. In this wireless communication system, communication is performed using any one of the wireless communication methods according to the above embodiments of the present disclosure or a combination thereof.

[0494] 12 is a diagram illustrating an example of a schematic configuration of a wireless communication system according to an embodiment. The wireless communication system 1 (which may be simply referred to as system 1) may be a system that realizes communication using Long Term Evolution (LTE) or 5th generation mobile communication system New Radio (5G NR) specified by the Third Generation Partnership Project (3GPP).

[0495] The wireless communication system 1 may also support dual connectivity between multiple Radio Access Technologies (RATs) (Multi-RAT Dual Connectivity (MR-DC)). MR-DC may include dual connectivity between LTE (Evolved Universal Terrestrial Radio Access (E-UTRA)) and NR (E-UTRA-NR Dual Connectivity (EN-DC)), dual connectivity between NR and LTE (NR-E-UTRA Dual Connectivity (NE-DC)), etc.

[0496] In EN-DC, the LTE (E-UTRA) base station (eNB) is the master node (Master Node (MN)), and the NR base station (gNB) is the secondary node (Secondary Node (SN)). In NE-DC, the NR base station (gNB) is the MN, and the LTE (E-UTRA) base station (eNB) is the SN.

[0497] The wireless communication system 1 may support dual connectivity between multiple base stations within the same RAT (for example, dual connectivity in which both the MN and SN are NR base stations (gNBs) (NR-NR Dual Connectivity (NN-DC))).

[0498] The wireless communication system 1 may include a base station 11 that forms a macrocell C1 with a relatively wide coverage, and base stations 12 (12a-12c) that are located within the macrocell C1 and form small cells C2 that are smaller than the macrocell C1. A user terminal 20 may be located within at least one of the cells. The locations and numbers of the cells and user terminals 20 are not limited to the embodiment shown in the figure. Hereinafter, when there is no need to distinguish between the base stations 11 and 12, they will be collectively referred to as base station 10.

[0499] The user terminal 20 may be connected to at least one of the multiple base stations 10. The user terminal 20 may utilize at least one of carrier aggregation (CA) using multiple component carriers (CCs) and dual connectivity (DC).

[0500] Each CC may be included in at least one of a first frequency band (Frequency Range 1 (FR1)) and a second frequency band (Frequency Range 2 (FR2)). The macro cell C1 may be included in FR1, and the small cell C2 may be included in FR2. For example, FR1 may be a frequency band of 6 GHz or less (sub-6 GHz), and FR2 may be a frequency band higher than 24 GHz (above-24 GHz). Note that the frequency bands and definitions of FR1 and FR2 are not limited to these, and for example, FR1 may correspond to a higher frequency band than FR2.

[0501] Furthermore, the user terminal 20 may perform communication using at least one of time division duplex (TDD) and frequency division duplex (FDD) in each CC.

[0502] The multiple base stations 10 may be connected by wire (e.g., optical fiber compliant with the Common Public Radio Interface (CPRI), an X2 interface, etc.) or wirelessly (e.g., NR communication). For example, when NR communication is used as a backhaul between the base stations 11 and 12, the base station 11 corresponding to the upper station may be called an Integrated Access Backhaul (IAB) donor, and the base station 12 corresponding to the relay station (relay) may be called an IAB node.

[0503] The base station 10 may be connected to the core network 30 directly or via another base station 10. The core network 30 may include, for example, at least one of an Evolved Packet Core (EPC), a 5G Core Network (5GCN), a Next Generation Core (NGC), and the like.

[0504] The core network 30 may include network functions (Network Functions (NF)) such as a User Plane Function (UPF), an Access and Mobility management Function (AMF), a Session Management Function (SMF), a Unified Data Management (UDM), an Application Function (AF), a Data Network (DN), a Location Management Function (LMF), and Operation, Administration and Maintenance (Management) (OAM). A single network node may provide multiple functions. Communication with an external network (e.g., the Internet) may also be performed via the DN.

[0505] The user terminal 20 may be a terminal that supports at least one of communication methods such as LTE, LTE-A, and 5G.

[0506] An Orthogonal Frequency Division Multiplexing (OFDM)-based radio access scheme may be used in the wireless communication system 1. For example, Cyclic Prefix OFDM (CP-OFDM), Discrete Fourier Transform Spread OFDM (DFT-s-OFDM), Orthogonal Frequency Division Multiple Access (OFDMA), Single Carrier Frequency Division Multiple Access (SC-FDMA), or the like may be used in at least one of the downlink (DL) and uplink (UL).

[0507] The radio access scheme may also be called a waveform. Note that in the wireless communication system 1, other radio access schemes (e.g., other single-carrier transmission schemes, other multi-carrier transmission schemes) may be used as the UL and DL radio access schemes.

[0508] In the wireless communication system 1, a downlink shared channel (Physical Downlink Shared Channel (PDSCH)) shared by each user terminal 20, a broadcast channel (Physical Broadcast Channel (PBCH)), a downlink control channel (Physical Downlink Control Channel (PDCCH)), etc. may be used as the downlink channel.

[0509] Furthermore, in the wireless communication system 1, an uplink shared channel (Physical Uplink Shared Channel (PUSCH)) shared by each user terminal 20, an uplink control channel (Physical Uplink Control Channel (PUCCH)), a random access channel (Physical Random Access Channel (PRACH)), or the like may be used as an uplink channel.

[0510] The PDSCH transmits user data, higher layer control information, a System Information Block (SIB), etc. The PUSCH may transmit user data, higher layer control information, etc. Furthermore, the PBCH may transmit a Master Information Block (MIB).

[0511] Lower layer control information may be transmitted by the PDCCH. The lower layer control information may include, for example, Downlink Control Information (DCI) including scheduling information for at least one of the PDSCH and the PUSCH.

[0512] Note that the DCI for scheduling the PDSCH may be referred to as a DL assignment, a DL DCI, etc., and the DCI for scheduling the PUSCH may be referred to as a UL grant, a UL DCI, etc. Note that the PDSCH may be replaced with DL data, and the PUSCH may be replaced with UL data.

[0513] A control resource set (CORESET) and a search space may be used to detect the PDCCH. The CORESET corresponds to resources for searching for DCI. The search space corresponds to a search region and a search method for PDCCH candidates. One CORESET may be associated with one or more search spaces. The UE may monitor the CORESET associated with a certain search space based on the search space configuration.

[0514] One search space may correspond to PDCCH candidates corresponding to one or more aggregation levels. One or more search spaces may be referred to as a search space set. Note that the terms "search space," "search space set," "search space configuration," "search space set configuration," "CORESET," "CORESET configuration," and the like in the present disclosure may be read interchangeably.

[0515] The PUCCH may transmit uplink control information (UCI) including at least one of channel state information (CSI), delivery confirmation information (which may be called, for example, Hybrid Automatic Repeat reQuest ACKnowledgement (HARQ-ACK), ACK / NACK, etc.), and scheduling request (SR). The PRACH may transmit a random access preamble for establishing a connection with a cell.

