Method for performance monitoring and device therefor

Abstract

A method according to one embodiment of the present specification comprises the steps of: receiving configuration information related to CSI; transmitting a first CSI report related to a first report configuration; and transmitting a second CSI report related to a second report configuration. The second CSI report includes fields related to SGCS values. A code point of each of the fields indicates a range related to SGCS quantization. The code point is one among a plurality of code points, and the plurality of code points include code points mapped to a plurality of ranges. The plurality of ranges include ranges defined by quantizing, on the basis of a step-size, a value range not smaller than a predefined value and not greater than 1.

Description

Method and apparatus for performance monitoring

[0001] This specification relates to a method and apparatus for performance monitoring.

[0002] Mobile communication systems were developed to provide voice services while ensuring user mobility. However, mobile communication systems have expanded their scope to include data services as well as voice. Currently, due to the explosive increase in traffic leading to resource shortages and users demanding higher-speed services, more advanced mobile communication systems are required.

[0003] The requirements for next-generation mobile communication systems largely include the ability to accommodate explosive data traffic, a dramatic increase in transmission rates per user, a significantly increased number of connected devices, very low end-to-end latency, and high energy efficiency. To achieve this, various technologies are being researched, such as dual connectivity, massive multiple input multiple output (MMIMO), in-band full duplex, non-orthogonal multiple access (NOMA), super wideband support, and device networking.

[0004] Performance monitoring for CSI prediction can be performed. Specifically, the terminal can transmit a CSI report related to prediction accuracy. The CSI report may include performance metric(s).

[0005] Meanwhile, regarding the aforementioned performance metric, it was agreed that the terminal calculates and reports Square Generalized Cosine Similarity (SGCS). However, since it has not yet been clearly defined how the calculated SGCS value should be reported, the following problems may arise. Specifically, as the UCI payload for CSI reporting is limited, reporting information that directly represents the SGCS value, which is a value between 0 and 1, may be inefficient in terms of signaling overhead.

[0006] The purpose of this specification is to propose a method for solving the aforementioned problems.

[0007] The technical problems to be solved in this specification are not limited to those mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art to which this specification belongs from the description below.

[0008] A method according to one embodiment of the present specification for solving the aforementioned problem comprises the steps of receiving configuration information related to channel state information (CSI), transmitting a first CSI report related to a first reporting configuration, and transmitting a second CSI report related to a second reporting configuration. The configuration information includes i) the first reporting configuration related to CSI prediction and ii) the second reporting configuration related to CSI prediction accuracy. The second CSI report includes fields related to Square Generalized Cosine Similarity (SGCS) values. Each code point of the fields represents a range related to SGCS quantization. The code point is one of a plurality of code points, and the plurality of code points include code points mapped to a plurality of ranges. The plurality of ranges are characterized by including ranges defined by quantizing a value range of greater than or equal to 1 based on a step size.

[0009] Therefore, by utilizing defined ranges based on quantization of a predefined range of values ​​from 1 to 1, the SGCS value is reported, thereby resolving ambiguity related to the reporting method of the SGCS value and reducing signaling overhead.

[0010] According to an embodiment of the present specification, reporting is performed by utilizing defined ranges by quantizing a range of values ​​from a predefined value to 1 or less based on a step-size, thereby reducing the number of bits required for performance metric reporting and reducing overhead.

[0011] In addition, since minute changes in SGCS values ​​can be absorbed into the same range, reporting stability can be improved, and the network can interpret reported values ​​based on more consistent and simple criteria.

[0012] In addition, by selectively providing precise distinctions based on step size only for specific value ranges, a balance between information representation efficiency and system performance can be secured.

[0013] The effects obtainable in this specification are not limited to those mentioned above, and other unmentioned effects will be clearly understood by those skilled in the art to which this specification belongs from the description below.

[0014] Figure 1 is a diagram illustrating the overall functions from the perspective of an AI / ML model.

[0015] Figure 2 illustrates a general form of an AI / ML-related procedure performed between a network and a terminal.

[0016] Figure 3 illustrates an example of AI / ML-based beam management operation.

[0017] Figure 4 illustrates an example of an AI / ML-based CSI measurement / reporting operation.

[0018] Figure 5 illustrates an example of an AI / ML-based positioning operation.

[0019] Figure 6 is a flowchart showing an example of a CSI-related procedure.

[0020] Figure 7 illustrates an AI / ML-based CSI reporting operation.

[0021] Figure 8 illustrates AI / ML-based CSI prediction.

[0022] Figure 9 illustrates a signaling procedure based on an AI / ML model.

[0023] Figure 10 is a diagram illustrating a mismatch between the actual ground-truth CSI and the prediction instance.

[0024] FIG. 11 is a flowchart illustrating a method according to one embodiment of the present specification.

[0025] FIG. 12 is a flowchart illustrating a method according to another embodiment of the present specification.

[0026] FIG. 13 is a drawing showing the configuration of a first device and a second device according to an embodiment of the present specification.

[0027] In this specification, "A or B" may mean "only A," "only B," or "both A and B." Alternatively, in this specification, "A or B" may be interpreted as "A and / or B." For example, in this specification, "A, B or C" may mean "only A," "only B," "only C," or "any combination of A, B and C."

[0028] A slash ( / ) or a comma used in this specification may mean "and / or." For example, "A / B" may mean "A and / or B." Accordingly, "A / B" may mean "only A," "only B," or "both A and B." For example, "A, B, C" may mean "A, B or C."

[0029] In this specification, "at least one of A and B" may mean "only A," "only B," or "both A and B." Additionally, in this specification, the expressions "at least one of A or B" or "at least one of A and / or B" may be interpreted as synonymous with "at least one of A and B."

[0030] Additionally, in this specification, "at least one of A, B and C" may mean "only A," "only B," "only C," or "any combination of A, B and C." Also, "at least one of A, B or C" or "at least one of A, B and / or C" may mean "at least one of A, B and C."

[0031] Additionally, parentheses used in this specification may mean "for example." Specifically, when indicated as "control information (PDCCH)," "PDCCH" may be proposed as an example of "control information." In other words, "control information" in this specification is not limited to "PDCCH," and "PDCCH" may be proposed as an example of "control information." Furthermore, even when indicated as "control information (i.e., PDCCH)," "PDCCH" may be proposed as an example of "control information."

[0032] In the following explanation, 'when, if, in case of' can be replaced with 'based on'.

[0033] Technical features described individually within a single drawing in this specification may be implemented individually or simultaneously.

[0034] Hereinafter, preferred embodiments according to the present specification will be described in detail with reference to the accompanying drawings. The detailed description disclosed below, together with the accompanying drawings, is intended to describe exemplary embodiments of the present specification and is not intended to represent the only embodiment in which the present specification may be practiced. The following detailed description includes specific details to provide a complete understanding of the present specification.

[0035] In this specification, a terminal is a user-side device (user equipment, UE) or a consumer-side device, and may also be referred to as a first node that receives / transmits signals from / to a base station / second node / IAB node / Transmission-Reception Point (TRP). A terminal may correspond to a physical node or a logical node. A terminal may correspond to a user-side endpoint or an intermediate point between other endpoints. In communication between two points not limited to endpoints (including one-to-one / many-to-one / one-to-many / many-to-many communication), a terminal may correspond to a served node. A terminal may be a fixed-location node or a non-fixed-location (or mobile) node.

[0036] In this specification, a Base Station (BS) is a device on the network side and may also be referred to as a second node / IAB node / x-NodeB (x-NodeB, where x may be an abbreviation related to Radio Access Technology (RAT)) / Transmission-Reception Point (TRP). A Base Station may correspond to a physical node or a logical node. A Base Station may correspond to an endpoint on the network side or an intermediate point between other endpoints. In communication between two points not limited to endpoints (including one-to-one / many-to-one / one-to-many / many-to-many communication), a Base Station may correspond to a serving node. A Base Station may be a node with a fixed location or a node with an indefinite location.

[0037] In this specification, higher layer parameters may be set for the terminal, pre-set, or pre-defined. For example, a base station may transmit higher layer parameters to the terminal. For example, the terminal may transmit parameters such as capability to the base station as higher layer parameters. For example, higher layer parameters may be transmitted via RRC (radio resource control) signaling or MAC (medium access control) signaling.

[0038] In this specification, information / state / parameters being "configured" or "pre-configured" may be interpreted as the information / state / parameters being provided / pre-provided to the terminal through pre-defined signaling (e.g., SIB, MAC, RRC) from the base station. In this specification, information / state / parameters being "defined" or "pre-defined" may be interpreted as being known or stored in advance by the base station and the terminal without signaling between the base station and the terminal.

[0039] < AI / ML for Wireless Communication >

[0040] With the advancement of computing technology, artificial intelligence (AI) and machine learning (ML) are being adopted across various industries and technological fields. In the field of wireless communication, various discussions are underway regarding the application of AI models trained on ML; notably, the 3GPP standardization process refers to this as AI / ML. In this specification, we use the term "AI / ML" following the terminology currently in use during the 3GPP standardization discussions; however, "AI / ML" may be referred to by various other terms depending on the progress of standardization and implementation in the future. For example, it may be referred to as a "transmission / reception mode" or a "signal / channel / operation / transmission / reception configuration" set for AI / ML, but is not limited thereto. The meanings of the terms currently used in the 3GPP standardization process are briefly summarized as follows.

[0041] - AI / ML Model: Refers to a data-driven algorithm that applies AI / ML technology to generate a set of outputs containing predictive information and / or decision parameters based on a set of inputs.

[0042] - Data collection: This is the process of collecting data necessary for AI / ML model training, data analysis, and inference from network nodes, management entities, or terminals.

[0043] - AI / ML Training: An online or offline process of training an AI model by learning features and patterns that best represent data and acquire an AI / ML model trained for inference.

[0044] - Offline training: A process of training a model based on previously collected data sets, where the trained model is used or provided for future inference.

[0045] - Online Training: A method in which the model is trained in real-time upon the acquisition of new training sample data and used for inference.

[0046] - AI / ML Inference: This is the process of making predictions or deriving decisions based on collected data and AI models using trained AI models. Meanwhile, depending on whether the AI / ML model is configured on both the transmitting and receiving devices or on only one, it can be classified into (i) two-sided models and (ii) one-sided models. In the case of (i) two-sided models, cooperative inference is performed through paired AI / ML models. Cooperative inference refers to cooperation between the network and the UE, where one side performs part of the inference and the other performs the remainder. (ii) One-sided models are divided into UE-side models and network-side models. In the case of one-sided models, inference is performed entirely by the UE / network-side models.

[0047] 1. Life Cycle Management (LCM) for AI / ML models

[0048] For AI / ML models, LCM is a concept that encompasses all overall procedures for the model, such as data collection, model training, model deployment, model inference, model monitoring, and model updates.

[0049] LCMs for AI / ML models can be broadly classified into functionality-based LCMs and model-ID-based LCMs. In functionality-based LCMs, the network can direct activation, deactivation, fallback, or switching for specific functions; in this case, the target AI / ML model may not be identified by the network. In model-ID (identifier)-based LCMs, the network can direct activation, deactivation, selection, or switching for AI / ML models identified based on their AI / ML model IDs.

[0050] Figure 1 is a diagram illustrating the overall functions from the perspective of an AI / ML model.

[0051] Referring to FIG. 1, a general AI / ML functional framework can be configured to include a data collection function (10), a model training function (20), a management function (30), an inference function (40), and a model storage function (50).

[0052] The Data Collection function (10) is a function that provides input data to the Model Training function (20), Management function (30), and Inference function (40). The Data Collection function (10) performs data preparation and can provide input data processed through data preparation.

[0053] Here, training data (11) refers to data required as input for the AI / ML model training function (20). monitoring data (12) refers to data required as input for the management (30) of the AI / ML model or AI / ML function. inference data (13) refers to data required as input for the AI / ML inference function (30).

[0054] The Model Training function (20) is a function that performs AI / ML model training, validation, and testing, and can generate model performance metrics that can be used as part of the AI / ML model testing procedure. If necessary, the Model Training function (20) can perform data preparation (e.g., data pre-processing and cleaning, forming and transformation) based on the Training Data (11) delivered from the Data Collection function (10).

[0055] Trained / Updated Model (21): If there is a Model Storage function (50), it is used to transfer trained, validated, and tested AI / ML models to the Model Storage function (50) or to transfer updated versions of the models to the Model Storage function (50).

[0056] The Management function (30) is a function that monitors the operation of an AI / ML model or an AI / ML function. Additionally, the Management function (30) may make decisions to ensure appropriate inference operations based on data received from the Data Collection function (10) (i.e., Monitoring Data (12)) and / or data received from the Inference function (40) (i.e., Inference Output (41)).

[0057] Management Instruction (32) is information required as input to manage the Inference function (40). The relevant information may include the selection / (de)activation / switching of an AI / ML model or an AI / ML-based function, and may also include a fallback to a non-AI / ML operation (i.e., not relying on the inference process).

[0058] A Model Transfer / Delivery Request (33) can be used to request model(s) from Model Storage (50).

[0059] Performance Feedback / Retraining Request (31) refers to information required as input to Model Training function (20) (e.g., for the purpose of retraining or updating the model).

[0060] The inference function (40) is a function that provides output from the process of applying an AI / ML model or AI / ML function using data (i.e., inference data (13)) provided by the data collection (10) as input. Data preparation (e.g., data preprocessing and cleaning, formatting and transformation) may also be performed based on the inference data (13) delivered by the data collection (10). If necessary, the inference function (40) may also perform data preparation (e.g., data pre-processing and cleaning, forming and transformation) based on the inference data (13) provided by the data collection function (10).

[0061] Inference Output (41) is data used in the Management function (30) to monitor the performance of an AI / ML model or AI / ML function. Inference Output (41) may include the inference output of an AI / ML model generated by the Inference function (30), and the details of the inference output may vary depending on the use case.

