Method and apparatus for CSI reporting
By defining separate timing conditions for AI/ML-based CSI reporting, the method clarifies terminal operations, reduces resource waste, and enhances CSI report reliability in mobile communication systems.
Patent Information
- Application Number
- PCT/KR2026/096663
- Authority / Receiving Office
- WO · WO
- Patent Type
- Applications
- Current Assignee / Owner
- Priority Date
- 2025-05-08
- Filing Date
- 2026-05-07
- Publication Date
- 2026-11-12
AI Technical Summary
In mobile communication systems, ambiguity in terminal operation arises during CSI reporting for AI/ML models due to differing timing conditions for non-AI/ML and AI/ML-based reporting, leading to potential delays and resource waste.
A method is proposed where CSI reports are transmitted based on a timing condition defined separately for AI/ML-based CSI, ensuring clarity and preventing resource waste by allowing defined information to be reported when time for inference is not secured.
This approach prevents ambiguity in terminal operation, ensures consistent communication continuity, and improves CSI report reliability by accurately reflecting processing times for AI/ML models.
Smart Images

Figure KR2026096663_12112026_PF_FP_ABST
Abstract
Description
Method and apparatus for CSI reporting
[0001] This specification relates to a method and apparatus for CSI reporting.
[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] In 6G, operations between base station terminals utilizing AI / ML are expected to expand further. For example, in 5G advance, an additional time offset value for AI / ML related CSI processing time required by AI / ML related functionality was introduced in addition to CSI processing time Z / Z'.
[0005] Meanwhile, regarding inference-related CSI reporting (e.g., CSI reporting related to inference of AI / ML models), the first number of symbols related to non-AI / ML-based reporting and the second number of symbols related to AI / ML-based reporting may be discussed in 6G as well. In such cases, ambiguity in terminal operation may arise if, in inference-related CSI reporting, the reporting time satisfies the timing condition based on the first number of symbols but does not satisfy the timing condition based on the second number of symbols. More specifically, it may be ambiguous whether the terminal should transmit an incomplete CSI report excluding the inference result or delay the reporting time until the inference result is generated.
[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 an embodiment of the present specification for solving the aforementioned problem comprises the steps of receiving a reporting setting related to Channel State Information (CSI) and transmitting a CSI report related to inference of an Artificial Intelligence / Machine Learning (AI / ML) model. The CSI report is transmitted as a valid CSI report based on the satisfaction of a timing condition based on a second number of symbols defined separately from a first number of symbols. The first number of symbols is related to a delay requirement for non-AI / ML-based CSI. The second number of symbols is related to a delay requirement for AI / ML-based CSI. The method is characterized by transmitting the CSI report containing defined information based on the satisfaction of the timing condition related to the first number of symbols and the non-satisfaction of the timing condition related to the second number of symbols.
[0009] Therefore, in CSI reporting, if the time required to derive inference results is not secured, the action to be performed can be clearly defined to prevent ambiguity in terminal operation.
[0010] According to the embodiments of the present specification, ambiguity in terminal operation can be prevented by ensuring that defined information is reported when the time required to derive an inference result is not secured. Additionally, inconsistency in interpretation between the terminal and the base station regarding the CSI reported in such a situation can be prevented.
[0011] In addition, as long as the timing condition related to the first number of symbols is satisfied, even information that can be reported at that time (i.e., defined information (e.g., non-AI / ML-based information)) can be reported, thereby preventing the waste of resources / terminal power consumed to calculate said information.
[0012] In addition, since the omission of the CSI report itself can be prevented, the base station can continuously perform control operations based on existing CSI information, thereby improving communication continuity and system stability.
[0013] In addition, by defining a second number of symbols (second Z / second Z') that reflects the time required for an AI / ML model to derive an inference output, the actual processing time required for a CSI report including the inference result can be accurately reflected. Accordingly, the reliability of a complete CSI report including the inference result can be improved.
[0014] 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.
[0015] Figure 1 is a diagram illustrating the overall functions from the perspective of an AI / ML model.
[0016] Figure 2 illustrates a general form of an AI / ML-related procedure performed between a network and a terminal.
[0017] Figure 3 illustrates an example of AI / ML-based beam management operation.
[0018] Figure 4 illustrates an example of an AI / ML-based CSI measurement / reporting operation.
[0019] Figure 5 illustrates an example of an AI / ML-based positioning operation.
[0020] Figure 6 is a flowchart showing an example of a CSI-related procedure.
[0021] Figure 7 is a diagram illustrating the CPU occupancy time associated with BM-Case 2.
[0022] FIG. 8 is a flowchart illustrating a method according to one embodiment of the present specification.
[0023] FIG. 9 is a flowchart illustrating a method according to another embodiment of the present specification.
[0024] FIG. 10 is a drawing showing the configuration of a first device and a second device according to an embodiment of the present specification.