[0516] In the present disclosure, downlink, uplink, etc. may be expressed without adding "link." Also, various channels may be expressed without adding "Physical" to the beginning.

[0517] In the wireless communication system 1, a synchronization signal (SS), a downlink reference signal (DL-RS), etc. may be transmitted. In the wireless communication system 1, as the DL-RS, a cell-specific reference signal (CRS), a channel state information reference signal (CSI-RS), a demodulation reference signal (DMRS), a positioning reference signal (PRS), a phase tracking reference signal (PTRS), etc. may be transmitted.

[0518] The synchronization signal may be, for example, at least one of a Primary Synchronization Signal (PSS) and a Secondary Synchronization Signal (SSS). A signal block including an SS (PSS, SSS) and a PBCH (and a DMRS for the PBCH) may be referred to as an SS / PBCH block, an SS Block (SSB), or the like. Note that the SS, SSB, and the like may also be referred to as a reference signal.

[0519] Furthermore, in the wireless communication system 1, a sounding reference signal (SRS), a demodulation reference signal (DMRS), or the like may be transmitted as an uplink reference signal (UL-RS). Note that the DMRS may also be called a user equipment-specific reference signal (UE-specific reference signal).

[0520] (Base Station) Fig. 13 is a diagram showing an example of the configuration of a base station according to an embodiment. The base station 10 includes a control unit 110, a transceiver unit 120, a transceiver antenna 130, and a transmission line interface 140. Note that the base station may include one or more of each of the control unit 110, the transceiver unit 120, the transceiver antenna 130, and the transmission line interface 140.

[0521] In this example, the functional blocks of the characteristic parts of the present embodiment are mainly shown, and it may be assumed that the base station 10 also has other functional blocks necessary for wireless communication. Some of the processing of each unit described below may be omitted.

[0522] The control unit 110 performs overall control of the base station 10. The control unit 110 can be configured from a controller, a control circuit, and the like that are explained based on common understanding in the technical field to which the present disclosure relates.

[0523] The control unit 110 may control signal generation, scheduling (e.g., resource allocation, mapping), etc. The control unit 110 may control transmission and reception using the transceiver unit 120, the transceiver antenna 130, and the transmission path interface 140, measurement, etc. The control unit 110 may generate data, control information, sequences, etc. to be transmitted as signals, and transfer them to the transceiver unit 120. The control unit 110 may perform call processing (setting up, releasing, etc.) of communication channels, status management of the base station 10, management of radio resources, etc.

[0524] The transceiver unit 120 may include a baseband unit 121, a radio frequency (RF) unit 122, and a measurement unit 123. The baseband unit 121 may include a transmission processing unit 1211 and a reception processing unit 1212. The transceiver unit 120 may be configured with a transmitter / receiver, an RF circuit, a baseband circuit, a filter, a phase shifter, a measurement circuit, a transceiver circuit, etc., which are described based on common understanding in the technical field related to the present disclosure.

[0525] The transmitting / receiving unit 120 may be configured as an integrated transmitting / receiving unit, or may be configured from a transmitting unit and a receiving unit. The transmitting unit may be configured from a transmission processing unit 1211 and an RF unit 122. The receiving unit may be configured from a reception processing unit 1212, the RF unit 122, and a measurement unit 123.

[0526] The transmitting and receiving antenna 130 can be configured from an antenna described based on common understanding in the technical field to which the present disclosure relates, such as an array antenna.

[0527] The transceiver 120 may transmit the above-mentioned downlink channel, synchronization signal, downlink reference signal, etc. The transceiver 120 may receive the above-mentioned uplink channel, uplink reference signal, etc.

[0528] The transceiver 120 may form at least one of the transmit beam and the receive beam using digital beamforming (e.g., precoding), analog beamforming (e.g., phase rotation), or the like.

[0529] The transmitter / receiver unit 120 (transmission processing unit 1211) may perform Packet Data Convergence Protocol (PDCP) layer processing, Radio Link Control (RLC) layer processing (e.g., RLC retransmission control), Medium Access Control (MAC) layer processing (e.g., HARQ retransmission control), etc. on data, control information, etc. obtained from the control unit 110, and generate a bit string to be transmitted.

[0530] The transmitter / receiver unit 120 (transmission processing unit 1211) may perform transmission processing such as channel coding (which may include error correction coding), modulation, mapping, filtering, Discrete Fourier Transform (DFT) processing (if necessary), Inverse Fast Fourier Transform (IFFT) processing, precoding, and digital-to-analog conversion on the bit string to be transmitted, and output a baseband signal.

[0531] The transceiver unit 120 (RF unit 122) may perform modulation, filtering, amplification, etc. on the baseband signal to a radio frequency band, and transmit the radio frequency band signal via the transceiver antenna 130.

[0532] On the other hand, the transceiver unit 120 (RF unit 122) may perform amplification, filtering, demodulation to a baseband signal, etc. on the radio frequency band signal received by the transceiver antenna 130.

[0533] The transceiver 120 (reception processing unit 1212) may apply reception processing such as analog-to-digital conversion, Fast Fourier Transform (FFT) processing, Inverse Discrete Fourier Transform (IDFT) processing (if necessary), filtering, demapping, demodulation, decoding (which may include error correction decoding), MAC layer processing, RLC layer processing, and PDCP layer processing to the acquired baseband signal, thereby acquiring user data, etc.

[0534] The transceiver 120 (measurement unit 123) may perform measurements on the received signal. For example, the measurement unit 123 may perform Radio Resource Management (RRM) measurements, Channel State Information (CSI) measurements, etc. based on the received signal. The measurement unit 123 may measure received power (e.g., Reference Signal Received Power (RSRP)), received quality (e.g., Reference Signal Received Quality (RSRQ), Signal to Interference plus Noise Ratio (SINR), Signal to Noise Ratio (SNR)), signal strength (e.g., Received Signal Strength Indicator (RSSI)), propagation path information (e.g., CSI), etc. The measurement results may be output to the control unit 110.

[0535] The transmission path interface 140 may transmit and receive signals (backhaul signaling) between devices included in the core network 30 (e.g., network nodes that provide NF), other base stations 10, etc., and may acquire and transmit user data (user plane data), control plane data, etc. for the user terminal 20.

[0536] The transmitting section and receiving section of the base station 10 in the present disclosure may be configured by at least one of the transmitting / receiving section 120, the transmitting / receiving antenna 130, and the transmission path interface 140.

[0537] The transceiver 120 may transmit information regarding specific conditions for lifecycle management fallbacks applied to specific use cases of artificial intelligence (AI) models to the terminal, and the control unit 110 may control reception of reports of fallback operations triggered by the terminal based on the specific conditions.

[0538] The transceiver 120 may transmit information regarding a trigger condition for a lifecycle management fallback that is applied to a specific use case of an artificial intelligence (AI) model and a specific condition for recovery from the fallback. The control unit 110 may control reception of a report on a fallback operation triggered by a terminal based on the trigger condition and an operation after the fallback operation triggered by the terminal based on the specific condition.