[0062] The Model Storage function (50) is a function that stores a trained / updated model that can be used to perform the Inference function (40). The Model Storage function (50) exemplified in FIG. 1 can be used as a reference point (if any) applicable to protocol termination, model transmission / delivery, and related processes. Additionally, the Model Storage function (50) is an example and is not intended to restrict the storage location of the actual AI / ML model, and may be omitted.

[0063] Model Transfer / Delivery (51) is used to transfer an AI / ML model to an inference function.

[0064] 2. General AI / ML related procedures between the network and the terminal

[0065] Figure 2 illustrates the general form of AI / ML-related procedures performed between a network and a terminal. While Figure 1 examined the LCM from the perspective of an AI / ML model, Figure 2 describes the general form of procedures performed between a terminal and a network from the perspective of signaling / protocols.

[0066] (1) AI / ML related setup procedure

[0067] Referring to FIG. 2, an AI / ML-related configuration procedure may be performed between the network and the terminal (B05). The AI / ML-related configuration procedure may include information exchange through at least one upper-layer signaling between the terminal and the network, and / or prior preparation / subsequent operations at the terminal / network respectively before / after the upper-layer signaling.

[0068] Specifically, the configuration procedure related to AI / ML may include, but is not limited to, at least one of the following: (i) reporting the capability of the AI / ML-related terminal, (ii) data collection, (iii) model training, (iv) model delivery / transmission, (v) selection of AI / ML functions / models, and (vi) configuration of various operations performed based on the AI / ML model (e.g., AI / ML-based CSI / Positioning / Beam Management).

[0069] (i) The terminal can inform the network of its capabilities, such as models and functions related to AI / ML, that it supports through UE Capability reporting. The network can provide AI / ML-related settings to the terminal based on the terminal's capabilities related to AI / ML reported by the terminal.

[0070] (ii) AI / ML-related configuration procedures may include data collection related to the training / inference of AI / ML models and / or the provision of configuration information regarding data collection. The configuration information regarding data collection may relate to how to configure the method / operation of data collection.

[0071] (iii) AI / ML-related configuration procedures may include training AI / ML models online or offline and / or providing configuration information for AI / ML model training. The configuration information for AI / ML model training may relate to how to configure the method / behavior, etc., of training the AI / ML model.

[0072] (iv) AI / ML-related configuration procedures may include transmitting / transmitting configuration information for a model. The configuration information for a model may include parameters that constitute the AI / ML model and / or an identifier (ID) for the AI / ML model.

[0073] The provided AI / ML model may be a model trained by the network or a model that requires self-training at the terminal. Even when a model trained by the network is provided, the terminal may perform fine-tuning or retraining as necessary. Meanwhile, if a model trained by the network is provided, the terminal may provide data for training to the network.

[0074] Meanwhile, AI / ML models can be classified into Type A models, which can be identified without OTA (over-the-air) signaling, and Type B models, which are identified through OTA signaling. A model ID may be assigned during the model identification process, and this process can be subdivided into methods initiated by the terminal and methods identified by the network.

[0075] (v) The configuration procedure related to AI / ML may include the configuration of how to select AI / ML Functionality / models and / or the selection process for AI / ML Functionality / models. In UE-side AI / ML models or two-sided AI / ML models, the selection of the UE part may be performed through instructions / signaling from the network or the terminal may select it itself. The selection of AI / ML Functionality / models may be performed when multiple AI / ML Functionality / models are configured / provided.

[0076] (vi) The configuration procedure related to AI / ML may include configuration information for various inference operations performed based on AI / ML models, e.g., AI / ML-based CSI measurement / reporting, AI / ML-based positioning, and / or AI / ML-based beam management.

[0077] (2) Operation based on inference by AI / ML models

[0078] Referring again to FIG. 2, the network and / or terminal can perform inference of the AI / ML model through the trained AI / ML model and perform various operations based on the inference of the AI / ML model (B10). If the AI / ML model is a one-sided model, the inference of the AI / ML model can be performed at either the network or the terminal where the AI / ML model is configured. If the AI / ML model is a two-sided model, each part of the inference of the AI / ML model can be performed at the network and the terminal, and depending on the implementation, such inference can be performed cooperatively between the network and the terminal.

[0079] (i) Actions performed based on the inference of an AI / ML model may include AI / ML-based CSI measurement / reporting. AI / ML-based CSI measurement / reporting is intended to improve CSI feedback and may be related to overhead reduction / CSI compression, accuracy improvement, and / or CSI prediction.

[0080] (ii) Actions performed based on the inference of an AI / ML model may include AI / ML-based beam management. AI / ML-based beam management may be related to beam prediction in the time domain, reduction of overhead / latency in the spatial domain, and / or improvement of beam selection accuracy.

[0081] (iii) Actions performed based on the inference of an AI / ML model may include AI / ML-based positioning. AI / ML-based positioning may be relevant to improving positioning accuracy in various scenarios, for example, in non-line-of-sight environments.

[0082] (3) Procedures for AI / ML management

[0083] The network and / or terminal can perform procedures for the management of AI / ML Functionality / model or the settings therefor (B15).

[0084] The network and / or terminal may perform monitoring of AI / ML Functionality / model during the AI / ML model inference or operation based thereon (B10) for the management procedure (B15).

[0085] The management procedure may include, for example, at least one of activation / deactivation, switching, model update, and / or fallback operation for AI / ML Functionality / model. For the signaling of the management procedure, various 3GPP signaling schemes, such as RRC, MAC-CE, DCI, etc., may be used.

[0086] As an example of model switching, multiple model groups are configured, and switching between them can be performed based on models having a common model structure or partially common substructures, and models within the same group may be associated with different input / output formats or processing.

[0087] Model updating involves modifying the parameters used by the model to suit channel conditions that change over time, and fine-tuning is an example of model updating.

[0088] Fallback: In a wireless communication system using an AI / ML model, this may refer to the operation of not using the AI / ML model or operating in a pre-configured / defined default mode when the reliability of the AI / ML model decreases due to internal or external environmental factors.

[0089] For example, the decision on whether to perform a management procedure can be made by the network. For instance, the network may decide to perform the management procedure upon network initiation, or the network may decide to perform the management procedure upon terminal initiation and request.

[0090] As another example, the decision on whether to perform a management procedure can be made by the terminal. For instance, the terminal's decision on the management procedure may be triggered when an event condition set by the network is satisfied, performed by reporting the terminal's decision to the network, or performed autonomously by the terminal.

[0091] 3. Specific operation examples based on AI / ML model inference

[0092] (1) Beam management

[0093] Figure 3 illustrates an example of AI / ML-based beam management operation.

[0094] Referring to FIG. 3, the network / terminal can perform a configuration procedure related to AI / ML-based beam management (C05). The network / terminal can perform a configuration procedure for an AI / ML model to be used for AI / ML-based beam management, and an exchange of configuration information for upper-layer signaling for AI / ML-based beam management. For example, at least one of information related to model inference, configuration for a first set / second set beam, monitoring performance, and assistance information for data collection and beam measurement may be signaled.

[0095] The network / terminal can perform measurements on the first set of beams (C10). The beam measurements may be related to RSRP measurements.

[0096] A network / terminal can obtain information about a second set of beams based on measurement results for a first set of beams (C15). For example, the network / terminal can perform AI / ML inference by using the measurement results for the first set of beams as AI / ML input data. Beam ID information may also be additionally provided as AI / ML input data. Information about the second set of beams may correspond to AI / ML output data. The AI / ML output data may be related to, for example, the probability that each beam will become a top-N beam, the predicted RSRP, etc., for predicting future beam quality, but is not limited thereto.

[0097] According to an embodiment, the network / terminal can transmit and receive information about the acquired second set of beams.

[0098] Specifically, AI / ML-based beam management operations may include at least one of the following BM-Case 1 and BM-Case 2.

[0099] - BM-Case 1: Prediction of the second set of DL beams in the spatial domain through the first set of beam measurements

[0100] - BM-Case 2: Prediction of the second set of DL beams in the time domain through the first set of beam measurements

[0101] In BM-Case 1 and / or 2, both AI / ML model training and inference may be performed on the network or on the terminal. The first set of beams and the second set of beams may be different beams. Or the first set of beams may be a subset of the second set of beams. Or, particularly in BM-Case 2, the first set of beams and the second set of beams may be the same beam.

[0102] The report corresponding to the inference of the UE-side model for BM-Case 1 may relate to the RSRP for the predicted top N beams. The report may include, for example, the predicted RSRP values, and as an example, the predicted RSRP values ​​may be reported together with the actual measured RSRP.

[0103] UE-side AI / ML model inference for BM-Case 2 can report inference results for N future time points through a single report. The report for each time point can correspond to the report in BM-Case 1.

[0104] For performance monitoring of the UE-side model for BM-Case 1 / 2, (i) network-side performance monitoring and / or (ii) UE-assisted performance monitoring may be supported. (i) For network-side performance monitoring, the terminal may report information necessary for the network to calculate performance metrics, for example, by reporting measurement results (e.g., RSRP) and / or RS index for a set of resources for monitoring. (ii) For UE-assisted performance monitoring, the terminal may calculate performance metrics.

[0105] With respect to the NW-side model for BM-Case 1 / 2, quantization of the reported RSRP may be supported, for example, differential RSRP reporting may be supported along existing quantization steps and ranges. The reported content may include information on the RSRP and the corresponding upper N beam, where N can be set by the network.

[0106] With respect to the configuration of the first set of beams and the second set of beams of the UE-side model of BM Case-1, two resource sets may be configured separately for each of the first set and the second set, and the resource sets may be provided through CSI reporting settings. The terminal may perform inference / measurement on the resource set of the first set of beams. The terminal may not be expected to perform measurement / inference on the resource set of the second set of beams. The beam information in the inference report may include resource set information for the first set.

[0107] In relation to the UE-side model, the associated ID may be provided through the CSI framework. The terminal may assume identical / similar characteristics for DL ​​transmit beams / sets (lists) for the same associated ID.

[0108] Regarding UE-assisted performance monitoring for the UE-side models of BM-Case 1 and 2, the following methods may be considered.

[0109] i) Compare prediction results based on resources for monitoring and use the top 1 or top K beam prediction accuracy.

[0110] ii) Use RSRP difference information based on RSRP measurements of resources for monitoring and actual RSRP measurements for at least one of the top N prediction beams.

[0111] iii) Use the difference information between the measured RSRP and the predicted RSRP for the corresponding beam of the resources for monitoring.

[0112] iv) Probability information that the predicted beam will become one of the top 1 or N beams

[0113] For reporting inference results for the UE-side model, quantization of RSRP may be supported, and differential RSRP with existing quantization steps may be supported. The scope of RSRP reporting is such that differential RSRP among multiple beams is supported in the case of BM-case 1, and differential RSRP among multiple beams at multiple time points is supported in the case of BM-case 2.

[0114] For BM-Case 2 of the UE-side model, the network can be configured to report inferences about N future times to the terminal.

[0115] (2) CSI prediction and / or compression

[0116] Figure 4 illustrates an example of an AI / ML-based CSI measurement / reporting operation.

[0117] Referring to FIG. 4, the network / terminal can perform a configuration procedure related to AI / ML-based CSI (D05). The network / terminal can perform a configuration procedure for an AI / ML model to be used for AI / ML-based CSI, and for the exchange of configuration information for upper-layer signaling for AI / ML-based CSI measurement / reporting. For example, at least one of information related to model inference, settings for RS / resources to be used for CSI measurement, monitoring performance, data collection, and conditions / resources for CSI reporting may be signaled.

[0118] The terminal can perform CSI measurements based on AI / ML model inference (D10). The AI / ML model used by the terminal for CSI measurements may be a UE-side AI / ML model corresponding to a one-side AI / ML model, or an AI / ML model corresponding to the terminal part of a two-side AI / ML model.

[0119] The terminal may report CSI to the network based on the results of CSI measurements (D15). CSI reporting may be performed periodically or non-periodically depending on the configuration, and in the case of non-period CSI reporting, network instructions (not shown), such as DCI, that trigger it may be additionally signaled. CSI reporting may include AI / ML-based CSI content and may additionally include legacy CSI content (e.g., non-AI / ML-based RI, PMI, CQI, etc.) (depending on the configuration / scheduling). AI / ML-based CSI content may be related to at least one of 1) CSI compression to reduce the overhead of CSI reporting and 2) CSI prediction for future time points in the time domain.

[0120] The network can acquire CSI based on the terminal's CSI report.

[0121] If a two-sided AI / ML model is configured, the network can reconstruct the CSI by using the terminal's CSI report as input data to the AI / ML model configured in the network (D20). The inference (output) of the AI / ML model configured in the network may be the reconstructed CSI. In such a two-sided AI / ML model, the terminal-side AI / ML model part can be understood as a CSI encoder, and the network-side AI / ML model part can be understood as a concept similar to a CSI decoder.

[0122] CSI compression is CSI compression in the spatial-frequency domain and can primarily be based on two-sided AI / ML models. CSI prediction can primarily be based on one-sided, specifically UE-side AI / ML models.

[0123] In CSI compression based on a two-sided AI / ML model, AI / ML model training may include at least one of the following: (i) Type 1, in which the two-sided AI / ML model is jointly trained at either the terminal or the network; (ii) Type 2, in which the terminal and the network each jointly train the corresponding parts of the two-sided AI / ML model; and (iii) Type 3, in which the terminal and the network each separately train the corresponding parts of the two-sided AI / ML model, wherein the training of the terminal is mainly related to CSI generation and the training of the network is mainly related to CSI reconstruction. Joint training means that the CSI generation / reconstruction models are trained in the same loop for forward / backward delays, and separate training may mean a sequential method in which one of the terminals or the network starts training first, and then the other performs training.