[0025] 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."
[0026] 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."
[0027] 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."
[0028] 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."
[0029] 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."
[0030] In the following explanation, 'when, if, in case of' can be replaced with 'based on'.
[0031] Technical features described individually within a single drawing in this specification may be implemented individually or simultaneously.
[0032] 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.
[0033] 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.
[0034] 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.
[0035] 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.
[0036] 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.
[0037] < AI / ML for Wireless Communication >
[0038] 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.
[0039] - 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.
[0040] - 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.
[0041] - 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.
[0042] - 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.
[0043] 1. Life Cycle Management (LCM) for AI / ML models
[0044] 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.
[0045] Figure 1 is a diagram illustrating the overall functions from the perspective of an AI / ML model.
[0046] 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).
[0047] 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.
[0048] 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).
[0049] 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).
[0050] 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).
[0051] The Management function (30) is a function that monitors the operation of an AI / ML model or an AI / ML function.
[0052] 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).
[0053] A Model Transfer / Delivery Request (33) can be used to request model(s) from Model Storage (50).
[0054] 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).
[0055] 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).
[0056] 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.
[0057] The Model Storage function (50) is a function that stores a trained / updated model that can be used to perform the Inference function (40).
[0058] Model Transfer / Delivery (51) is used to transfer an AI / ML model to an inference function.
[0059] 2. General AI / ML related procedures between the network and the terminal
[0060] 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.
[0061] (1) AI / ML related setup procedure
[0062] 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.
[0063] 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).
[0064] (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.
[0065] (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.
[0066] (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.
[0067] (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.
[0068] 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.
[0069] (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.
[0070] (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.
[0071] (2) Operation based on inference by AI / ML models
[0072] 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.
[0073] (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.
[0074] (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.
[0075] (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.
[0076] (3) Procedures for AI / ML management
[0077] The network and / or terminal can perform procedures for the management of AI / ML Functionality / model or the settings therefor (B15).
[0078] 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).
[0079] 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.
[0080] 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.
[0081] 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.
[0082] 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.
[0083] 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.
[0084] 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.
[0085] 3. Specific operation examples based on AI / ML model inference
[0086] (1) Beam management
[0087] Figure 3 illustrates an example of AI / ML-based beam management operation.
[0088] 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.
[0089] The network / terminal can perform measurements on the first set of beams (C10). The beam measurements may be related to RSRP measurements.
[0090] 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.
[0091] According to an embodiment, the network / terminal can transmit and receive information about the acquired second set of beams.
[0092] Specifically, AI / ML-based beam management operations may include at least one of the following BM-Case 1 and BM-Case 2.
[0093] - BM-Case 1: Prediction of the second set of DL beams in the spatial domain using the first set of beam measurements
[0094] - BM-Case 2: Prediction of the second set of DL beams in the time domain through the first set of beam measurements
[0095] 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.
[0096] 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.
[0097] 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.
[0098] 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.
[0099] 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.
[0100] 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.
[0101] 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.
[0102] Regarding UE-assisted performance monitoring for the UE-side models of BM-Case 1 and 2, the following methods may be considered.
[0103] i) Compare prediction results based on resources for monitoring and use the top 1 or top K beam prediction accuracy.
[0104] 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.
[0105] iii) Use the difference information between the measured RSRP and the predicted RSRP for the corresponding beam of the resources for monitoring
[0106] iv) Probability information that the predicted beam will become one of the top 1 or N beams
[0107] 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.
[0108] For BM-Case 2 of the UE-side model, the network can be configured to report inferences about N future times to the terminal.
[0109] (2) CSI prediction and / or compression
[0110] Figure 4 illustrates an example of an AI / ML-based CSI measurement / reporting operation.
[0111] 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.
[0112] 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.
[0113] 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.
[0114] The network can acquire CSI based on the terminal's CSI report.
[0115] 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.
[0116] 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.
[0117] 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.
[0118] (3) Positioning
[0119] Figure 5 illustrates an example of an AI / ML-based positioning operation.
[0120] Referring to FIG. 5, the network / terminal can perform a setup procedure related to AI / ML-based positioning (E05). The network / terminal can perform measurements for positioning (E10). The measurements for positioning may be related to PRS and / or SRS measurements. Based on the measurement results, the network / terminal can obtain information regarding terminal positioning (E15). For example, the network / terminal can perform AI / ML inference by using the measurement results for PRS / SRS as AI / ML input data. The information regarding 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.
[0121] < CSI Related Actions >
[0122] 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).
[0123] 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).
[0124] 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).
[0125] Figure 6 is a flowchart showing an example of a CSI-related procedure.
[0126] 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).
[0127] 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, CSI-RS resource information (e.g., M≥1 CSI-ResourceConfig resource setting), 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.
[0128] 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.
[0129] 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.