[0539] (User Terminal) Fig. 14 is a diagram showing an example of the configuration of a user terminal according to one embodiment. The user terminal 20 includes a control unit 210, a transceiver unit 220, and a transceiver antenna 230. Note that the user terminal 20 may include one or more of each of the control unit 210, the transceiver unit 220, and the transceiver antenna 230.

[0540] In this example, the functional blocks of the characteristic parts of the present embodiment are mainly shown, and it may be assumed that the user terminal 20 also has other functional blocks necessary for wireless communication. Some of the processing of each unit described below may be omitted.

[0541] The control unit 210 performs overall control of the user terminal 20. The control unit 210 can be configured from a controller, a control circuit, etc., which are described based on common understanding in the technical field to which the present disclosure relates.

[0542] The control unit 210 may control signal generation, mapping, etc. The control unit 210 may control transmission and reception, measurement, etc. using the transceiver unit 220 and the transceiver antenna 230. The control unit 210 may generate data, control information, sequences, etc. to be transmitted as signals and transfer them to the transceiver unit 220.

[0543] The transceiver unit 220 may include a baseband unit 221, an RF unit 222, and a measurement unit 223. The baseband unit 221 may include a transmission processing unit 2211 and a reception processing unit 2212. The transceiver unit 220 may be configured with a transmitter / receiver, an RF circuit, a baseband circuit, a filter, a phase shifter, a measurement circuit, a transceiver circuit, etc., which are described based on common understanding in the technical field related to the present disclosure.

[0544] The transmitting / receiving unit 220 may be configured as an integrated transmitting / receiving unit, or may be composed of a transmitting unit and a receiving unit. The transmitting unit may be composed of a transmission processing unit 2211 and an RF unit 222. The receiving unit may be composed of a reception processing unit 2212, an RF unit 222, and a measurement unit 223.

[0545] The transmitting / receiving antenna 230 can be configured from an antenna described based on common understanding in the technical field to which the present disclosure relates, such as an array antenna.

[0546] The transceiver 220 may receive the above-mentioned downlink channel, synchronization signal, downlink reference signal, etc. The transceiver 220 may transmit the above-mentioned uplink channel, uplink reference signal, etc.

[0547] The transceiver unit 220 may form at least one of the transmit beam and the receive beam using digital beamforming (e.g., precoding), analog beamforming (e.g., phase rotation), or the like.

[0548] The transceiver unit 220 (transmission processing unit 2211) may perform PDCP layer processing, RLC layer processing (e.g., RLC retransmission control), MAC layer processing (e.g., HARQ retransmission control), etc. on data, control information, etc. obtained from the control unit 210, and generate a bit string to be transmitted.

[0549] The transmitter / receiver unit 220 (transmission processing unit 2211) may perform transmission processing such as channel coding (which may include error correction coding), modulation, mapping, filtering, DFT processing (if necessary), IFFT processing, precoding, and digital-to-analog conversion on the bit string to be transmitted, and output a baseband signal.

[0550] Whether or not to apply DFT processing may be based on the setting of transform precoding. When transform precoding is enabled for a certain channel (e.g., PUSCH), the transceiver unit 220 (transmission processing unit 2211) may perform DFT processing as the transmission processing to transmit the channel using a DFT-s-OFDM waveform, and if not, it may not be necessary to perform DFT processing as the transmission processing.

[0551] The transceiver unit 220 (RF unit 222) may perform modulation, filtering, amplification, etc. on the baseband signal to a radio frequency band, and transmit the radio frequency band signal via the transceiver antenna 230.

[0552] On the other hand, the transceiver unit 220 (RF unit 222) may perform amplification, filtering, demodulation to a baseband signal, etc. on the radio frequency band signal received by the transceiver antenna 230.

[0553] The transceiver unit 220 (reception processing unit 2212) may apply reception processing such as analog-to-digital conversion, FFT processing, IDFT processing (if necessary), filtering, demapping, demodulation, decoding (which may include error correction decoding), MAC layer processing, RLC layer processing, and PDCP layer processing to the acquired baseband signal, and acquire user data, etc.

[0554] The transceiver 220 (measurement unit 223) may perform measurements on the received signal. For example, the measurement unit 223 may perform RRM measurements, CSI measurements, etc. based on the received signal. The measurement unit 223 may measure received power (e.g., RSRP), received quality (e.g., RSRQ, SINR, SNR), signal strength (e.g., RSSI), propagation path information (e.g., CSI), etc. The measurement results may be output to the control unit 210.

[0555] The measurement unit 223 may derive channel measurements for CSI calculation based on the channel measurement resources. The channel measurement resources may be, for example, non-zero power (NZP) CSI-RS resources. The measurement unit 223 may also derive interference measurements for CSI calculation based on the interference measurement resources. The interference measurement resources may be at least one of an NZP CSI-RS resource for interference measurement, a CSI-Interference Measurement (IM) resource, etc. Note that CSI-IM may be referred to as CSI-Interference Management (IM) or may be interchangeably read as Zero Power (ZP) CSI-RS. Note that in the present disclosure, CSI-RS, NZP CSI-RS, ZP CSI-RS, CSI-IM, CSI-SSB, etc. may be interchangeably read as interchangeable.

[0556] The transmitting unit and receiving unit of the user terminal 20 in the present disclosure may be configured by at least one of the transmitting / receiving unit 220 and the transmitting / receiving antenna 230.

[0557] The transceiver 220 may receive information regarding a specific condition for lifecycle management fallback applied to a specific use case of an artificial intelligence (AI) model. The controller 210 may control triggering of a fallback operation based on the specific condition. The specific condition may be reception of an instruction regarding at least one of a radio link failure, a beam failure detection, and a cell switch triggered by the terminal. The controller 210 may execute at least one of deactivating a functionality or model, switching a model, and an operation not using the AI ​​model as the fallback operation. The specific use case may be any of AI-based channel state information (CSI) feedback and beam management.

[0558] The transceiver 220 may receive information regarding a trigger condition for a lifecycle management fallback and a specific condition for recovery from the fallback, which are applied to a specific use case of an artificial intelligence (AI) model. The controller 210 may control triggering of a fallback operation based on the trigger condition and control operation after the fallback operation based on the specific condition. The specific condition may be receiving a predetermined number of in-sync instructions from a lower layer. The controller 210 may determine whether to reactivate a functionality or model based on the specific condition. The specific use case may be AI-based channel state information (CSI) feedback and beam management.

[0559] (Hardware Configuration) Note that the block diagrams used to explain the above embodiments show functional blocks. These functional blocks (components) are realized by any combination of at least one of hardware and software. Furthermore, the method for realizing each functional block is not particularly limited. That is, each functional block may be realized using a single device that is physically or logically coupled, or may be realized using two or more physically or logically separated devices that are directly or indirectly connected (for example, using wires, wirelessly, etc.) and these multiple devices. The functional block may be realized by combining software with the single device or the multiple devices.