[0124] (3) Positioning

[0125] Figure 5 illustrates an example of an AI / ML-based positioning operation.

[0126] Referring to FIG. 5, the network / terminal can perform a configuration procedure related to AI / ML-based positioning (E05). The network / terminal can perform a configuration procedure for an AI / ML model to be used for AI / ML-based positioning, and an exchange of configuration information for upper-layer signaling for AI / ML-based positioning. For example, at least one of information related to model inference, configuration for positioning RS, monitoring performance, and assistance information for data collection and positioning measurement may be signaled.

[0127] The network / terminal can perform a measurement for positioning (E10). The measurement for positioning may be related to PRS and / or SRS measurements.

[0128] The network / terminal can obtain information about terminal positioning based on measurement results (E15). For example, the network / terminal can perform AI / ML inference by using measurement results for PRS / SRS as AI / ML input data. Information about terminal positioning may correspond to AI / ML output data. The AI / ML output data may be, for example, terminal location or assistance information that serves as the basis for determining terminal location, but is not limited thereto.

[0129] According to an embodiment, the network / terminal can transmit and receive information regarding the acquired terminal positioning.

[0130] Specifically, AI / ML-based positioning may include at least one of (i) direct AI / ML positioning and / or (ii) AI / ML-assisted positioning. In (i) direct AI / ML positioning, the output of the AI / ML model includes the UE location. In (ii) AI / ML-assisted positioning, the output of the AI / ML model may be a new measurement result and / or an improvement on an existing measurement (e.g., LoS / NLoS identification, timing and / or angle of the measurement, likelihood of the measurement, etc.).

[0131] The data sample collected for training data collection may include at least one of the following parts.

[0132] - Part A: Channel Measurement, Quality Indicators of Channel Measurement, Time Stamps of Channel Measurement

[0133] - Part B: ground truth label (or its approximation), label quality indicators, label timestamp

[0134] The training data samples in Part A and Part B are for the same terminal (e.g., PRU or Non-PRU UE) and may be for the same location associated with Part B.

[0135] Measurements for positioning can be classified into sample-based measurements and path-based measurements.

[0136] - In sample-based measurements, the measurement consists of Nt' samples of the channel response estimated in the time domain. Timing information for the Nt' samples is reported as the timing granularity T, where T = 2k x Tc. k represents the timing reporting granularity factor, and Tc represents the base time unit of the corresponding wireless communication system. Nt' and k may be signaling parameters. Timing information can be defined as a value relative to a reference time.

[0137] - Path-based measurement refers to a measurement defined in existing wireless communication systems.

[0138] AI / ML-based positioning may be related to at least one of the following detailed cases.

[0139] - (i) Case 1: UE-based positioning using a UE-side model, in the case of direct AI / ML or AI / ML assisted positioning

[0140] - (ii) Case 2a: UE-assisted / LMF-based positioning using a UE-side model, in the case of AI / ML-assisted positioning

[0141] - (iii) Case 2b: UE-assisted / LMF-based positioning using an LMF-side model, in the case of direct AI / ML positioning

[0142] - (iv) Case 3a: NG-RAN node assisted positioning using a gNB-side model, in the case of AI / ML assisted positioning

[0143] - (v) Case 3b: NG-RAN node-assisted positioning using an LMF-side model, in the case of direct AI / ML positioning

[0144] (i) Regarding model performance monitoring in Case 1, the following options may be considered for calculating model performance metrics in label-based model monitoring.

[0145] 1) Option A. Calculate monitoring metrics on the target terminal side

[0146] - Option A-1: ​​At least part of the information regarding the ground truth label of the target terminal is generated in the LMF and provided to the target terminal. For example, the target terminal and / or base station sends measurement results to the LMF, and the LMF can derive information regarding the ground truth label from this.

[0147] - Option A-2: At least some of the information regarding location calculation assistance data is provided from the LMF to the target UE.

[0148] - Option A-3: As a method to reuse assistance data previously provided from the LMF to the target terminal, the PRU measurement results and the corresponding PRU location information are provided from the LMF to the target terminal.

[0149] - Option A-4: PRU measurement and PRU location are provided from the PRU to the target UE.

[0150] 2) Option B. LMF calculates monitoring metrics

[0151] - Option B-1: At least the inference result of the target terminal (i.e., the model output corresponding to the channel measurement of the target terminal) can be transmitted to the LMF by the target terminal.

[0152] - Option B-2: Channel measurements of the PRU are provided to the target terminal through the LMF, and inference results (i.e., model output corresponding to the channel measurements of the PRU) can be transmitted from the target terminal to the LMF.

[0153] (iv) Regarding the generation of training data in Case 3a, at least LMF can generate labels and related data (e.g., time stamp).

[0154] (iv) For calculating model performance monitoring metrics in label-based model monitoring in Case 3a, the following options A and B may be considered.

[0155] - Option A: The NG-RAN node performs monitoring metric calculations for its model

[0156] - Option B: LMF performs monitoring metric calculations for models located on the NG-RAN node.

[0157] (iv) For generating training data for Case 3a and (v) Case 3b, measurements and related data (e.g., timestamps) may be generated at TRP / base stations.

[0158] (v) For base station channel measurements reported to the LMF (location management server) in Case 3b, timing information can be expressed as a value relative to the UL RTOA reference time T0+tSRS.

[0159] (iii) Time domain channel measurements supported for reporting in Case 2b and (v) Case 3b may include timing information and / or power information corresponding to the timing information.

[0160] For the definition of a sample-based measurement, Nt' samples may be selected from a list of consecutive Nt samples, and the Nt samples may have a particle size at time T.

[0161] In relation to sample-based measurements, the Nt' samples selected for network measurement may be those with the highest power.

[0162] Regarding positioning based on Case 3b, for sample-based measurements, LMF can signal parameter values ​​such as Nt, Nt', and k to the base station.

[0163] < CSI Related Operations >

[0164] Channel state information (CSI) may include at least one of a channel quality indicator (CQI), a precoding matrix indicator (PMI), a CSI-RS resource indicator (CRI), an SS / PBCH block resource indicator (SSBRI), a layer indicator (LI), a rank indicator (RI), Layer 1-RSRP (Layer 1-Reference Signal Received Power), and / or Layer 1-SINR (Layer 1-Signal-to-Interference-plus-Noise Ratio).

[0165] In the case of the CSI prediction described below, the CSI related to the prediction may include at least one of the predicted CQI (predicted CQI, P-CQI), predicted PMI (predicted PMI, P-PMI), predicted CRI (predicted CRI, P-CRI), predicted SSBRI (predicted SSBRI, P-SSBRI), predicted LI (predicted LI, P-LI), predicted RI (predicted RI, P-RI), predicted L1-RSRP (predicted L1-RSRP, P-L1-RSRP) and / or predicted L1-SINR (predicted L1-SINR, P-L1-SINR).

[0166] When monitoring the performance / accuracy of the CSI prediction described below, the CSI related to prediction accuracy may include a Prediction Accuracy Indicator (PAI). For example, the PAI may indicate the accuracy of predicted downlink reference signal(s) (e.g., predicted CRI(s) and / or predicted SSBRI(s)), and the PAI may be interpreted / replaced as a Reference Signal-Prediction Accuracy Indicator (RS-PAI). For example, the PAI may indicate the accuracy of predicted CSI (e.g., predicted PMI), and the PAI may be interpreted / replaced as a Channel State Information-Prediction Accuracy Indicator (CSI-PAI).

[0167] Figure 6 is a flowchart showing an example of a CSI-related procedure.

[0168] Referring to FIG. 6, to perform one of the uses of CSI-RS, a terminal (e.g., user equipment, UE) receives configuration information related to CSI from a base station (e.g., general Node B, gNB) via radio resource control (RRC) signaling (S610).

[0169] The configuration information related to the above CSI may include at least one of CSI-IM (interference management) resource information, CSI measurement configuration information, CSI resource configuration information (e.g., M≥1 CSI-ResourceConfig resource setting), CSI-RS resource information, or CSI report configuration information (e.g., N≥1 CSI-ReportConfig reporting setting). As an example, the configuration information may include at least one of one or more CSI resource settings and / or one or more CSI reporting settings.

[0170] For example, the configuration information may include a first CSI resource setting for measurement and a second CSI resource setting for prediction. As a specific example, the measurement related to the prediction of CSI described below may be performed based on the first CSI resource setting. The terminal may perform L1-RSRP measurements on CSI-RS resources or SS / PBCH block resources associated with the first CSI resource setting. As a specific example, the prediction of CSI described below may be performed based on the second CSI resource setting. Based on the L1-RSRP measurements, the terminal may perform predictions on CSI-RS resources or SS / PBCH block resources associated with the second CSI resource setting. In other words, using L1-RSRPs as measurement metrics, the best CRI / best SSBRI (e.g., P-CRI(s), P-SSBRI(s)) may be predicted.

[0171] For example, the above configuration information may include a first CSI reporting setting related to prediction and a second CSI reporting setting related to prediction accuracy.

[0172] Information related to CSI resource configuration can be expressed as CSI-ResourceConfig IE. Information related to CSI resource configuration defines a group including at least one of an NZP (non-zero power) CSI-RS resource set, a CSI-IM resource set, or a CSI-SSB resource set. That is, the information related to CSI resource configuration includes a CSI-RS resource set list, and the CSI-RS resource set list may include at least one of an NZP CSI-RS resource set list, a CSI-IM resource set list, or a CSI-SSB resource set list. A CSI-RS resource set is identified by a CSI-RS resource set ID, and one resource set includes at least one CSI-RS resource. Each CSI-RS resource is identified by a CSI-RS resource ID.

[0173] Information related to CSI report configuration (e.g., CSI-ReportConfig IE) includes a reportConfigType parameter representing time domain behavior and a reportQuantity parameter representing the CSI-related quantity to be reported. The time domain behavior may be periodic, aperiodic, or semi-persistent.

[0174] The above reportQuantity parameter includes the channel quality indicator (CQI), precoding matrix indicator (PMI), CSI-RS resource indicator (CRI), SS / PBCH block resource indicator (SSBRI), layer indicator (LI), rank indicator (RI), L1-RSRP (Layer 1-Reference Signal Received Power), L1-SINR (Layer 1-Signal-to-Interference-plus-Noise Ratio), predicted CQI (P-CQI), predicted PMI (P-PMI), predicted CRI (P-CRI), predicted SSBRI (P-SSBRI), predicted LI (P-LI), predicted RI (P-RI), predicted L1-RSRP (P-L1-RSRP), and / or predicted L1-SINR (predicted It can be set to a value representing at least one of L1-SINR, P-L1-SINR) and / or a Prediction Accuracy Indicator (PAI) (e.g., CSI-PAI).

[0175] For example, the reportQuantity parameter can be set to cri, ssb-Index, cri-RSRP, or ssb-Index-RSRP. cri represents the CSI-RS resource indicator (CRI). RSRP represents the L1-RSRP (Layer 1-Reference Signal Received Power). ssb-Index represents the SS / PBCH block resource indicator (SSBRI).

[0176] For example, the reportQuantity parameter can be set to p-cri, p-ssb-index, p-cri-RSRP, or p-ssb-index-RSRP. p-cri represents the predicted CRI (predicted CRI, P-CRI). p-ssb-index represents the predicted SSBRI (predicted SSBRI, P-SSBRI). p-cri-RSRP represents the predicted CRI (predicted CRI, P-CRI) and predicted L1-RSRP (predicted L1-RSRP, P-L1-RSRP). p-ssb-index-RSRP represents the predicted SSBRI (predicted SSBRI, P-SSBRI) and predicted L1-RSRP (predicted L1-RSRP, P-L1-RSRP).

[0177] For example, the reportQuantity parameter can be set to pai (or rs-pai). pai (e.g., csi-pai) represents PAI (e.g., CSI-PAI).

[0178] The measurement resource may include settings for downlink signals and / or downlink resources for which the terminal will perform measurements to determine feedback information. The measurement resource may be set as a set of ZP and / or NZP CSI-RS resources associated with a CSI reporting setting. The NZP CSI-RS resource set may include a CSI-RS set or an SSB set. For example, L1-RSRP may be measured against a CSI-RS set or against an SSB set.

[0179] The terminal measures the CSI based on configuration information related to the above CSI (S620). The CSI measurement may include (1) a process of receiving the terminal's CSI-RS (S621) and (2) a process of computing the CSI through the received CSI-RS (S622). The terminal reports the CSI to the base station (S630).

[0180] resource setting

[0181] Each CSI resource setting 'CSI-ResourceConfig' contains a configuration for S≥1 CSI resource sets (given by the higher layer parameter csi-RS-ResourceSetList). The CSI resource setting corresponds to the CSI-RS-resourcesetlist, where S represents the number of configured CSI-RS resource sets. Here, the list of S≥1 CSI resource sets includes either or both of the NZP CSI-RS resource set(s) and the SS / PBCH block (SSB) set(s) used for L1-RSRP computation, or includes CSI-IM resource set(s).

[0182] Next, one or more CSI resource settings for channel measurement (CM) and interference measurement (IM) are established through higher layer signaling.

[0183] - CSI-IM resource for interference measurement.

[0184] - NZP CSI-RS resources for interference measurement.

[0185] - NZP CSI-RS resources for channel measurement.

[0186] That is, the CMR (channel measurement resource) may be an NZP CSI-RS for CSI acquisition, and the IMR (Interference measurement resource) may be an NZP CSI-RS for CSI-IM and IM.

[0187] Here, CSI-IM (or ZP CSI-RS for IM) is primarily used for inter-cell interference measurements.

[0188] Also, the NZP CSI-RS for IM is mainly used for intra-cell interference measurement from multi-users.

[0189] A UE can assume that the CSI-RS resource(s) for channel measurement set for one CSI reporting and the CSI-IM / NZP CSI-RS resource(s) for interference measurement (when the NZP CSI-RS resource(s) are used for interference measurement) have a QCL relationship with respect to 'QCL-TypeD' on a resource-by-resource basis.