[0130] 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.
[0131] 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.
[0132] 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 Prediction Accuracy Indicator (PAI) (or RS-PAI).
[0133] 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).
[0134] 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).
[0135] For example, the reportQuantity parameter can be set to pai (or rs-pai). pai (or rs-pai) represents PA (or RS-PAI).
[0136] 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.
[0137] 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).
[0138] resource setting
[0139] 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).
[0140] Next, one or more CSI resource settings for channel measurement (CM) and interference measurement (IM) are established through higher layer signaling.
[0141] - CSI-IM resource for interference measurement.
[0142] - NZP CSI-RS resources for interference measurement.
[0143] - NZP CSI-RS resources for channel measurement.
[0144] 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.
[0145] Here, CSI-IM (or ZP CSI-RS for IM) is primarily used for inter-cell interference measurements.
[0146] Also, the NZP CSI-RS for IM is mainly used for intra-cell interference measurement from multi-users.
[0147] 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.
[0148] As examined, resource setting can refer to a resource set list.
[0149] 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.
[0150] 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.
[0151] Beam Management (BM)
[0152] 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.
[0153] - Beam measurement: An operation in which a base station or UE measures the characteristics of a received beamforming signal.
[0154] - Beam determination: The operation in which a base station or UE selects its transmit beam (Tx beam) / receive beam (Rx beam).
[0155] - Beam sweeping: An operation that covers a spatial area using transmitting and / or receiving beams for a set time interval in a predetermined manner.
[0156] - Beam report: An operation in which the UE reports information about the beam-formed signal based on beam measurements.
[0157] 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).
[0158] In addition, each BM procedure may include Tx beam sweeping to determine the Tx beam and Rx beam sweeping to determine the Rx beam.
[0159] DL BM
[0160] 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.
[0161] Here, beam reporting may include preferred DL RS ID(identifier)(s) and the corresponding L1-RSRP(Reference Signal Received Power).
[0162] The above DL RS ID may be SSBRI (SSB Resource Indicator) or CRI (CSI-RS Resource Indicator).
[0163] An example of beam forming using SSB and CSI-RS will be examined in detail below.
[0164] 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.
[0165] 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.
[0166] The DL BM procedure is examined below.
[0167] Configuration for beam reporting using SSB is performed during CSI / beam configuration in the RRC connected state (or RRC connected mode).
[0168] - 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.
[0169] 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.
[0170]
[0171] 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.
[0172] - The terminal receives a DownLink Reference Signal (DL RS) from the base station. As a specific example, the terminal receives an SSB resource from the base station based on the CSI-SSB-ResourceSetList.
[0173] - 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.
[0174] 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.
[0175] 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'.
[0176] 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.
[0177] 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.
[0178] - 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).
[0179] - 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.
[0180] - The terminal selects (or determines) the best beam.
[0181] - 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'.
[0182] Explanation regarding Rel-17 / 18 beam management >
[0183] 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.
[0184] 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.
[0185] 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.).
[0186] 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).
[0187] 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).
[0188] < Background related to AI / ML beam management >
[0189] In this document, ' / ' means 'and', 'or', or 'and / or' depending on the context.
[0190] In the Rel-18 AI / ML study item, performance analysis and potential specification impact were analyzed through evaluation when NW and / or UE-side AI / ML models were operating in three use cases: CSI compression / prediction, beam management, and positioning. In the beam management use case, the study was conducted by dividing the sub-use cases into BM-case1 and BM-case2 to analyze performance and potential specification impact regarding spatial domain beam prediction and temporal beam prediction (refer to TR38.843).
[0191] In the Rel-18 AI / ML study, discussions were held regarding NW / UE-sided AI / ML operations that predict the best beam of Set A based on Set B measurements. For UE-sided AI / ML, an operation is required where the terminal measures Set B and reports the predicted Set A beam, while for NW-sided AI / ML, an operation is required where the terminal reports the Set B measurements.
[0192] The standardization agreements for Rel-19 UE-sided AI / ML to date are as follows.
[0193] Agreement
[0194] For UE-side models, at least for BM-Case1, the following is supported regarding the content of the inference result report.
[0195] Option 1: Beam information for the predicted Top K beams within the beam set
[0196] Option 2: Beam information for the predicted Top K beams within the beam set, and the RSRP of the predicted Top K beams
[0197] At least K=1, and for the maximum value, use FFS
[0198] For beam information, FFS
[0199] For the definition of the predicted Top K beam, FFS
[0200] For the definition of the reported RSRP where applicable, FFS
[0201] For other information within the report along with potential down selections among the following options, FFS
[0202] Option 3: Beam information for the predicted Top K beams within the beam set, and probability information for the predicted Top K beams
[0203] Regarding the quantization method of probability information, FFS
[0204] The probability information is the probability that the beam will become the Top 1 or Top K beam.