[0560] Here, the functions include, but are not limited to, judgment, determination, judgment, calculation, computation, processing, derivation, investigation, search, confirmation, reception, transmission, output, access, resolution, selection, election, establishment, comparison, assumption, expectation, deeming, broadcasting, notifying, communicating, forwarding, configuring, reconfiguring, allocating, mapping, and assignment. For example, a functional block (component) that performs transmission may be called a transmitting unit, transmitter, etc. As described above, the implementation method of each is not particularly limited.

[0561] For example, a base station, a user terminal, etc. according to an embodiment of the present disclosure may function as a computer that performs processing of the wireless communication method of the present disclosure. Fig. 15 is a diagram illustrating an example of the hardware configuration of a base station and a user terminal according to an embodiment. The above-described base station 10 and user terminal 20 may be physically configured as a computer device including a processor 1001, a memory 1002, a storage 1003, a communication device 1004, an input device 1005, an output device 1006, a bus 1007, etc.

[0562] In the present disclosure, the terms apparatus, circuit, device, section, unit, etc. may be used interchangeably. The hardware configurations of the base station 10 and the user terminal 20 may be configured to include one or more of the devices shown in the drawings, or may be configured to exclude some of the devices.

[0563] For example, although only one processor 1001 is shown, there may be multiple processors. Furthermore, processing may be performed by one processor, or processing may be performed by two or more processors simultaneously, serially, or in other ways. Furthermore, processor 1001 may be implemented by one or more chips.

[0564] Each function in the base station 10 and the user terminal 20 is realized, for example, by loading specified software (programs) onto hardware such as a processor 1001 and a memory 1002, causing the processor 1001 to perform calculations, control communication via the communication device 1004, and control at least one of reading and writing data in the memory 1002 and the storage 1003.

[0565] The processor 1001, for example, runs an operating system to control the entire computer. The processor 1001 may be configured as a central processing unit (CPU) including an interface with peripheral devices, a control device, an arithmetic unit, a register, etc. For example, at least a part of the above-mentioned control unit 110 (210), transceiver unit 120 (220), etc. may be realized by the processor 1001.

[0566] The processor 1001 also reads programs (program codes), software modules, data, etc. from at least one of the storage 1003 and the communication device 1004 into the memory 1002 and executes various processes in accordance with these. The programs used are those that cause a computer to execute at least some of the operations described in the above-described embodiments. For example, the control unit 110 (210) may be implemented by a control program stored in the memory 1002 and running on the processor 1001, and the other functional blocks may be implemented in a similar manner.

[0567] The memory 1002 is a computer-readable recording medium and may be configured by at least one of, for example, Read Only Memory (ROM), Erasable Programmable ROM (EPROM), Electrically EEPROM (EEPROM), Random Access Memory (RAM), or other suitable storage medium. The memory 1002 may also be referred to as a register, cache, main memory, etc. The memory 1002 may store executable programs (program codes), software modules, etc. for implementing a wireless communication method according to an embodiment of the present disclosure.

[0568] Storage 1003 is a computer-readable recording medium and may be composed of at least one of, for example, a flexible disk, a floppy disk, a magneto-optical disk (e.g., a compact disc (e.g., a Compact Disc ROM (CD-ROM)), a digital versatile disc, a Blu-ray disc), a removable disk, a hard disk drive, a smart card, a flash memory device (e.g., a card, a stick, a key drive), a magnetic stripe, a database, a server, or other suitable storage medium. Storage 1003 may also be referred to as an auxiliary storage device.

[0569] The communication device 1004 is hardware (transmission / reception device) for communicating between computers via at least one of a wired network and a wireless network, and is also referred to as, for example, a network device, a network controller, a network card, or a communication module. The communication device 1004 may be configured to include a high-frequency switch, a duplexer, a filter, a frequency synthesizer, etc. to realize at least one of frequency division duplex (FDD) and time division duplex (TDD). For example, the above-mentioned transmission / reception unit 120 (220), transmission / reception antenna 130 (230), etc. may be realized by the communication device 1004. The transmission / reception unit 120 (220) may be implemented as a transmission unit 120a (220a) and a reception unit 120b (220b) that are physically or logically separated.

[0570] The input device 1005 is an input device (e.g., a keyboard, a mouse, a microphone, a switch, a button, a sensor, etc.) that accepts input from the outside. The output device 1006 is an output device (e.g., a display, a speaker, a light emitting diode (LED) lamp, etc.) that outputs to the outside. Note that the input device 1005 and the output device 1006 may be integrated into one device (e.g., a touch panel).

[0571] Furthermore, each device, such as the processor 1001 and the memory 1002, is connected by a bus 1007 for communicating information. The bus 1007 may be configured using a single bus, or may be configured using different buses between each device.

[0572] Furthermore, the base station 10 and the user terminal 20 may be configured to include hardware such as a microprocessor, a digital signal processor (DSP), an application specific integrated circuit (ASIC), a programmable logic device (PLD), or a field programmable gate array (FPGA), and some or all of the functional blocks may be realized using this hardware. For example, the processor 1001 may be implemented using at least one of these pieces of hardware.

[0573] (Modifications) Note that terms described in the present disclosure and terms necessary for understanding the present disclosure may be replaced with terms having the same or similar meanings. For example, a channel, a symbol, and a signal (signal or signaling) may be interchangeable. A signal may also be a message. A reference signal may be abbreviated as RS, and may also be called a pilot, pilot signal, etc. depending on the applicable standard. A component carrier (CC) may also be called a cell, frequency carrier, carrier frequency, etc.

[0574] A radio frame may be composed of one or more periods (frames) in the time domain. Each of the one or more periods (frames) constituting a radio frame may be called a subframe. Furthermore, a subframe may be composed of one or more slots in the time domain. A subframe may have a fixed time length (e.g., 1 ms) that is independent of numerology.

[0575] Here, the numerology may be a communication parameter applied to at least one of transmission and reception of a signal or channel, and may indicate at least one of, for example, Subcarrier Spacing (SCS), bandwidth, symbol length, cyclic prefix length, Transmission Time Interval (TTI), number of symbols per TTI, radio frame structure, specific filtering performed by a transceiver in the frequency domain, and specific windowing performed by a transceiver in the time domain.

[0576] A slot may be composed of one or more symbols (such as an Orthogonal Frequency Division Multiplexing (OFDM) symbol or a Single Carrier Frequency Division Multiple Access (SC-FDMA) symbol) in the time domain. A slot may also be a time unit based on numerology.

[0577] A slot may include multiple minislots. Each minislot may consist of one or multiple symbols in the time domain. A minislot may also be called a subslot. A minislot may consist of fewer symbols than a slot. A PDSCH (or PUSCH) transmitted in a time unit larger than a minislot may be called PDSCH (PUSCH) mapping type A. A PDSCH (or PUSCH) transmitted using a minislot may be called PDSCH (PUSCH) mapping type B.

[0578] A radio frame, a subframe, a slot, a minislot, and a symbol all represent time units for transmitting signals. The radio frame, the subframe, the slot, the minislot, and the symbol may be referred to by other names corresponding to the radio frame, the subframe, the slot, the minislot, and the symbol. Note that the time units such as a frame, a subframe, a slot, a minislot, and a symbol in the present disclosure may be interchangeable.