[0190] As examined, resource setting can refer to a resource set list.

[0191] For aperiodic CSI, each trigger state set using the higher layer parameter CSI-AperiodicTriggerState is associated with one or more CSI-ReportConfigs, and each CSI-ReportConfig is linked to a periodic, semi-persistent, or aperiodic resource setting.

[0192] One reporting setting (e.g., CSI-ReportConfig) can be associated with up to three resource settings (e.g., CSI-ResourceConfig). For example, one CSI reporting setting may include the ID (e.g., CSI-ResourceConfigId) of at least one CSI resource setting. The at least one CSI resource setting may include a CSI resource setting associated with a measurement.

[0193] Beam Management (BM)

[0194] BM procedures are L1 (layer 1) / L2 (layer 2) procedures for acquiring and maintaining a set of base station (e.g., gNB, TRP, etc.) and / or terminal (e.g., UE) beams that can be used for downlink (DL) and uplink (UL) transmission / reception, and may include the following procedures and terms.

[0195] - Beam measurement: An operation in which a base station or UE measures the characteristics of a received beamforming signal.

[0196] - Beam determination: The operation in which a base station or UE selects its transmit beam (Tx beam) / receive beam (Rx beam).

[0197] - Beam sweeping: An operation that covers a spatial area using transmitting and / or receiving beams for a set time interval in a predetermined manner.

[0198] - Beam report: An operation in which the UE reports information about the beam-formed signal based on beam measurements.

[0199] The BM procedure can be divided into (1) a DL BM procedure using an SS (synchronization signal) / PBCH (physical broadcast channel) Block or CSI-RS, and (2) a UL BM procedure using an SRS (sounding reference signal).

[0200] In addition, each BM procedure may include Tx beam sweeping to determine the Tx beam and Rx beam sweeping to determine the Rx beam.

[0201] DL BM

[0202] The DL BM procedure may include (1) transmission to beamformed DL RS (reference signals) of the base station (e.g., CSI-RS or SS Block (SSB)) and (2) beam reporting of the terminal.

[0203] Here, beam reporting may include preferred DL RS ID(identifier)(s) and the corresponding L1-RSRP(Reference Signal Received Power).

[0204] The above DL RS ID may be SSBRI (SSB Resource Indicator) or CRI (CSI-RS Resource Indicator).

[0205] An example of beam forming using SSB and CSI-RS will be examined in detail below.

[0206] SSB beams and CSI-RS beams can be used for beam measurement. The measurement metric is L1-RSRP per resource / block. SSB is used for coarse beam measurement, while CSI-RS can be used for fine beam measurement. SSB can be used for both Tx beam sweeping and Rx beam sweeping.

[0207] Rx beam sweeping using SSBs can be performed as the UE changes the Rx beam across multiple SSB bursts for the same SSBRI. Here, one SS burst includes one or more SSBs, and one set of SS bursts includes one or more SSB bursts.

[0208] The DL BM procedure is examined below.

[0209] Configuration for beam reporting using SSB is performed during CSI / beam configuration in the RRC connected state (or RRC connected mode).

[0210] - The terminal receives configuration information from the base station. As a specific example, the terminal receives from the base station a CSI-ResourceConfig IE containing a CSI-SSB-ResourceSetList containing SSB resources used for BM.

[0211] Table 1 shows an example of CSI-ResourceConfig IE. As shown in Table 1, BM configuration using SSB is not defined separately, and SSB is configured like a CSI-RS resource.

[0212]

[0213] In Table 1, the csi-SSB-ResourceSetList parameter represents a list of SSB resources used for beam management and reporting in a single CSI-RS resource set. Here, the SSB resource set can be set to {SSBx1, SSBx2, SSBx3, SSBx4, …}. For example, the SSB index can be defined from 0 to 63.

[0214] - The terminal receives a DownLink Reference Signal (DL RS) from the base station. As a specific example, the terminal receives an SSB from the base station based on the CSI-SSB-ResourceSetList.

[0215] - The terminal transmits a beam report to the base station. As a specific example, if a CSI-ReportConfig related to reporting on SSBRI (SSB Resource Indicator) and L1-RSRP is configured, the terminal reports the best SSBRI and the corresponding L1-RSRP to the base station.

[0216] That is, if the reportQuantity of the above CSI-ReportConfig IE is set to 'ssb-Index-RSRP', the terminal reports the best SSBRI and the corresponding L1-RSRP to the base station.

[0217] And, if the terminal has a CSI-RS resource configured in the same OFDM symbol(s) as the SSB (SS / PBCH Block) and 'QCL-TypeD' is applicable, the terminal can assume that the CSI-RS and SSB are quasi-co-located in terms of 'QCL-TypeD'.

[0218] Here, the above QCL Type D may mean that the antenna ports are QCL-connected in terms of spatial Rx parameters. When a terminal receives multiple DL antenna ports that are in a QCL Type D relationship, it is acceptable to apply the same receiving beam. Additionally, the terminal does not expect CSI-RS to be established in an RE that overlaps with the RE of the SSB.

[0219] The configuration for beam reporting using CSI-RS is performed in the same manner as the configuration for beam reporting using SSB described above, so a redundant explanation is omitted. The operation of the beam reporting procedure using CSI is described below.

[0220] - The terminal receives configuration information from the base station. As a specific example, the terminal receives from the base station a CSI-ResourceConfig IE containing a CSI-SSB-ResourceSetList containing CSI resources used for BM (e.g., NZP CSI-RS resource set IE).

[0221] - The terminal receives CSI-RS resources within the NZP CSI-RS resource set through different Tx beams (DL spatial domain transmission filters) of the base station.

[0222] - The terminal selects (or determines) the best beam.

[0223] - The terminal reports the ID and associated quality information (e.g., L1-RSRP) for the selected beam to the base station. In this case, the reportQuantity of the CSI report config can be set to 'cri-RSRP'.

[0224] Explanation regarding Rel-17 / 18 beam management >

[0225] In Rel-17, DL DCI (e.g., DCI format 1-1 or 1-2) can indicate both the DL TCI state and the UL TCI state, or it can indicate only the UL TCI state without specifying the DL TCI state. Consequently, the methods used in the existing R15 / R16 for configuring UL beam and power control (PC) are replaced in Rel-17 by the aforementioned method of indicating the UL TCI state. More specifically, in R17, a single UL TCI state can be indicated through the TCI field of the DL DCI; this UL TCI state is applied to all PUSCHs and all PUCCHs after a certain period known as the beam application time, and can be applied to some or all of the indicated SRS resource sets. Additionally, the base station can utilize DCI and / or MAC-CE to perform a terminal common beam update, which performs indication / updates for multiple specific DL / UL channel / RS combinations using a single beam (utilizing joint or separate TCI states). For the target channel / RS of the common beam update, UE-dedicated CORESET and UE-dedicated reception on PDSCH are available for DL, and DG / CG-PUSCH and all or subset of dedicated PUCCH are available for UL, and additionally, AP CSI-RS for tracking / BM and SRS can be set as target channel / RS.In Rel-18, considering the M-TRP environment, the method of indicating multiple UL TCI states (and / or DL ​​TCI states) through the TCI field of DL DCI was standardized, and depending on the S-DCI based M-TRP environment and the M-DCI based M-TRP environment, uplink and downlink resources to which each indicated TCI is applied can be defined / configured.

[0226] In this specification, 'beam' may refer to a source RS for a 'spatial filter' or 'spatial relation', and may be interpreted as a QCL (type-D) RS, a TCI state, or (in the case of an uplink) a spatial relation RS.

[0227] For example, in this specification, 'beam' may refer to a spatial filter determined based on the reference RS or the source RS. The spatial filter may include a spatial domain filter, a spatial domain transmission filter, and a spatial domain receive filter. For example, in this specification, 'beam' may be interpreted or substituted with a reference signal index (RS index), a reference signal resource index (RS resource index), and / or a resource indicator (e.g., RS index, SSB index, CSI-RS resource index, SRS resource index, SSB Resource Indicator (SSBRI), CSI-RS Resource Indicator (CRI), etc.).

[0228] For example, a beam associated with a UL may be referred to as i) a spatial filter (for uplink transmission or uplink reception), ii) a spatial domain filter (for uplink transmission or uplink reception), iii) an uplink spatial domain transmission filter, iv) an uplink spatial domain receive filter, v) an uplink transmission spatial filter (UL Tx spatial filter) or vi) an uplink receive spatial filter (UL Rx spatial filter).

[0229] For example, a beam associated with DL may be referred to as i) a spatial filter (for downlink transmission or downlink reception), ii) a spatial domain filter (for downlink transmission or downlink reception), iii) a downlink spatial domain transmission filter, iv) a downlink spatial domain receive filter, v) a downlink transmission spatial filter (DL Tx spatial filter), or vi) a downlink receive spatial filter (DL Rx spatial filter).

[0230] In this specification, ' / ' means 'and', 'or', or 'and / or' depending on the context.

[0231] This specification considers AI / ML-based CSI reporting. This will be explained below with reference to FIG. 7.

[0232] Figure 7 illustrates an AI / ML-based CSI reporting operation.

[0233] Referring to FIG. 7, the terminal is equipped with an AI encoder (e.g., a CSI encoder), and the base station is equipped with an AI decoder (e.g., a CSI decoder). Each of the terminal and the base station performs AI / ML model inference. A model in which AI / ML model inference is performed at each of the two nodes (terminal and base station) in this manner is called a two-sided model. FIG. 7 illustrates CSI compression aimed at overhead reduction.

[0234] In other words, two-sided AI / ML means that an AI / ML model is deployed / configured at both the terminal and the base station (or network), and each performs inference. For example, an AI / ML model in the form of an auto-encoder may be configured. On the terminal side (CSI encoder side), the AI / ML model inference output is calculated using channel information (e.g., channel matrix / channel covariance matrix / channel eigenvector) or information that has undergone pre-processing of the corresponding channel information as input. The terminal feeds back information regarding the output (e.g., CSI) to the base station. For example, the information regarding the output may be generated based on specific post-processing. For example, the information regarding the output may be generated without post-processing.

[0235] The base station (CSI decoder) uses the corresponding feedback information (e.g., CSI) as input to an AI / ML model. For example, the feedback information to which pre-processing has been applied may be used as the input. For example, the feedback information may be used as the input without pre-processing. The base station calculates the inference output and decodes the final CSI with or without post-processing.

[0236] As another use case, the Rel-18 AI / ML study investigated CSI prediction based on a UE (User Equipment)-sided model. The CSI prediction use case is explained with reference to Fig. 8.

[0237] Figure 8 illustrates AI / ML-based CSI prediction.

[0238] Specifically, FIG. 8 illustrates an inference operation based on a UE-sided model. An AI / ML (Artificial Intelligence / Machine Learning) model (8B) is provided only on the terminal side, and model inference is performed by said model (8B). Historical measurements are applied as the AI / ML input (8A) of said model (8B). For example, said historical measurements may be performed based on a CSI resource configuration (e.g., CSI-ResourceConfig) connected to a CSI report configuration related to prediction (e.g., CSI-ReportConfig). -x , t -x+1,.. t0 represents historical time instances associated with multiple measurements. The AI / ML model output (8C) is a singular or multiple CSIs (e.g., one or more RSRPs and / or one or more beam indices). t N , t N+1 ,.. t N+K represents future time instances associated with the above CSI. The CSI type of the above input may be a raw channel matrix or a precoder type (e.g., eigenvector).

[0239] The following examines the operation of non-AI-based CSI reporting. For example, in Rel-15 Type I / II CSI reporting, the following operations may be performed for the purpose of inter-cell interference management. The base station may transmit settings / instructions to the terminal that restrict the PMIs to be used for CSI calculation based on RRC signaling. The terminal excludes the corresponding PMI(s) from the CSI calculation, calculates the preferred CSI (e.g., CQI / RI / PMI), and reports it to the base station.

[0240] In AI / ML-based CSI reporting, operations such as codebook subset restriction (CBSR) can be performed for the purpose of controlling inter-cell interference. Furthermore, AI / ML can be utilized to predict rank restrictions, CBSRs, and codebook types. This allows for the effective reduction of feedback overhead. This specification presents effective methods for settings and operations related to restrictions in AI / ML-based CSI reporting operations.

[0241] First, let's look at the legacy CBSR.

[0242] 1) Type 1 CSI

[0243] A. PMI restriction - bitmap based ( ) indication

[0244] i. In the above bitmap , In this case, it represents the number of antennas in the 1st domain and 2nd domain of the base station antenna port (x-pol antenna port). * α becomes the length of the DFT-vector (1D / 2D) corresponding to the ports corresponding to one slant of the x-pol antenna. The length of the final DL precoder is The final DL precoder consists of two * The DFT vector has a form in which it is connected to the cophase. The Rank 1 codebook is It has the form of is length * It is a DFT vector, and has a co-phase value of {1,j,-1,-j}.

[0245] ii. O1 and O2 are oversampling factors applied to the 1st domain and 2nd domain, respectively.

[0246] iii. represents the total number of DFT vectors, and each specific bit corresponds to a specific DFT vector.

[0247] B. RI restriction - An 8-bit bitmap is used.

[0248] i. The size of the RI restriction can be determined based on the max rank value set by the base station. For example, if the max rank set by the base station is 8, the rank value that the terminal should not use is indicated based on 8 bits. As a specific example, even if the max rank is 8, if "11110000" is indicated as the RI restriction, the terminal can only calculate and report CSIs corresponding to ranks 1, 2, 3, and 4.

[0249] 2) Type 2 CSI

[0250] A. PMI restriction

[0251] i. The entire system is not restricted. Specifically, a specific beam group is selected, and restrictions are applied to the DFT vectors within that beam group. For the beams in question, the amplitude available for Type 2 CSI configuration is also limited to 2 bits based on Table 2 below, depending on the beams being combined.

[0252] 1. The above The beams Partition into groups. Each group -by- It includes adjacent beams.

[0253] 2. P beam groups are selected based on the indicator.

[0254] 3. , here, is length Bitmap(length- bitmap) and, -by- The beam Limited by the beam ( -by- beam is restricted by beam), each beam is 2-bit soft power restricted.

[0255]

[0256] B. RI restriction - A 4-bit bitmap is used.

[0257] The AI / ML model-based signaling operation will be explained in detail below with reference to Fig. 9.

[0258] Figure 9 illustrates a signaling procedure based on an AI / ML model.

[0259] Step 1: In the description of this specification, signaling (e.g., information / data / channel / signal, etc.) or a set of signaling between a specific node (e.g., terminal, network, etc.) and another node may be interpreted as the signaling or a set of signaling of Step 1 used to perform an operation based on an AI / ML model, unless otherwise noted. For example, the signaling may correspond to training data for training (i.e., generation and / or reconstruction) the AI / ML model of FIG. 1, or to inference data used for inference of the AI / ML model, or to feedback to the AI / ML model, etc. If, in this specification, signaling between nodes is not required prior to an operation based on an AI / ML model, Step 1 may be omitted.

[0260] In cases where a one-side model is used in this specification, unidirectional / bidirectional signaling (set) in this specification may correspond to a one-stage signaling. Additionally, in cases where a two-side model is used in this specification, unidirectional / bidirectional signaling in this specification may correspond to a one-stage signaling, and repetitive signaling operations may also correspond to a one-stage signaling.

[0261] For example, in AI / ML model-based beam management, when a base station predicts (i.e., infers) high-quality beam(s) based on an AI / ML model, the base station can receive quality / intensity information for multiple beams from the terminal. Additionally, when a terminal predicts (i.e., infers) high-quality beam(s) based on an AI / ML model, the terminal can receive multiple beams from the base station.

[0262] Step 2: In the description of this specification, operations (e.g., computation, selection, prediction, etc.) at a specific node (e.g., terminal, network, etc.) or joint operations (e.g., computation, selection, prediction, etc.) at multiple nodes (e.g., terminal, network, etc.) may correspond to Step 2 operations based on one or more functions in the functional framework of an AI / ML model, unless otherwise noted. For example, said operations may correspond to the training (i.e., creation and / or reconstruction) of the AI / ML model of FIG. 2 or to inference of the AI / ML model. When a one-side model is used, operations performed by a single node in this specification may correspond to Step 2 operations. Additionally, when a two-side model is used, joint operations performed by multiple nodes in this specification may correspond to Step 2 operations.

[0263] For example, in an AI / ML model-based BM, a base station can predict (i.e., infer) high-quality beam(s) based on an AI / ML model by using quality / intensity information for multiple beams received from a terminal as inference data. Additionally, a terminal can measure multiple beams received from a base station and predict (i.e., infer) high-quality beam(s) based on an AI / ML model by using the measurement results as inference data.

[0264] Step 3: In the description of this specification, signaling (e.g., information / data / channel / signal, etc.) or a set of signaling between a specific node (e.g., terminal, network, etc.) and another node may be interpreted as the signaling or set of signaling of Step 3 generated as a result of an operation based on an AI / ML model, unless otherwise noted. For example, the signaling may correspond to the output resulting from the inference of the AI / ML model of FIG. 2. If signaling between nodes is not required as a result of an operation based on an AI / ML model in this specification, Step 3 may be omitted. If a one-side model is used in this specification, unidirectional / bidirectional signaling (set) in this specification may correspond to the signaling of Step 3. Additionally, if a two-side model is used in this specification, unidirectional / bidirectional signaling in this specification may correspond to the signaling of Step 3, and repetitive signaling operations may also correspond to the signaling of Step 3.

[0265] For example, in an AI / ML model-based BM, the base station may transmit beam(s) predicted based on the AI / ML model as candidates to the terminal so that the terminal can determine the optimal beam. Additionally, the terminal may report the beam(s) predicted based on the AI / ML model to the base station to request the base station to transmit candidate beams as candidates for determining the optimal beam.

[0266] The information included within the CSI related to prediction and the information included within the CSI related to monitoring (prediction accuracy), as described below, are explained as follows based on the background of the aforementioned CSI-related operations.

[0267] For example, CSIs reported based on existing reporting settings may include a Precoding Matrix Indicator (PMI). In other words, the reportQuantity parameter within CSI-Reportconfig can be set to a value representing the PMI.

[0268] For example, a CSI reported based on a forecast-related reporting setting may include forecasted information / forecasted parameter(s) / indicator(s). Specifically, the forecast-related CSI may include a predicted PMI. In other words, the reportQuantity parameter within CSI-Reportconfig may be set to a value representing the predicted PMI.

[0269] For example, a CSI reported based on a monitoring-related reporting setting may include parameter(s) or indicator(s) related to monitoring / prediction accuracy. As a specific example, the CSI related to monitoring / prediction accuracy may include a prediction accuracy indicator (e.g., a performance metric based on an example described below). In other words, the reportQuantity parameter in CSI-Reportconfig may be set to a value representing the prediction accuracy indicator. For convenience of explanation, the Prediction Accuracy Indicator may be expressed / referred to as PAI in this specification. The Prediction Accuracy Indicator (PAI) may be replaced with the CSI Prediction Accuracy Indicator (CSI-PAI).

[0270] Below, we will specifically examine how to calculate and report performance metrics to effectively monitor performance in AI / ML-based CSI prediction.

[0271] Proposal 1

[0272] We examine how to handle cases where the ground-truth CSI calculation time differs from the predicted CSI calculation time in the report.

[0273] Ground-truth CSI is required to calculate the metric / output for performance monitoring. To this end, the base station can set a new monitoring RS to the terminal (e.g., periodic, semi-persistent, aperiodic CSI-RS) and transmit it.

[0274] In this specification, ground-truth CSI refers to a CSI generated or calculated in the same manner as conventional methods. In other words, since ground-truth CSI is a CSI not based on prediction, ground-truth CSI may be interpreted or replaced with non-predicted CSI in this specification to distinguish it from predicted CSI. Additionally, in this specification, predicted CSI may be interpreted or replaced with predicted CSI parameters (e.g., PMI). For example, non-predicted CSI may be interpreted or replaced with non-predicted PMI, and predicted CSI may be interpreted or replaced with predicted PMI.

[0275] Figure 10 illustrates a mismatch between the actual ground-truth CSI and a prediction instance. Referring to Figure 10, the prediction instance within the prediction window and the actual ground-truth CSI may not be at the same point in time. In other words, the prediction instance and the ground-truth CSI may be based on different points in time. When a mismatch occurs between the ground-truth CSI and the prediction instance as described above, the metric can be calculated and reported as follows.

[0276] Proposal 1-1

[0277] The terminal can calculate and report a metric / output based on the time of receiving the RS to measure the Ground-truth CSI or the predicted CSI closest to the time of calculating the Ground-truth CSI from the RS.

[0278] Proposal 1-2

[0279] The terminal can calculate and report a metric / output based on the closest predicted CSI before / after the time of receiving the RS to measure the Ground-truth CSI or the time of calculating the ground truth CSI from the RS.

[0280] The above proposal 1-2 can be applied in different ways depending on the timing of the ground-truth CSI calculation. As shown in FIG. 10, if the first ground-truth CSI is later than the first prediction instance, the terminal can calculate and report a metric based on the nearest predicted CSI prior to the ground-truth CSI calculation time. Conversely, if the first ground-truth CSI is earlier than the first prediction instance, the terminal can calculate and report a metric based on the nearest predicted CSI after the ground-truth CSI calculation time.

[0281] Proposal 1-3

[0282] The terminal can calculate and report the filtered value (e.g., average / weighted average) of the Predicted CSI as a monitoring metric.

[0283] Consider reporting the filtered value as a metric when a mismatch occurs between the ground-truth CSI and the prediction instance in even a single instance. For example, after calculating the metric in the same manner as in Proposal 1-1 / 2 above, a weighted average value can be calculated by assigning weights to each metric and reported. For example, metric value for the nth instance, If we say it is the i-th weight value, then the weighted average is It can be calculated as follows. In this case, the sum of the weights becomes 1 (e.g., ), weights can be set. Each weight value can be determined and reported by the terminal or set / instructed by the base station.

[0284] Proposal 2

[0285] When a mismatch occurs between the ground-truth CSI and an instance within the prediction window, the terminal can request to accurately receive the monitoring RS through separate signaling.

[0286] Calculating the metric in the manner of Proposal 1 above does not result in the calculation of an accurate metric because the mismatch between the ground-truth CSI and the prediction instance is not resolved. Therefore, a method to solve this problem through separate signaling may be considered.

[0287] For example, the terminal may transmit a 1-bit indicator to the base station indicating that a mismatch has occurred between the Ground-truth CSI and an instance within the prediction window. Based on the 1-bit indicator, a request for timing alignment between the Ground-truth CSI and the prediction instance may be directed. Additionally, the terminal may transmit information such as timing advance / delaying and / or the number of monitoring occasions to the base station via a separate signaling. The base station may interpret such terminal actions as a request for a change in the timing of RS transmission related to the ground-truth CSI and perform actions accordingly. Through the aforementioned actions, the terminal can receive the RS related to the ground-truth CSI at the correct time and calculate an accurate metric.

[0288] More specifically, the terminal may request changes to information regarding the following items (Resource Setting configuration). For example, if it is a periodic / semi-persistet CSI, the terminal may request changes to information regarding periodicityAndOffset. For example, if it is an aperiodic CSI, the terminal may request changes to the number of AP CSI-RS resources K and separation information m between two consecutive CSI-RS resources.

[0289] Below, we examine the existing behavior related to the settings of periodicityAndOffset, K, and m described above.

[0290] 5.2.1.4.1 Resource Setting Configuration

[0291] For aperiodic CSI, each trigger state configured using the higher layer parameter CSI-AperiodicTriggerState is associated with a CSI-ReportConfig that is not configured with groupBasedBeamReporting-r17 or groupBasedBeamReporting-v18, and said CSI-ReportConfig is linked to a periodic, semi-persistent, or aperiodic resource setting.

[0292] When a single resource setting is configured, that resource setting (given by the upper-level parameter resourcesForChannelMeasurement) is for channel measurement for L1-RSRP, or channel and interference measurement for L1-SINR computation.

[0293] When two resource settings are configured, the first resource setting (given by the upper layer parameter resourcesForChannelMeasurement) is for channel measurement, and the second resource setting (given by either the upper layer parameter csi-IM-ResourcesForInterference or the upper layer parameter nzp-CSI-RS-ResourcesForInterference) is for interference measurement performed on CSI-IM or NZP CSI-RS.

[0294] When three resource settings are configured, the first resource setting (upper layer parameter resourcesForChannelMeasurement) is for channel measurement, the second resource setting (upper layer parameter csi-IM-ResourcesForInterference) is for CSI-IM based interference measurement, and the third resource setting (upper layer parameter nzp-CSI-RS-ResourcesForInterference) is for NZP CSI-RS based interference measurement.

[0295] For aperiodic CSI and periodic and semi-persistent CSI resource settings, each trigger state configured using the upper-level parameter CSI-AperiodicTriggerState is associated with a CSI-ReportConfig configured as groupBasedBeamReporting-r17 or groupBasedBeamReporting-v18, said CSI-ReportConfig is linked to the periodic or semi-persistent setting.

[0296] When a single resource setting is configured, that resource setting is given by resourcesForChannelMeasurement for L1-RSRP measurement. In this case, the number of CSI resource sets configured within the resource setting is S=2.

[0297] For aperioditic CSI and aperioditic CSI resource settings, each trigger state configured using the upper-level parameter CSI-AperiodicTriggerState is associated with a CSI-ReportConfig configured with groupBasedBeamReporting-r17 or groupBasedBeamReporting-v18, said CSI-ReportConfig being associated with resourcesForChannel and resourcesForChannel2, which correspond to the first resource set and second resource set, respectively, for L1-RSRP measurement. For periodic or semi-persistent CSI, each CSI-ReportConfig is linked to a periodic or semi-persistent resource setting.

[0298] When a single resource setting (given by the upper layer parameter resourcesForChannelMeasurement) is configured, said resource setting is for channel measurement for L1-RSRP, or channel and interference measurement for L1-SINR computation.

[0299] When two resource settings are configured, the first resource setting (given by the upper layer parameter resourcesForChannelMeasurement) is for channel measurement, and the second resource setting (given by the upper layer parameter csi-IM-ResourcesForInterference) is used for interference measurement performed on CSI-IM. For L1-SINR computation, the second resource setting (given by the upper layer parameter csi-IMResourcesForInterference or the upper layer parameter nzp-CSI-RS-ResourceForInterference) is used for interference measurement performed on CSI-IM or NZP CSI-RS.

[0300] For aperiodic CSI, a user equipment (UE) configured with a CSI-ReportConfig in which the upper-level parameter reportQuantity is set to 'tdcp' is expected to be configured with a single CSI Resource Setting (given by the upper-level parameter resourcesForChannelMeasurement). The said CSI Resource Setting may be periodic, having K_TRS∈{1,2,3} CSI-RS Resource Sets configured by the upper-level parameter trs-Info. Support for K_TRS=2 or 3 depends on the UE capability indication. For a periodic CSI-ResourceConfig, the UE may assume that all CSI-RS resources within K_TRS CSI-RS resource sets share the same QCL-Type A / C and, where applicable, Type D. The UE expects that all CSI-RS resources within the CSI-RS resource set(s) will be configured with the same bandwidth and subcarrier locations. A UE configured with a CSI-ReportConfig where the upper layer parameter reportQuantity is set to 'tdcp' is not expected to be configured with interference measurements on CSI-IM and / or NZP-CSI-RS.