[0579] For example, one subframe may be referred to as a TTI, or multiple consecutive subframes may be referred to as a TTI, or one slot or one minislot may be referred to as a TTI. That is, at least one of the subframe and the TTI may be a subframe (1 ms) in existing LTE, a period shorter than 1 ms (for example, 1-13 symbols), or a period longer than 1 ms. Note that the unit representing the TTI may be called a slot, minislot, etc. instead of a subframe.

[0580] Here, TTI refers to, for example, the smallest time unit for scheduling in wireless communication. For example, in an LTE system, a base station performs scheduling to allocate radio resources (such as frequency bandwidth and transmission power that can be used by each user terminal) to each user terminal in TTI units. Note that the definition of TTI is not limited to this.

[0581] The TTI may be a transmission time unit for a channel-encoded data packet (transport block), a code block, a code word, etc., or may be a processing unit for scheduling, link adaptation, etc. When a TTI is given, the time interval (e.g., the number of symbols) to which a transport block, a code block, a code word, etc. is actually mapped may be shorter than the TTI.

[0582] When one slot or one minislot is called a TTI, one or more TTIs (i.e., one or more slots or one or more minislots) may be the minimum time unit for scheduling. Also, the number of slots (minislots) constituting the minimum time unit for scheduling may be controlled.

[0583] A TTI having a time length of 1 ms may be called a regular TTI (TTI in 3GPP Rel. 8-12), normal TTI, long TTI, regular subframe, normal subframe, long subframe, slot, etc. A TTI shorter than a regular TTI may be called a shortened TTI, short TTI, partial or fractional TTI, shortened subframe, short subframe, minislot, subslot, slot, etc.

[0584] In addition, a long TTI (e.g., a normal TTI, a subframe, etc.) may be interpreted as a TTI having a time length of more than 1 ms, and a short TTI (e.g., a shortened TTI, etc.) may be interpreted as a TTI having a TTI length shorter than the TTI length of a long TTI and greater than or equal to 1 ms.

[0585] A resource block (RB) is a resource allocation unit in the time domain and the frequency domain, and may include one or more consecutive subcarriers in the frequency domain. The number of subcarriers included in an RB may be the same regardless of numerology, for example, 12. The number of subcarriers included in an RB may be determined based on numerology.

[0586] In addition, an RB may include one or more symbols in the time domain and may have a length of one slot, one minislot, one subframe, or one TTI, each of which may be composed of one or more resource blocks.

[0587] In addition, one or more RBs may be referred to as a physical resource block (PRB), a sub-carrier group (SCG), a resource element group (REG), a PRB pair, an RB pair, etc.

[0588] Furthermore, a resource block may be composed of one or more resource elements (REs). For example, one RE may be a radio resource region of one subcarrier and one symbol.

[0589] A Bandwidth Part (BWP), which may also be referred to as a partial bandwidth, may represent a subset of contiguous common resource blocks (RBs) for a given numerology on a given carrier, where the common RBs may be identified by their index relative to a Common Reference Point of the carrier. PRBs may be defined in a BWP and numbered within the BWP.

[0590] The BWP may include a UL BWP (BWP for UL) and a DL BWP (BWP for DL). One or more BWPs may be configured for a UE within one carrier.

[0591] At least one of the configured BWPs may be active, and the UE may not expect to transmit or receive a given signal / channel outside the active BWP. Note that the terms "cell," "carrier," etc. in this disclosure may be read as "BWP."

[0592] The above-described structures of radio frames, subframes, slots, minislots, symbols, etc. are merely examples. For example, the number of subframes included in a radio frame, the number of slots per subframe or radio frame, the number of minislots included in a slot, the number of symbols and RBs included in a slot or minislot, the number of subcarriers included in an RB, the number of symbols in a TTI, the symbol length, the cyclic prefix (CP) length, etc. may be changed in various ways.

[0593] Furthermore, the information, parameters, etc. described in the present disclosure may be expressed using absolute values, may be expressed using relative values ​​from a predetermined value, or may be expressed using other corresponding information. For example, a radio resource may be indicated by a predetermined index.

[0594] The names used for parameters and the like in this disclosure are not intended to be limiting in any way. Furthermore, the mathematical expressions and the like using these parameters may differ from those explicitly disclosed in this disclosure. The various channels (PUCCH, PDCCH, etc.) and information elements may be identified by any suitable names, and therefore the various names assigned to these various channels and information elements are not intended to be limiting in any way.

[0595] The information, signals, etc. described in this disclosure may be represented using any of a variety of different technologies. For example, data, instructions, commands, information, signals, bits, symbols, chips, etc. that may be referred to throughout the above description may be represented by voltages, currents, electromagnetic waves, magnetic fields or magnetic particles, optical fields or photons, or any combination thereof.

[0596] Furthermore, information, signals, etc. may be output from a higher layer to a lower layer and / or from a lower layer to a higher layer. Information, signals, etc. may be input / output via multiple network nodes.

[0597] Input and output information, signals, etc. may be stored in a specific location (for example, memory) or may be managed using a management table. Input and output information, signals, etc. may be overwritten, updated, or added. Output information, signals, etc. may be deleted. Input information, signals, etc. may be transmitted to another device.

[0598] The notification of information is not limited to the aspects / embodiments described in the present disclosure, and may be performed using other methods. For example, the notification of information in the present disclosure may be performed by physical layer signaling (e.g., Downlink Control Information (DCI) and Uplink Control Information (UCI)), higher layer signaling (e.g., Radio Resource Control (RRC) signaling, broadcast information (Master Information Block (MIB) and System Information Block (SIB)), Medium Access Control (MAC) signaling), other signals, or a combination thereof.

[0599] Note that the physical layer signaling may be referred to as Layer 1 / Layer 2 (L1 / L2) control information (L1 / L2 control signal), L1 control information (L1 control signal), etc. Furthermore, the RRC signaling may be referred to as an RRC message, such as an RRC Connection Setup message or an RRC Connection Reconfiguration message. Furthermore, the MAC signaling may be notified using, for example, a MAC Control Element (CE).

[0600] Furthermore, notification of specified information (e.g., notification that "it is X") is not limited to explicit notification, but may be made implicitly (e.g., by not notifying the specified information or by notifying other information).

[0601] The determination may be made by a value represented by one bit (0 or 1), by a Boolean value represented by true or false, or by a comparison of numerical values ​​(e.g., comparison with a predetermined value).

[0602] Software shall be construed broadly to mean instructions, instruction sets, code, code segments, program code, programs, subprograms, software modules, applications, software applications, software packages, routines, subroutines, objects, executable files, threads of execution, procedures, functions, etc., whether referred to as software, firmware, middleware, microcode, hardware description language, or otherwise.

[0603] Software, instructions, information, etc. may also be transmitted or received over a transmission medium. For example, if software is transmitted from a website, server, or other remote source using wired technologies (such as coaxial cable, fiber optic cable, twisted pair, Digital Subscriber Line (DSL)), and / or wireless technologies (such as infrared, microwave), these wired and / or wireless technologies are included within the definition of transmission media.

[0604] As used in this disclosure, the terms "system" and "network" may be used interchangeably. A "network" may refer to devices included in the network (e.g., base stations).

[0605] In this disclosure, terms such as "precoding," "precoder," "weight (precoding weight)," "Quasi-Co-Location (QCL)," "Transmission Configuration Indication state (TCI state)," "spatial relation," "spatial domain filter," "transmit power," "phase rotation," "antenna port," "layer," "number of layers," "rank," "resource," "resource set," "beam," "beam width," "beam angle," "antenna," "antenna element," "panel," "UE panel," "transmitting entity," "receiving entity," etc. may be used interchangeably.

[0606] In the present disclosure, the term "antenna port" may be interchangeably read as an antenna port for any signal / channel (e.g., a demodulation reference signal (DMRS) port). In the present disclosure, the term "resource" may be interchangeably read as a resource for any signal / channel (e.g., a reference signal resource, an SRS resource, etc.). The resource may include time / frequency / code / space / power resources. Furthermore, the spatial domain transmission filter may include at least one of a spatial domain transmission filter and a spatial domain reception filter.

[0607] The group may include, for example, at least one of a spatial relationship group, a Code Division Multiplexing (CDM) group, a Reference Signal (RS) group, a Control Resource Set (CORESET) group, a PUCCH group, an antenna port group (e.g., a DMRS port group), a layer group, a resource group, a beam group, an antenna group, a panel group, and the like.

[0608] In addition, in the present disclosure, beam, SRS Resource Indicator (SRI), CORESET, CORESET pool, PDSCH, PUSCH, codeword (CW), transport block (TB), RS, etc. may be read as interchangeable terms.

[0609] In addition, in the present disclosure, the terms TCI state, downlink TCI state (DL TCI state), uplink TCI state (UL TCI state), unified TCI state, common TCI state, joint TCI state, etc. may be read interchangeably.

[0610] Furthermore, in the present disclosure, terms such as "QCL," "QCL assumption," "QCL relationship," "QCL type information," "QCL property / properties," "specific QCL type (e.g., Type A, Type D) property," and "specific QCL type (e.g., Type A, Type D)" may be interchangeable.

[0611] In the present disclosure, terms such as index, identifier (ID), indicator, indication, and resource ID may be interchangeable. In the present disclosure, terms such as sequence, list, set, group, cluster, and subset may be interchangeable.

[0612] Furthermore, the spatial relationship information identifier (ID) (TCI state ID) and the spatial relationship information (TCI state) may be interchangeable. The "spatial relationship information (TCI state)" may be interchangeable with "set of spatial relationship information (TCI state)", "one or more pieces of spatial relationship information", etc. The TCI state and the TCI may be interchangeable. The spatial relationship information and the spatial relationship may be interchangeable.

[0613] In the present disclosure, terms such as "base station (BS)," "radio base station," "fixed station," "NodeB," "eNB (eNodeB)," "gNB (gNodeB)," "access point," "transmission point (TP)," "reception point (RP)," "transmission / reception point (TRP)," "panel," "cell," "sector," "cell group," "carrier," "component carrier," etc. may be used interchangeably. Base stations may also be referred to by terms such as macrocell, small cell, femtocell, picocell, etc.

[0614] A base station can accommodate one or more (e.g., three) cells. When a base station accommodates multiple cells, the overall coverage area of ​​the base station can be partitioned into multiple smaller areas, and each smaller area can be provided with communication service by a base station subsystem (e.g., a small indoor base station (Remote Radio Head (RRH))). The terms "cell" or "sector" refer to part or all of the coverage area of ​​a base station and / or base station subsystem that provides communication service within that coverage.

[0615] In the present disclosure, a base station transmitting information to a terminal may be interpreted as the base station instructing the terminal to control / operate based on the information.

[0616] In this disclosure, the terms "Mobile Station (MS)," "user terminal," "User Equipment (UE)," "terminal," etc. may be used interchangeably.

[0617] A mobile station may also be referred to as a subscriber station, mobile unit, subscriber unit, wireless unit, remote unit, mobile device, wireless device, wireless communication device, remote device, mobile subscriber station, access terminal, mobile terminal, wireless terminal, remote terminal, handset, user agent, mobile client, client, or some other suitable terminology.

[0618] At least one of the base station and the mobile station may be called a transmitting device, a receiving device, a wireless communication device, etc. Note that at least one of the base station and the mobile station may be a device mounted on a moving object, the moving object itself, etc.

[0619] The mobile body is a movable object that can move at any speed and naturally includes cases where the mobile body is stationary. Examples of the mobile body include, but are not limited to, vehicles, transport vehicles, automobiles, motorcycles, bicycles, connected cars, excavators, bulldozers, wheel loaders, dump trucks, forklifts, trains, buses, handcars, rickshaws, ships and other watercraft, airplanes, rockets, satellites, drones, multicopters, quadcopters, balloons, and objects mounted thereon. The mobile body may also be a mobile body that moves autonomously based on an operation command.

[0620] The mobile object may be a vehicle (e.g., a car, an airplane, etc.), an unmanned mobile object (e.g., a drone, an autonomous vehicle, etc.), or a robot (manned or unmanned). Note that at least one of the base station and the mobile station may also include devices that do not necessarily move during communication operations. For example, at least one of the base station and the mobile station may be an Internet of Things (IoT) device such as a sensor.

[0621] 16 is a diagram showing an example of a vehicle according to an embodiment. The vehicle 40 includes a drive unit 41, a steering unit 42, an accelerator pedal 43, a brake pedal 44, a shift lever 45, left and right front wheels 46, left and right rear wheels 47, axles 48, an electronic control unit 49, various sensors (including a current sensor 50, an RPM sensor 51, an air pressure sensor 52, a vehicle speed sensor 53, an acceleration sensor 54, an accelerator pedal sensor 55, a brake pedal sensor 56, a shift lever sensor 57, and an object detection sensor 58), an information service unit 59, and a communication module 60.

[0622] The drive unit 41 is configured with at least one of an engine, a motor, and a hybrid of an engine and a motor, for example. The steering unit 42 includes at least a steering wheel (also called a handle) and is configured to steer at least one of the front wheels 46 and the rear wheels 47 based on the operation of the steering wheel operated by a user.

[0623] The electronic control unit 49 is composed of a microprocessor 61, memory (ROM, RAM) 62, and a communication port (for example, an input / output (IO) port) 63. Signals are input to the electronic control unit 49 from various sensors 50-58 provided in the vehicle. The electronic control unit 49 may also be called an Electronic Control Unit (ECU).