[0301] For a UE configured with [LTM-CSI-ReportConfig], aperioditic, semi-persistent, or periodic CSI is associated with a single resource setting provided by [ltm-ResourcesForChannelMeasurement] for L1-RSRP measurement.

[0302] For a CSI-ReportConfig where the upper layer parameter codebookType is set to 'typeII', 'typeII-PortSelection', 'typeII-r16', 'typeII-PortSelection-r16', or 'typeII-PortSelection-r17', the UE is not expected to be composed of two or more CSI-RS resources within the resource set for channel measurement. For a CSI-ReportConfig where the upper-level parameter reportQuantity is set to 'none', 'cri-RI-CQI', 'cri-RSRP', 'ssb-Index-RSRP', 'cri-SINR', 'ssb-Index-SINR', 'cri-RSRP-Index', 'ssb-Index-RSRP-Index', 'cri-SINR-Index', or 'ssb-Index-SINR-Index', the UE is not expected to be configured with more than 64 NZP CSI-RS resources and / or SS / PBCH block resources within the resource setting for channel measurement. When interference measurement is performed on CSI-IM, each CSI-RS resource for channel measurement is associated with a CSI-IM resource in a resource-wise manner according to the ordering of CSI-RS resources and CSI-IM resources within the corresponding resource sets. The number of CSI-RS resources for channel measurement is equal to the number of CSI-IM resources.

[0303] A UE configured with a CSI-ReportConfig in which the upper-level parameter reportQuantity is set to 'cri-RI-PMI-CQI' and codebookType is set to 'typeII-CJT-r18' or 'typeII-CJT-PortSelection-r18' is expected to have 1 ≤ K ≤ 4 CSI-RS resources within the resource set for channel measurements. If interference measurements are performed on CSI-IM, only one resource is configured within the corresponding csi-IM-ResourceSet. If interference measurements are performed on NZP CSI-RS, only one resource is configured within the corresponding NZP-CSI-RS-ResourceSet for interference measurements.

[0304] A UE configured with a CSI-ReportConfig in which the upper-level parameters N4 and reportQuantity are set to 'cri-RI-PMI-CQI' is expected to consist of K∈{4,8,12} aperioditic CSI-RS resources or one periodic or semi-persistent CSI-RS resource within the resource set for channel measurement. For the aperioditic CSI-RS resource set for channel measurement, the K CSI-RS resources are triggered by the same triggering instance, and the separation between two consecutive CSI-RS resources is m∈{1,2} in slots, which is configured by the upper-level parameters in the NZP-CSI-RS-ResourceSet. The above K non-periodic CSI-RS resources are transmitted according to the order of the CSI-RS resource IDs configured within the CSI-RS resource set. The UE shall assume that the antenna ports having the same port index of the above K non-periodic CSI-RS resources are identical. If interference measurement is performed on CSI-IM, only one resource is configured within the corresponding csi-IM-ResourceSet. If interference measurement is performed on NZP CSI-RS, only one resource is configured within the corresponding NZP-CSI-RS-ResourceSet for interference measurement.

[0305] 5.2.2.3 Reference signal (CSI-RS)

[0306] 5.2.2.3.1 NZP CSI-RS

[0307] The UE may be configured with one or more NZP CSI-RS resource set configurations indicated by the higher layer parameters CSI-ResourceConfig and NZP-CSI-RS-ResourceSet. Each NZP CSI-RS resource set consists of K≥1 NZP CSI-RS resource(s). For each CSI-RS resource configuration, the following parameters, for which the UE must assume non-zero transmission power for the CSI-RS resource, are configured through the higher layer parameters NZP-CSI-RS-Resource, CSI-ResourceConfig, and NZP-CSI-RS-ResourceSet:

[0308] nzp-CSI-RS-ResourceId determines the CSI-RS resource configuration identity.

[0309] periodicityAndOffset defines the CSI-RS periodicity and slot offset for periodic / semi-persistent CSI-RS. All CSI-RS resources within a set are configured with the same periodicity, and the slot offset may be the same or different for different CSI-RS resources.

[0310] The above signaling may include information regarding a mismatch in the frequency domain as well as the timing. By transmitting offset information regarding the transmission frequency domain to the base station, the terminal can request that the frequency domains be closer or overlap.

[0311] Proposal 3

[0312] The following embodiments may be considered regarding information on rank / layer / instance / subband that the terminal assumes when calculating a performance metric (e.g., SGCS (Square Generalized Cosine Similarity)) when performing CSI prediction (for multiple time instances).

[0313] For example, information regarding the above rank / layer / instance / subband can be agreed upon / defined in advance.

[0314] For example, information regarding the rank / layer / instance / subband above can be set in the terminal by the base station.

[0315] For example, information regarding the rank / layer / instance / subband above is determined by the terminal, and the terminal can report the information to the base station.

[0316] A use case related to Proposal 3 is explained in detail with reference to Fig. 10. This use case applies CSI for K time instances within an observation window as AI / ML input and within a prediction window This is an AI / ML-based channel estimation use case that estimates future CSIs using CSIs for time instances as the AI / ML output. In this case, performance metrics (e.g., SGCS, NMSE) are calculated to monitor the performance of the model; however, since metrics can be calculated using multiple ranks, layers, instances, and subbands, a decision must be made on which one to use.

[0317] For example, a performance metric can be calculated as follows.

[0318] In the case of SGCS, instance , subband , layer SGCS can be calculated based on. As a specific example, the above SGCS is and It can be calculated based on. is the predicted CSI for the s-th SB and l-th layer (e.g., predicted PMI or the predicted precoder represented by the predicted PMI), and is the ground-truth CSI for the s-th SB and l-th layer (e.g., non-predicted PMI or the precoder represented by non-predicted PMI).

[0319] Here, is # of prediction instance (number of prediction instances) (e.g., N4 consecutive slot intervals), is # of SBs (number of subbands), and R is Rank information.

[0320] For example, the above , At least one of the and / or R information may be based on information set in the terminal by the base station to calculate the performance monitoring metric.

[0321] For example, the above , At least one of the and / or R information may be a value determined by a specific rule. If determined by a specific rule, the value ( , and / or R information) can be determined based on the value (e.g., RI) reported in the most recent triggered / reported inference report.

[0322] also, In this case, it is channel information in the form of a precoder determined by preprocessing, etc. from ground truth CSI measurement results. This channel information in the form of a precoder may refer to eigenvectors or singular vectors corresponding to the s-th SB and l-th layer.

[0323] From the above definition, the performance metric is defined as follows.

[0324] Performance metric

[0325]

[0326]

[0327]

[0328] Performance metrics are not limited to the aforementioned SGCS, NMSE, and Frobenius norm, and can be extended to others. The following methods for determining rank / layer / instance / subband can be considered.

[0329] As an additional method for calculating the above metric, the NW can also trigger the terminal with the calculated value of the monitoring metric / output for the reference scheme. Here, the reference scheme is a non-AI-based scheme. As the reference scheme, the sample and hold method (i.e., a method of applying CSI values ​​calculated based on the latest measurement to the prediction window (PW)) or CSI values ​​calculated based on Rel-18 CSI can be used. The base station can set which reference scheme to use for the terminal, or it can be agreed upon in advance.

[0330] In such cases, reference scheme-based metric calculation is It can be used by replacing it with the value calculated by the reference scheme. For notational convenience, the precoder corresponding to the s-th SB, l-th layer, and n-th prediction occasion calculated by the reference scheme It is denoted as . The above SGCS can be defined as follows.

[0331]

[0332] The above monitoring / output calculation method is a method of calculating the absolute value of the monitoring metric according to the precoder calculation method.

[0333] In another embodiment, the monitoring metric / output may be defined as performance relative to the reference scheme. For example, the monitoring metric / output may consist of the differential value of the metric or the gain value of the metric. In the case of the differential value, it may be defined as SGCS - SGCS(ref), and the gain value or It is a value that can be calculated in units such as dB. This method can be extended and applied to values ​​such as NMSE. If a negative value is calculated for the gain and / or differential, it means that the performance of AI / ML-based CSI prediction is very poor. Since reporting negative values ​​to the base station is not very helpful, the following actions can be considered. For example, the terminal can report 0 to the base station. Alternatively, instead of a specific value for the negative number, a specific code point called "negative gain" can be defined. The terminal can report this specific code point to the base station. This can reduce reporting overhead.

[0334] Proposal 3-1

[0335] When the terminal calculates the monitoring metric, a method of limiting it to a specific rank / layer (e.g., rank / layer 1) separately from / different from the reported rank value may be considered.

[0336] For example, a terminal can determine a specific rank / layer and calculate a performance metric without separate settings or instructions. For example, a base station can set or instruct the terminal to a value for the specific rank / layer. For example, if the terminal determines a specific rank / layer, calculates a metric, and reports it to the base station, the terminal can report information about the specific rank / layer to the base station along with a monitoring metric.

[0337] Proposal 3-2

[0338] For the number of layers corresponding to the reported Rank value, a method using the weighted average value of the metric calculated per layer may be considered.

[0339] A weighted average value can be used by multiplying the metric calculated for each layer by a weight. For example, second Let it be called the metric value for the i-th layer, and second If we say the i-th weight value, the weighted average is It can be represented as follows. In this case, the weight value It can be configured to satisfy [the condition]. Each weight value may be determined and reported by the terminal or set / instructed by the base station. Additionally, during the operation of Proposal 3-2 above, a method may be considered in which the base station does not set a separate weight or the terminal does not report a weight value, and the same weight is applied to all layers to calculate and report an average metric for all layers. By applying the same weight to all layers, there is no need for the base station to set or instruct different weights for each layer, and the terminal also does not need to report separate weight information, thus simplifying signaling at the level of RRC or MAC CE.

[0340] Proposal 3-3

[0341] For the number of instances corresponding to the reported predicted CSI, a method using the weighted average value of the metric calculated per instance may be considered.

[0342] A weighted average value can be used by multiplying the metric calculated for each subband by a weight. The weight value for the metric can be set in the same way as in Proposal 3-2 above. Each weight value can be determined and reported by the terminal or set / instructed by the base station. Additionally, during the operation of Proposal 3-3 above, a method can be considered in which the base station does not set a separate weight or the terminal does not report a weight value, and the same weight is applied to all instances to calculate and report the average metric for all instances.

[0343] Proposal 3-4

[0344] For the number of subbands corresponding to the reported frequency bands, a method using a weighted average value for the metric calculated for each subband may be considered.

[0345] A weighted average value can be used by multiplying the metric calculated for each subband by a weight. Weight values ​​for the metric can be set in the same manner as in Proposal 3-2 above, and each weight value can be determined and reported by the terminal or set / instructed by the base station. Furthermore, during the operation of Proposal 3-4 above, a method may be considered in which the base station does not set a separate weight or the terminal does not report a weight value, but instead applies the same weight to all subbands to calculate and report the average metric for all subbands. As a specific example, the terminal may report the arithmetic mean of the values ​​calculated for all subbands (e.g., similarities) as the performance metric. To explain in more detail, in the formula of the aforementioned SGCS, weight values ​​are not separately defined (e.g., =1, =1, =1), for defined instances, each layer based on the R value (yes: Assuming the case where metrics are calculated for each ), the metric for each layer (SGCS l )silver It can be expressed as follows.

[0346] The above monitoring metric can be reported based on UCI. In this case, both 1-part encoding and 2-part encoding can be considered.

[0347] In the case of 1-part encoding, the size of the UCI can be determined based on the (configured) maximum number of monitoring occasions and the number of monitoring metrics configured to be reported. If the number of times actually reported is less than the number of configured monitoring metrics, zero-padding can be applied to the remaining bits not used for reporting. This eliminates ambiguity in the payload.

[0348] In the case of 2-part encoding, the part 1 CSI (fixed payload) may include at least the number of reported metrics (# of metric). Additionally, the part 1 CSI may include the value of the first metric. The part 2 CSI (variable payload) may include the values ​​for the remaining metrics and report them (if the first metric is not reported in the part 1 CSI, all metrics may be reported in the part 2 CSI). If only a single metric is reported, the report in the part 2 CSI is omitted.

[0349] In addition, metrics may be reported together with inference (e.g., sending a CSI report containing an inference result (predicted CSI) and a monitoring result (performance metric)). An indicator indicating whether monitoring-related reporting is performed may be included in the part 1 CSI.

[0350] Proposal 4

[0351] The following embodiments may be considered regarding the quantity to be used by the terminal when calculating performance metrics (e.g., SGCS, NMSE) when performing CSI prediction (for multiple time instances). For example, the quantity associated with the performance metric may be set on the terminal by the base station. For example, the terminal may determine the quantity associated with the performance metric and report the quantity to the base station.

[0352] For example, in the above proposal 1 / 2 / 3, the performance metric can be calculated based on the raw channel of the predicted CSI and ground-truth CSI. For example, in the above proposal 1 / 2 / 3, the performance metric can be calculated based on the PMI (e.g., predicted PMI, non-predicted PMI).

[0353] Using a raw channel enables more accurate monitoring because there is no information loss or quantization error that can occur when calculating PMI. On the other hand, using PMI has the advantage of reducing overhead compared to a raw channel when reporting CSI. For example, the quantity can be set via a 1-bit indication from the base station. For example, the terminal can determine the quantity and report it through the reportQuantity of the predicted CSI.

[0354] According to one embodiment, 0 or 1 can be set / instructed as a parameter for the quantity used in metric calculation through RRC or MAC-CE. If 0 is set / instructed, the terminal can calculate a performance metric based on PMI. If 1 is set / instructed, the terminal can calculate a performance metric based on the raw channel.