[0624] The signals from the various sensors 50-58 include a current signal from a current sensor 50 that senses the current of the motor, a rotation speed signal of the front wheels 46 / rear wheels 47 obtained by a rotation speed sensor 51, an air pressure signal of the front wheels 46 / rear wheels 47 obtained by an air pressure sensor 52, a vehicle speed signal obtained by a vehicle speed sensor 53, an acceleration signal obtained by an acceleration sensor 54, a depression amount signal of the accelerator pedal 43 obtained by an accelerator pedal sensor 55, a depression amount signal of the brake pedal 44 obtained by a brake pedal sensor 56, an operation signal of the shift lever 45 obtained by a shift lever sensor 57, and a detection signal for detecting obstacles, vehicles, pedestrians, etc. obtained by an object detection sensor 58.

[0625] The information service unit 59 is composed of various devices, such as a car navigation system, an audio system, speakers, a display, a television, and a radio, for providing (outputting) various information such as driving information, traffic information, and entertainment information, and one or more ECUs for controlling these devices. The information service unit 59 uses information acquired from external devices via the communication module 60 or the like to provide various information / services (e.g., multimedia information / multimedia services) to the occupants of the vehicle 40.

[0626] The information service unit 59 may include input devices (e.g., keyboards, mice, microphones, switches, buttons, sensors, touch panels, etc.) that accept input from the outside, and may also include output devices (e.g., displays, speakers, LED lamps, touch panels, etc.) that output to the outside.

[0627] The driving assistance system unit 64 includes various devices for providing functions to prevent accidents and reduce the driver's driving burden, such as millimeter-wave radar, Light Detection and Ranging (LiDAR), cameras, positioning locators (e.g., Global Navigation Satellite System (GNSS)), map information (e.g., High Definition (HD) maps, Autonomous Vehicle (AV) maps), gyro systems (e.g., Inertial Measurement Units (IMUs), Inertial Navigation Systems (INSs)), artificial intelligence (AI) chips, and AI processors, as well as one or more ECUs that control these devices. The driving assistance system unit 64 also transmits and receives various information via the communication module 60 to realize driving assistance functions or autonomous driving functions.

[0628] The communication module 60 can communicate with the microprocessor 61 and components of the vehicle 40 via the communication port 63. For example, the communication module 60 transmits and receives data (information) via the communication port 63 to and from the drive unit 41, steering unit 42, accelerator pedal 43, brake pedal 44, shift lever 45, left and right front wheels 46, left and right rear wheels 47, axles 48, the microprocessor 61 and memory (ROM, RAM) 62 in the electronic control unit 49, and the various sensors 50-58, which are provided in the vehicle 40.

[0629] The communication module 60 is a communication device that can be controlled by the microprocessor 61 of the electronic control unit 49 and can communicate with an external device. For example, it transmits and receives various information to and from the external device via wireless communication. The communication module 60 may be located either inside or outside the electronic control unit 49. The external device may be, for example, the base station 10 or the user terminal 20 described above. Furthermore, the communication module 60 may be, for example, at least one of the base station 10 and the user terminal 20 described above (or may function as at least one of the base station 10 and the user terminal 20).

[0630] The communication module 60 may transmit at least one of signals from the above-mentioned various sensors 50-58 input to the electronic control unit 49, information obtained based on the signals, and information based on input from the outside (user) obtained via the information service unit 59 to an external device via wireless communication. The electronic control unit 49, the various sensors 50-58, the information service unit 59, etc. may be referred to as input units that accept input. For example, the PUSCH transmitted by the communication module 60 may include information based on the above-mentioned input.

[0631] The communication module 60 receives various information (traffic information, traffic signal information, vehicle distance information, etc.) transmitted from an external device and displays it on an information service unit 59 provided in the vehicle. The information service unit 59 may also be called an output unit that outputs information (for example, outputs information to a device such as a display or speaker based on the PDSCH received by the communication module 60 (or data / information decoded from the PDSCH)).

[0632] Furthermore, the communication module 60 stores various information received from external devices in a memory 62 that can be used by the microprocessor 61. Based on the information stored in the memory 62, the microprocessor 61 may control the drive unit 41, steering unit 42, accelerator pedal 43, brake pedal 44, shift lever 45, left and right front wheels 46, left and right rear wheels 47, axles 48, various sensors 50-58, and the like provided in the vehicle 40.

[0633] Furthermore, a base station in the present disclosure may be read as a user terminal. For example, the aspects / embodiments of the present disclosure may be applied to a configuration in which communication between a base station and a user terminal is replaced with communication between multiple user terminals (which may be called, for example, Device-to-Device (D2D) or Vehicle-to-Everything (V2X)). In this case, the user terminal 20 may be configured to have the functions of the base station 10 described above. Furthermore, terms such as "uplink" and "downlink" may be read as terms corresponding to terminal-to-terminal communication (for example, "sidelink"). For example, terms such as an uplink channel and a downlink channel may be read as a sidelink channel.

[0634] Similarly, the user terminal in the present disclosure may be read as a base station, in which case the base station 10 may be configured to have the functions of the user terminal 20 described above.

[0635] In the present disclosure, an operation described as being performed by a base station may be performed by its upper node in some cases. It is apparent that in a network including one or more network nodes having a base station, various operations performed for communication with a terminal may be performed by the base station, one or more network nodes other than the base station (such as, but not limited to, a Mobility Management Entity (MME), a Serving-Gateway (S-GW), etc.), or a combination thereof.

[0636] Each aspect / embodiment described in this disclosure may be used alone, in combination, or switched depending on the implementation. Furthermore, the order of the processing procedures, sequences, flowcharts, etc. of each aspect / embodiment described in this disclosure may be changed unless inconsistent. For example, the methods described in this disclosure present elements of various steps using an example order, and are not limited to the particular order presented.

[0637] Each aspect / embodiment described in the present disclosure may be a technology other than Long Term Evolution (LTE), LTE-Advanced (LTE-A), LTE-Beyond (LTE-B), SUPER 3G, IMT-Advanced, 4th generation mobile communication system (4G), 5th generation mobile communication system (5G), 6th generation mobile communication system (6G), xth generation mobile communication system (xG (x is, for example, an integer or decimal number)), Future Radio Access (FRA), New-Radio Access Technology (RAT), New Radio (NR), New radio access (NX), Future generation radio access (FX), Global System for Mobile communications (GSM (registered trademark)), CDMA2000, Ultra Mobile Broadband (UMB), IEEE 802.11 (Wi-Fi (registered trademark)), IEEE 802.16 (WiMAX (registered trademark)), IEEE 802.17 (WiMAX (registered trademark)), IEEE 802.19 (WiMAX (registered trademark)), IEEE 802.20 (WiMAX (registered trademark)), IEEE 802.21 (Wi-Fi (registered trademark)), IEEE 802.22 (WiMAX (registered trademark)), IEEE 802.23 (WiMAX (registered trademark)), IEEE 802.24 (WiMAX (registered trademark)), IEEE 802.25 (WiMAX (registered trademark)), IEEE 802.26 (WiMAX (registered trademark)), IEEE 802.27 (WiMAX (registered trademark)), IEEE 802.28 (WiMAX (registered trademark)), IEEE 802.29 (WiMAX (registered trademark)), IEEE 802.30 (WiMAX (registered trademark)), IEEE 802.31 (Wi-Fi (registered trademark)), IEEE 802.32 (WiMAX (registered trademark)), IEEE 802.33 (WiMAX (registered trademark)), IEEE 802. The present invention may be applied to systems that use IEEE 802.20, Ultra-WideBand (UWB), Bluetooth (registered trademark), or other suitable wireless communication methods, or to next-generation systems that are expanded, modified, created, or defined based on these. Furthermore, the present invention may be applied to a combination of multiple systems (e.g., a combination of LTE or LTE-A and 5G).