[0355] According to one embodiment, when a terminal determines reportquantity, it can calculate a metric as follows. If the reportQuantity of the predicted CSI includes a raw channel, the terminal can calculate a metric based on the raw channel. If the reportQuantity of the predicted CSI does not include a raw channel, the terminal can calculate a performance metric based on PMI.

[0356] Proposal 4-1

[0357] A method of performing performance monitoring by reporting the difference between Raw Channel and PMI-based performance metrics may be considered.

[0358] Proposal 4-1 above concerns a method for additionally reporting the difference between performance metrics calculated using an unselected method (e.g., Raw channel-based or PMI-based) and the method determined by the terminal or configured by the base station when calculating metrics for performance monitoring. As previously explained, there are two methods for calculating performance metrics: one that calculates directly in the raw channel domain and another that calculates using channel information precoded based on PMI. However, depending on the actual operating environment or channel conditions of the terminal, there may be cases where a performance metric calculated using a method different from the configured method is superior (i.e., reflects the conditions more accurately or shows better correlation). For example, even if the base station instructed the terminal to calculate a PMI-based metric, a metric calculated based on the raw channel may be advantageous in actual performance evaluation, such as reflecting the channel conditions more accurately than a metric calculated based on PMI in a specific actual channel environment. Conversely, there may also be cases where the PMI-based performance is superior even though the base station instructed the calculation of a raw channel-based metric. In such a situation, it is highly likely that optimal performance evaluation cannot be performed using only the single metric set by the base station. To resolve this problem, the following operations may be performed.

[0359] The terminal calculates and reports a metric from a specific metric calculation quantity (Raw channel or PMI) instructed by the base station, and at the same time, may calculate the difference in performance metrics between the configured method and other unselected methods and report it additionally. In this case, the difference in performance metrics may be defined to be additionally reported to the base station only when the performance metric of the configured method is poor.

[0360] As a specific example of the above method, if a base station commands a terminal to perform a PMI-based metric calculation, the terminal first reports a performance metric (m_PMI) calculated based on PMI using the predicted CSI and ground-truth CSI. At this time, the terminal additionally calculates a performance metric (m_RAW) based on the raw channel as well, and if the m_RAW value is higher, calculates the difference between the two methods (e.g., △m = |m_RAW - m_PMI|) and additionally reports it. Compared to reporting both m_RAW and m_PMI, this method has the advantage of reducing reporting overhead because the process of comparing the two performance metric values ​​can be omitted and the value range of △m will be smaller than that of m_RAW (or m_PMI).

[0361] Proposal 5

[0362] Below, we examine examples related to quantization information (X,Y) when a terminal reports a Performance metric calculated by SGCS.

[0363] According to one embodiment, when a terminal reports SGCS, which is used as a performance indicator between the predicted CSI and the ground-truth CSI, to a base station, the terminal may report a quantized SGCS value based on a quantization lower limit value (X) and a quantization step size (Y). Generally, the SGCS value has a value between 0 and 1. The closer it is to 1, the higher the similarity between the predicted CSI and the ground-truth CSI and the better the performance. However, if the SGCS value is below a certain value (e.g., indicating very low performance), receiving an accurate value is not very meaningful from the base station's perspective and may instead only add unnecessary overhead to the reporting. To solve this problem, the quantization lower limit (X) and step size (Y) for SGCS reporting may be set / defined in advance. Based on this, the terminal may quantize and report the SGCS value. The degree of quantization varies depending on X and Y, and the payload may also vary accordingly. For example, if the lower limit X and step-size Y are set / defined to 0.6 and 0.1, respectively, the terminal can quantize the SGCS value and report it to the base station in the following way. When 00, When 01, When 10, When it is 11, it can be reported. In other words, the terminal can report a code point representing a range related to SGCS quantization (e.g., one of four code points (00, 01, 10, 11) for a 2-bit field, one of 16 code points for a 4-bit field).

[0364] As a specific embodiment of the above method, if the terminal calculates the SGCS of the predicted CSI and ground-truth CSI as 0.85 at time n, it quantizes and reports the SGCS as 10 according to the quantization lower limit X = 0.6 and step size Y = 0.1.

[0365] As the wireless channel environment changes dynamically over time, the range and magnitude of change of SGCS values ​​may vary. The above proposal provides the flexibility to actively adjust the SGCS quantization method according to the purpose of base station performance monitoring, thereby offering the advantage of efficiently reporting SGCS metrics in response to changes in the wireless environment without additional overhead. is a value derived from the predicted CSI, and the following various implementations may exist.

[0366] 1. may be an eigenvector corresponding to the l-th layer of the predicted CSI of the s-th SB and n-th instance calculated from the inference output before conversion to PMI.

[0367] 2. It may be a PMI corresponding to the s-th SB, n-th instance, and l-th layer converted into a PMI based on the inference output (using Rel-18 doppler CSI).

[0368] For example, which of the definitions 1 and 2 above determines the metric to be calculated can be determined based on instructions from the base station.

[0369] For example, according to which of the definitions 1 and 2 above the metric is to be calculated, it can be agreed / defined in advance.

[0370] For example, which of the definitions 1 and 2 above determines the metric to be calculated is determined based on the terminal's capability, and the base station can determine which definition the metric was calculated according to through the terminal's capability report.

[0371] Below, we examine matters related to the performance metric (SGCS) associated with Proposal 5.

[0372] During the discussions for Releases 18 and 19, NMSE and SGCS were reviewed as candidate metrics for monitoring the performance of AI / ML-based CSI prediction. SGCS is defined as the squared magnitude of the normalized inner product and directly indicates how accurately the predicted CSI vector matches the ground-truth CSI vector in terms of direction. In contrast, NMSE measures the overall magnitude-based deviation between two vector / channel matrices. Codebook selection and PMI reporting procedures evaluate how well the reported precoding vector represents the actual channel vector in terms of direction rather than absolute amplitude. From this perspective, the SGCS metric, which explicitly measures the directional correlation between two complex channel vectors, aligns directly with the NR CSI feedback principle and existing codebook structures. High SGCS values ​​clearly imply that the channel direction has been accurately predicted, which directly leads to improvements in PMI selection and precoding accuracy.

[0373] Furthermore, the two metrics can also be compared in terms of their value range, which is important as it directly affects the number of quantization bits required for reporting. Specifically, NMSE can theoretically range from values ​​close to zero (perfect prediction) to very large values, i.e., values ​​close to infinity; this phenomenon is particularly pronounced when prediction performance is low or received power is low. Such an unrestricted value range may require more quantization bits for reporting or cause significant quantization errors, thereby degrading reporting accuracy. On the other hand, SGCS is limited by definition to a fixed range between 0 and 1. This clearly restricted value range enables accurate and efficient representation with only a small number of quantization bits, significantly reducing quantization errors and reporting overhead. Therefore, SGCS is supported as an intermediate KPI for monitoring metrics when performance monitoring type 1 or 3 is supported.

[0374] proposal #:

[0375] If performance monitoring type 1 or 3 is supported, SGCS is supported as a monitoring metric rather than NMSE for CSI prediction monitoring.

[0376] The range of reported values ​​is from X to 1, and the step size is Y.

[0377] X and Y values ​​will be reviewed later (FFS).

[0378] Below, we examine the operation procedure of a base station and a terminal to which a method based on at least one embodiment of Proposal 1 to Proposal 5 described above can be applied.

[0379] UE-sided model

[0380] Terminal operation:

[0381] Step 1: A step of reporting capabilities including the maximum number of CSI-RS resources supported by the base station, the number of CSI-RS ports, the total number of CSI-RS ports that can be supported simultaneously, the number of Rx antenna groups, etc.

[0382] Step 2: Receiving CSI-RS transmission, CSI reports, and related configuration information from the base station

[0383] Step 3: A step of receiving CSI-RS from the base station and measuring / predicting / calculating CSI using an AI / ML model.

[0384] Step 3-1: A step of performing CSI omission based on configuration information and CSI priority from the base station.

[0385] Step 4: Reporting the above measured / predicted / calculated CSI to the base station

[0386] Step 5: The step of receiving scheduling for downlink channels (e.g., PDCCH, PDSCH) from the base station.

[0387] Step 6: Receiving the downlink channel / signal transmitted by the base station

[0388] Base station operation:

[0389] Step 1: A step of receiving a capability report from the terminal including the maximum number of supported CSI-RS resources, the number of CSI-RS ports, the total number of simultaneously supported CSI-RS ports, the number of Rx antenna groups, etc.

[0390] Step 2: Transmitting CSI-RS to the terminal and sending CSI reports and related configuration information.

[0391] Step 3: Step of transmitting CSI-RS to the terminal

[0392] Step 4: The step where the terminal receives the measured / predicted / calculated CSI from the terminal.

[0393] Step 5: A step of scheduling downlink channels (e.g., PDCCH, PDSCH) based on the CSI reported from the terminal and transmitting them to the terminal.

[0394] NW-sided model

[0395] Terminal operation:

[0396] Step 1: A step of reporting capabilities including the maximum number of CSI-RS resources supported by the base station, the number of CSI-RS ports, the total number of CSI-RS ports that can be supported simultaneously, the number of Rx antenna groups, etc.

[0397] Step 2: Receiving CSI-RS transmission, CSI reports, and related configuration information from the base station

[0398] Step 3: A step of receiving CSI-RS from the base station and measuring / predicting / calculating CSI based on it.

[0399] Step 3-1: A step of performing CSI omission based on configuration information and CSI priority from the base station.

[0400] Step 4: Reporting the above measured / predicted / calculated CSI to the base station

[0401] Step 5: The step of receiving scheduling for downlink channels (e.g., PDCCH, PDSCH) from the base station.

[0402] Step 6: Receiving the downlink channel / signal transmitted by the base station

[0403] Base station operation:

[0404] Step 1: A step of receiving a capability report from the terminal including the maximum number of supported CSI-RS resources, the number of CSI-RS ports, the total number of simultaneously supported CSI-RS ports, the number of Rx antenna groups, etc.

[0405] Step 2: Transmitting CSI-RS to the terminal and sending CSI reports and related configuration information.

[0406] Step 3: Step of transmitting CSI-RS to the terminal

[0407] Step 4: The step where the terminal receives the measured / predicted / calculated CSI from the terminal.

[0408] Step 5: A step of predicting CSI using an AI / ML model based on CSI reported from the terminal, scheduling downlink channels (e.g., PDCCH, PDSCH), and transmitting this to the terminal.

[0409] In the above terminal / base station operation, (some) specific steps may be omitted.

[0410] In terms of implementation, the operations of the base station / terminal according to the embodiments described above (e.g., operations based on at least one of proposals 1 to 5) can be processed by the device of FIG. 13 (e.g., the processor (110, 210) of FIG. 13).

[0411] In addition, the operations of the base station / terminal according to the above-described embodiment (e.g., operations based on at least one of proposals 1 to 5) may be stored in memory (e.g., 140, 240 of FIG. 13) in the form of instructions / programs (e.g., instruction, executable code) for driving at least one processor (e.g., 110, 210 of FIG. 13).

[0412] The embodiments described above will be explained in detail below with reference to FIGS. 11 and FIGS. 12 in terms of the operation of the terminal and base station. The methods described below are distinguished only for convenience of explanation, and it is obvious that a part of one method may be substituted with a part of another method or combined with one another and applied.

[0413] FIG. 11 is a flowchart illustrating a method according to one embodiment of the present specification.

[0414] Referring to FIG. 11, a method according to one embodiment of the present specification includes a setting information receiving step (S1110), a first CSI report transmission step (S1120), and a second CSI report transmission step (S1130).

[0415] In S1110, the terminal receives configuration information related to Channel State Information (CSI) from the base station.

[0416] For example, the above configuration information may include information based on at least one of the aforementioned CSI-related operations and proposals 1 through 5. As a specific example, the above configuration information may include at least one of one or more resource configurations (e.g., M≥1 CSI-ResourceConfig resource setting) and / or one or more reporting configurations (e.g., N≥1 CSI-ReportConfig reporting setting). Each of the one or more reporting configurations may be associated with up to three resource configurations. In other words, each reporting configuration may include the IDs (e.g., CSI-ResourceConfigId) of up to three resource configurations.

[0417] For example, the above one or more reporting settings may include i) a first reporting setting related to CSI prediction and ii) a second reporting setting related to CSI prediction accuracy. As a specific example, the setting information may include i) a first reporting setting related to CSI prediction and ii) a second reporting setting related to CSI prediction accuracy.

[0418] In S1120, the terminal transmits a first CSI report related to the first report setting to the base station.

[0419] In S1130, the terminal transmits a second CSI report related to the second report setting to the base station.

[0420] The above-mentioned second CSI report may include a metric based on at least one of the aforementioned proposals 1 to 5. This will be explained in detail below.

[0421] According to one embodiment, the second CSI report may include fields related to Square Generalized Cosine Similarity (SGCS) values. Each code point of the fields may represent a range related to SGCS quantization. The code point may be one of a plurality of code points. The plurality of code points may include code points mapped to a plurality of ranges. The plurality of ranges may include ranges defined by quantizing a value range of greater than or equal to 1 and less than or equal to a predefined value (e.g., X described above) based on a step-size (e.g., Y described above). This embodiment may be based on Proposal 5. For example, the predefined value (e.g., X described above) may be greater than 0.

[0422] The above SGCS values ​​can be calculated based on at least one of the embodiments of Proposal 1 to Proposal 5.

[0423] According to one embodiment, the SGCS values ​​may include the SGCS values ​​for each layer among the layers. The SGCS value for each layer may be the average of the values ​​calculated for the subbands. This embodiment may be based on Proposal 3-4. In this case, each of the values ​​may be the square of the magnitude of the normalized inner product based on the PMI.

[0424] For example, each of the above values ​​may be the squared magnitude of the normalized inner product calculated based on the predicted PMI and the non-predicted PMI. The present embodiment may be based on Proposal 4-1 and Proposal 5.