[0638] As used in this disclosure, the phrase "based on" does not mean "based only on," unless expressly stated otherwise. In other words, the phrase "based on" means both "based only on" and "based at least on."

[0639] As used in this disclosure, any reference to an element using a designation such as "first," "second," etc. does not generally limit the quantity or order of those elements. These designations may be used in this disclosure as a convenient method of distinguishing between two or more elements. Thus, a reference to a first and a second element does not imply that only two elements may be employed or that the first element must in some way precede the second element.

[0640] The term "determining" as used in this disclosure may encompass a wide variety of actions. For example, "determining" may be considered to be judging, calculating, computing, processing, deriving, investigating, looking up, search, inquiry (e.g., looking up in a table, database, or another data structure), ascertaining, etc.

[0641] Additionally, "determining" may be considered to be "determining" receiving (e.g., receiving information), transmitting (e.g., sending information), input, output, accessing (e.g., accessing data in memory), etc.

[0642] Furthermore, "determination" may be considered to be "determining" resolving, selecting, choosing, establishing, comparing, etc. In other words, "determination" may be considered to be "determining" some kind of action. In the present disclosure, "determination" may be read interchangeably with the above-mentioned actions.

[0643] Furthermore, in this disclosure, "determine / determining" may be interchangeably read as "assume / assuming," "expect / expecting," "consider / considering," etc. Furthermore, in this disclosure, "does not expect to do..." may be interchangeably read as "assumes not to do...."

[0644] In the present disclosure, "expect" may be interchangeably read as "be expected." For example, "expect(s) ..." ("..." may be expressed, for example, as a that clause, a to-infinitive, etc.) may be interchangeably read as "be expected ...." "does not expect ..." may be interchangeably read as "be not expected ...." Furthermore, "An apparatus A is not expected ..." may be interchangeably read as "an apparatus B other than apparatus A does not expect ... from apparatus A" (e.g., if apparatus A is a UE, apparatus B may be a base station).

[0645] The "maximum transmit power" in this disclosure may mean the maximum value of transmit power, the nominal UE maximum transmit power, or the rated UE maximum transmit power.

[0646] As used in this disclosure, the terms "connected," "coupled," or any variation thereof, refer to any direct or indirect connection or coupling between two or more elements, and may include the presence of one or more intermediate elements between two elements that are "connected" or "coupled" to each other. The coupling or connection between elements may be physical, logical, or a combination thereof. For example, "connected" may be read as "access."

[0647] In this disclosure, when two elements are connected, they may be considered to be "connected" or "coupled" to one another using one or more wires, cables, printed electrical connections, etc., as well as using electromagnetic energy having wavelengths in the radio frequency range, microwave range, light (both visible and invisible) range, etc., as some non-limiting and non-exhaustive examples.

[0648] In the present disclosure, the term "A and B are different" may mean "A and B are different from each other." The term may also mean "A and B are each different from C." Terms such as "separate" and "coupled" may also be interpreted in the same way as "different."

[0649] When the terms "include," "including," and variations thereof are used in this disclosure, these terms are intended to be inclusive, similar to the term "comprising." Furthermore, when the term "or" is used in this disclosure, it is not intended to be an exclusive or.

[0650] In this disclosure, where articles are added by translation, such as a, an, and the in English, the disclosure may include that the nouns following these articles are in the plural form.

[0651] In the present disclosure, terms such as "less than or equal to," "less than," "greater than," "more than," "equal to," etc. may be interchangeable. Furthermore, in the present disclosure, terms meaning "good," "bad," "big," "small," "high," "low," "fast," "slow," "wide," "narrow," etc. may be interchangeable, not limited to the positive, comparative, and superlative. Furthermore, in the present disclosure, terms meaning "good," "bad," "big," "small," "high," "low," "fast," "slow," "wide," "narrow," etc. may be interchangeable, not limited to the positive, comparative, and superlative, as expressions with "i-th" (i is an arbitrary integer) attached (for example, "highest" may be interchangeable with "i-th highest").

[0652] In this disclosure, the terms "of," "for," "regarding," "related to," "associated with," etc. may be read interchangeably.

[0653] In the present disclosure, terms such as "when A, B," "if A, (then) B," "B upon A," "B in response to A," "B based on A," "B during / while A," "B before A," "B at (the same time as) / on A," "B after A," "B since A," and "B until A" may be interchangeable. Note that A, B, and the like herein may be replaced with appropriate expressions such as nouns, gerunds, and regular sentences, depending on the context. Note that the time difference between A and B may be approximately zero (immediately after or immediately before). A time offset may also be applied to the time at which A occurs. For example, "A" may be interchangeable with "before / after a time offset at which A occurs." The time offset (eg, one or more symbols / slots) may be predefined or may be specified by the UE based on signaled information.

[0654] In the present disclosure, timing, time, duration, time instance, any time unit (e.g., slot, subslot, symbol, subframe), period, occasion, resource, etc. may be read interchangeably.

[0655] Although the invention according to the present disclosure has been described in detail above, it is clear to those skilled in the art that the invention according to the present disclosure is not limited to the embodiments described in the present disclosure. The description of the present disclosure is for illustrative purposes only and does not impose any limiting meaning on the invention according to the present disclosure.

Claims

1. A receiving unit that receives information regarding a fallback trigger condition for lifecycle management applied to a specific use case of an artificial intelligence (AI) model and specific conditions for recovery from the fallback, and a control unit that controls triggering of a fallback operation based on the trigger condition and controls an operation after the fallback operation based on the specific conditions.

2. The terminal according to claim 1, wherein the specific condition is receiving a predetermined number of synchronization instructions from a lower layer.

3. The terminal according to claim 1, wherein the control unit determines whether to execute reactivation of functionality or a model based on the specific conditions.

4. The terminal according to claim 1, wherein the specific use case is either AI-based channel state information (CSI) feedback or beam management.

5. A wireless communication method of a terminal, comprising: receiving information regarding a fallback trigger condition for lifecycle management applied to a specific use case of an artificial intelligence (AI) model and specific conditions for recovery from the fallback; and controlling triggering of a fallback operation based on the trigger condition and controlling an operation after the fallback operation based on the specific conditions.

6. A base station, comprising: a transmitting unit that transmits information regarding a fallback trigger condition for lifecycle management applied to a specific use case of an artificial intelligence (AI) model and specific conditions for recovery from the fallback; and a control unit that controls reception of a report of a fallback operation triggered by a terminal based on the trigger condition and an operation after the fallback operation triggered by the terminal based on the specific conditions.

Images