[0425] Specifically, each of the above values ​​can be calculated based on the following mathematical formula.

[0426] [Mathematical Formula]

[0427]

[0428] Here, is a vector based on the above-mentioned predicted PMI associated with each subband (e.g., the predicted precoder represented by the above-mentioned predicted PMI), and is a vector based on the above non-predicted PMI associated with each subband (e.g., the precoder represented by the above non-predicted PMI), and is the magnitude of the vector (L2-norm), and can be a Hermitian transpose operation.

[0429] For example, the above SGCS value can be calculated based on the following mathematical formula.

[0430] [Mathematical Formula]

[0431]

[0432] Here, s is a subband index, and S may be the number of the subbands.

[0433] According to one embodiment, the number of layers (e.g., R) may be determined based on a Rank Indicator (RI) reported based on the first CSI report. This embodiment may be based on Proposal 3.

[0434] According to one embodiment, the first CSI report may include a Precoding Matrix Indicator (PMI). The PMI may indicate predicted precoder matrices associated with one or more intervals (e.g., N4 slot intervals).

[0435] For example, the 1st / 2nd CSI report can be interpreted / substituted with the 1st / 2nd CSI.

[0436] For convenience of explanation, the embodiment of Proposal 5 was mentioned first, but the essential configuration of the method according to the embodiment of this specification is not intended to be limited to the configuration according to the embodiment of Proposal 5. In other words, the essential configuration of the method according to the embodiment of this specification is not defined by the order in which the embodiments are mentioned. The essential configuration of the method according to the embodiment of this specification may be based on at least one of the above-described embodiments / operations.

[0437] For example, the method according to the embodiment of the present specification may essentially include an operation based on the embodiment of Proposal 5 (SGCS quantization-based feature).

[0438] For example, the method according to the embodiment of the present specification may essentially include features based on the embodiment of Proposal 3-4 (where the SGCS value for each layer is the average of the values ​​calculated for the subbands). In this case, the method according to the embodiment of the present specification includes the steps of receiving configuration information related to channel state information (CSI), transmitting a first CSI report related to a first reporting setting, and transmitting a second CSI report related to a second reporting setting. The configuration information includes i) the first reporting setting related to CSI prediction and ii) the second reporting setting related to CSI prediction accuracy. The second CSI report includes fields related to Square Generalized Cosine Similarity (SGCS) values. The SGCS values ​​include the SGCS value for each layer among the layers. The SGCS value for each layer is characterized as being the average of the values ​​calculated for the subbands.

[0439] Operations based on the above-described S1110 to S1130 can be implemented by the device of FIG. 13. For example, referring to FIG. 13, the terminal (200) can control one or more transceivers (230) and / or one or more memories (240) to perform operations based on S1110 to S1130.

[0440] The embodiments described above will be explained in detail below in terms of base station operation.

[0441] S1210 to S1230 described below correspond to S1110 to S1130 described in FIG. 11. Considering the above correspondence, redundant descriptions are omitted. That is, the specific description of the base station operation described below can be replaced by the description / embodiment of FIG. 11 corresponding to the operation.

[0442] FIG. 12 is a flowchart illustrating a method according to another embodiment of the present specification.

[0443] Referring to FIG. 12, a method according to another embodiment of the present specification includes a setting information transmission step (S1210), a first CSI report reception step (S1220), and a second CSI report reception step (S1230).

[0444] In S1210, the base station transmits configuration information related to Channel State Information (CSI) to the terminal. For example, the configuration information may include i) a first reporting setting related to CSI prediction and ii) a second reporting setting related to CSI prediction accuracy.

[0445] In S1220, the base station receives a first CSI report related to the first report setting from the terminal.

[0446] In S1230, the base station receives a second CSI report related to the second report setting from the terminal.

[0447] The above-mentioned second CSI report may include a metric based on at least one of the aforementioned proposals 1 to 5.

[0448] According to one embodiment, the second CSI report may include fields related to Square Generalized Cosine Similarity (SGCS) values. Each code point of the fields may represent a range related to SGCS quantization. The code point may be one of a plurality of code points. The plurality of code points may include code points mapped to a plurality of ranges. The plurality of ranges may include ranges defined by quantizing a value range of 1 or more and 1 or less from a predefined value (e.g., X described above) based on a step-size (e.g., Y described above). This embodiment may be based on Proposal 5.

[0449] According to one embodiment, the SGCS values ​​may include SGCS values ​​for each layer among the layers. The SGCS value for each layer may be the average of the values ​​calculated for the subbands. This embodiment may be based on Proposal 3-4.

[0450] Operations based on the above-described S1210 to S1230 can be implemented by the device of FIG. 13. For example, referring to FIG. 13, a base station (100) can control one or more transceivers (130) and / or one or more memories (140) to perform operations based on S1210 to S1230.

[0451] The operations / terms based on the embodiments described above are described under the assumption of an existing system (e.g., a 5G system). However, this is for the convenience of explanation and is not intended to limit the scope of application of the technical problems and means for solving problems that are to be solved by this specification to a specific system. That is, the technical problems / technical issues / problems mentioned in this specification may exist in other systems (e.g., a 6G system). It is evident that the embodiments of this specification can be extended to solve problems that exist in other systems as well. Therefore, for the extended application of the embodiments of this specification to other systems, terms defined / described based on a 5G system may be replaced / changed with terms defined in said other systems (or generalized terms not specific to one system). For example, PRACH, PUSCH, PUCCH, or SRS may be replaced / changed to uplink signals (or uplink channels). For example, SSB, CSI-RS, PDSCH, and PDCCH may be replaced / changed to downlink signals (or downlink channels).

[0452] Hereinafter, an apparatus to which the embodiments of the present specification can be applied (an apparatus implementing the method / operation according to the embodiments of the present specification) will be described with reference to FIG. 13.

[0453] FIG. 13 is a drawing showing the configuration of a first device and a second device according to an embodiment of the present specification.

[0454] The first device (100) may include a processor (110), an antenna unit (120), a transceiver (130), and a memory (140).

[0455] The processor (110) performs baseband-related signal processing and may include an upper layer processing unit (111) and a physical layer processing unit (115). The upper layer processing unit (111) may process operations of the MAC layer, RRC layer, or higher upper layers. The physical layer processing unit (115) may process operations of the PHY layer. For example, if the first device (100) is a base station device in base station-terminal communication, the physical layer processing unit (115) may perform uplink reception signal processing, downlink transmission signal processing, etc. For example, if the first device (100) is a first terminal device in terminal-terminal communication, the physical layer processing unit (115) may perform downlink reception signal processing, uplink transmission signal processing, sidelink transmission signal processing, etc. In addition to performing baseband-related signal processing, the processor (110) may also control the overall operation of the first device (100).

[0456] The antenna section (120) may include one or more physical antennas, and if it includes multiple antennas, it may support MIMO transmission and reception. The transceiver (130) may include an RF (Radio Frequency) transmitter and an RF receiver. The memory (140) may store information processed by the processor (110) and software, operating systems, applications, etc. related to the operation of the first device (100), and may include components such as a buffer.

[0457] The processor (110) of the first device (100) may be configured to implement the operation of the base station in base station-terminal communication (or the operation of the first terminal device in terminal-terminal communication) in the embodiments described in this disclosure.

[0458] The second device (200) may include a processor (210), an antenna unit (220), a transceiver (230), and a memory (240).

[0459] The processor (210) performs baseband-related signal processing and may include an upper layer processing unit (211) and a physical layer processing unit (215). The upper layer processing unit (211) may process operations of the MAC layer, RRC layer, or higher upper layers. The physical layer processing unit (215) may process operations of the PHY layer. For example, if the second device (200) is a terminal device in base station-terminal communication, the physical layer processing unit (215) may perform downlink reception signal processing, uplink transmission signal processing, etc. For example, if the second device (200) is a second terminal device in terminal-terminal communication, the physical layer processing unit (215) may perform downlink reception signal processing, uplink transmission signal processing, sidelink reception signal processing, etc. In addition to performing baseband-related signal processing, the processor (210) may also control the overall operation of the second device (210).

[0460] The antenna section (220) may include one or more physical antennas, and may support MIMO transmission and reception if it includes multiple antennas. The transceiver (230) may include an RF transmitter and an RF receiver. The memory (240) may store information processed by the processor (210) and software, operating systems, applications, etc. related to the operation of the second device (200), and may include components such as a buffer.

[0461] The processor (210) of the second device (200) may be configured to implement the operation of the terminal in base station-terminal communication (or the operation of the second terminal device in terminal-terminal communication) in the embodiments described in this disclosure.

[0462] In the operation of the first device (100) and the second device (200), the details described in the examples of the present disclosure regarding the base station and terminal (or the first terminal and the second terminal in terminal-to-terminal communication) in base station-to-terminal communication may be applied in the same way, and redundant descriptions are omitted.

[0463] Here, the wireless communication technology implemented in the device of the present disclosure may include LTE, NR, and 6G, as well as Narrowband Internet of Things (NB-IoT) for low-power communication. For example, NB-IoT technology may be an example of Low Power Wide Area Network (LPWAN) technology and may be implemented according to standards such as LTE Cat NB1 and / or LTE Cat NB2, but is not limited to the names mentioned above.

[0464] Additionally or alternatively, the wireless communication technology implemented in the device of the present disclosure may perform communication based on LTE-M technology. For example, LTE-M technology may be an example of LPWAN technology and may be referred to by various names such as eMTC (enhanced Machine Type Communication). For example, LTE-M technology may be implemented in at least one of various standards such as 1) LTE CAT 0, 2) LTE Cat M1, 3) LTE Cat M2, 4) LTE non-BL (non-Bandwidth Limited), 5) LTE-MTC, 6) LTE Machine Type Communication, and / or 7) LTE M, and is not limited to the names mentioned above.

[0465] Additionally or generally, the wireless communication technology implemented in the device of the present disclosure may include at least one of ZigBee, Bluetooth, and a Low Power Wide Area Network (LPWAN) for low-power communication, but is not limited to the names mentioned above. For example, ZigBee technology can create personal area networks (PANs) related to small / low-power digital communication based on various standards such as IEEE 802.15.4 and may be referred to by various names.

Claims

1. Regarding the method, A step of receiving configuration information related to channel state information (CSI) from a base station by a terminal, The above setting information includes i) a first reporting setting related to CSI prediction and ii) a second reporting setting related to CSI prediction accuracy; A step of transmitting a first CSI report related to the first report setting to the base station by the terminal; and The method includes the step of transmitting a second CSI report related to the second report setting to the base station by the terminal; The above second CSI report includes fields related to SGCS (Square Generalized Cosine Similarity) values, and Each code point of the above fields represents a range related to SGCS quantization, and The above code point is one of a plurality of code points, and The above plurality of code points include code points mapped to a plurality of ranges, and A method characterized in that the above plurality of ranges include ranges defined by quantizing a value range of greater than or equal to 1 based on a step size.

2. In Paragraph 1, The above SGCS values ​​include the SGCS values ​​for each layer among the layers, and A method characterized in that the SGCS value for each of the above layers is the average of the values ​​calculated for the subbands.

3. In Paragraph 2, A method characterized in that each of the above values ​​is the squared magnitude of a normalized inner product calculated based on predicted PMI and non-predicted PMI.

4. In Paragraph 3, Each of the above values ​​is calculated based on the following mathematical formula, and [Mathematical Formula] Here, is a vector based on the above predicted PMI associated with each subband, and is a vector based on the above non-predicted PMI associated with each subband, and is the magnitude of the vector (L2-norm), and A method characterized by being a Hermitian transpose operation.

5. In Paragraph 4, The above SGCS value is calculated based on the following mathematical formula, and [Mathematical Formula] A method characterized in that, where s is a subband index and S is the number of said subbands.

6. In Paragraph 2, A method characterized in that the number of the above layers is determined based on a Rank Indicator (RI) reported based on the first CSI report.

7. In Paragraph 1, A method characterized by the above-mentioned predefined value being greater than 0.

8. In Paragraph 1, The above first CSI report includes a Precoding Matrix Indicator (PMI), and A method characterized in that the above PMI indicates predicted precoder matrices associated with one or more intervals.

9. Regarding the terminal, One or more transmitters / receivers; One or more processors; and It includes one or more memories connected to the above one or more processors and storing instructions, A terminal characterized by the above instructions, based on execution by the one or more processors, causing the terminal to perform all steps of the method according to any one of claims 1 to 8.

10. A device comprising one or more memories and one or more processors connected to the one or more memories, An apparatus characterized in that the above one or more memories store instructions that cause the apparatus to perform all steps of the method according to any one of claims 1 to 8, based on execution by the above one or more processors.

11. In a non-transitory computer-readable storage medium for storing instructions, A non-transitory computer-readable storage medium characterized by instructions executable by one or more processors such that the terminal performs all steps of the method according to any one of claims 1 through 8.

12. Regarding the method, A step of transmitting configuration information related to channel state information (CSI) to a terminal by a base station, The above setting information includes i) a first reporting setting related to CSI prediction and ii) a second reporting setting related to CSI prediction accuracy; A step of receiving a first CSI report related to the first reporting setting from the terminal by the base station; and The method includes the step of receiving a second CSI report related to the second report setting from the terminal by the base station; The above second CSI report includes fields related to SGCS (Square Generalized Cosine Similarity) values, and Each code point of the above fields represents a range related to SGCS quantization, and The above code point is one of a plurality of code points, and The above plurality of code points include code points mapped to a plurality of ranges, and A method characterized in that the above plurality of ranges include ranges defined by quantizing a value range of greater than or equal to 1 based on a step size.

13. Regarding base stations, One or more transmitters / receivers; One or more processors; and It includes one or more memories connected to the above one or more processors and storing instructions, A base station characterized by the above instructions, based on execution by one or more processors, having the base station perform all steps of the method according to claim 12.

Images