[0205] Option 4: Beam information for the predicted Top K beams within the beam set, the RSRP of the predicted Top K beams, and the reliability information of the corresponding RSRPs
[0206] Regarding the definition of the reported RSRP, FFS
[0207] Regarding the definition of reliability information and quantization methods, FFS
[0208] Other options are not excluded either.
[0209] Here, the beam set is Set A, which refers to the beams for UE prediction.
[0210] Agreement
[0211] For UE-side AI / ML models in BM-Case1 and BM-Case2:
[0212] Support Type 1 Performance Monitoring, includes the following two options:
[0213] Option 1 (NW-side performance monitoring):
[0214] UE sends reports to NW (to calculate performance metrics in NW)
[0215] Measurement results from resource sets for monitoring (e.g., L1-RSRP and / or RS index) are supported as report content.
[0216] Other content is FFS
[0217] The report is set / triggered by at least NW
[0218] Note: This may or may not have an additional spec impact.
[0219] Option 2 (UE-supported performance monitoring):
[0220] UE calculates performance metrics
[0221] Regarding how to report and what to report, FFS
[0222] Whether to trigger reporting based on events for Option 1 and / or Option 2 is FFS
[0223] FFS Type 2 Performance Monitoring
[0224] Agreement
[0225] The following working assumptions have been established.
[0226] Working Assumption
[0227] In the inference result report for the UE-side model of BM-Case 2, the predicted RSRP of the beam is the predicted RSRP, and this predicted RSRP is based on the AI / ML output.
[0228] Agreement
[0229] For UE-side models, at least for the quantization of RSRP values in inference result reports, the following is supported:
[0230] Support for differential RSRP reporting using existing quantization steps and ranges for L1-RSRP reporting
[0231] For BM-Case1, differential RSRP reporting between multiple beams is supported.
[0232] For BM-Case2, support for differential RSRP reporting between multiple beams across multiple time points.
[0233] Details are FFS
[0234] Agreement
[0235] For the UE-side model, at least for BM Case-1, two resource sets can be configured separately for Set A and Set B in the CSI report configuration in the inference result report.
[0236] Whether to support configuring a resource set solely for Set B is FFS.
[0237] The UE performs measurements on the resource set of Set B for inference, and the UE is not expected to measure the resource set of Set A for inference.
[0238] The beam information in the inference report refers to the resource set of Set A.
[0239] Agreement
[0240] With respect to the UE-side AI / ML models in BM-Case1 and BM-Case2, for Option 2 (UE-supported performance monitoring), support at least Alternative 1: determine Top 1 or Top K beam prediction accuracy (with or without margins) by comparing the Top 1 or Top K beams based on the prediction results and measurements from the resource set / resources.
[0241] FFS: Detailed definition of the metric, including whether to configure or define a window for calculation
[0242] FFS: Includes other details related to how to configure resource sets / resources for monitoring, e.g.
[0243] Example: Whether / how to use the entire set of Set A for measurement. If the entire Set A is not configured, whether / how to define a metric.
[0244] FFS: Other Alternatives
[0245] Agreement
[0246] In BM-Case 2 of the UE-side model, the reference time of the earliest time instance for the prediction result considers at least the following potential down-selection alternatives:
[0247] Option 1: Based on the UL slot for reporting
[0248] Option 2: Based on CSI reference resources corresponding to the report
[0249] Option 3: Based on the latest transmission time of the CSI-RS / SSB resource within Set B for measurement for reporting, and this transmission time is not later than the CSI reference resource
[0250] Agreement
[0251] For UE-side AI / ML models, for BM-Case1, at least for inference, and for at least Set B, the following CSI-RS resource types for CMR are supported:
[0252] Periodic (P) CSI-RS
[0253] Semi-persistent (SP) CSI-RS
[0254] Aperiodic (AP) CSI-RS
[0255] For UE-side AI / ML models, for BM-Case 2, at least for inference, and for at least Set B, the following CSI-RS resource types for CMR are supported:
[0256] Periodic (P) CSI-RS
[0257] Semi-persistent (SP) CSI-RS
[0258] FFS: Aperiodic (AP) CSI-RS
[0259] Note: The above CSI-RS resources refer to resources used for beam management.
[0260] Agreement
[0261] For at least UE-side model monitoring of Monitoring Type 1 Option 2 (if applicable), consider at least the following options and include potential down-selection for monitoring configuration:
[0262] Option 1: Resource set(s) for monitoring and reporting configuration are configured within the CSI reporting configuration used for inference (if applicable).
[0263] FFS: Resource set(s) for monitoring
[0264] The UE measures resource set(s) for monitoring
[0265] FFS: When / how to report monitoring results
[0266] Option 2: Dedicated resource set(s) and reporting configuration for monitoring are configured within the dedicated CSI reporting configuration used for monitoring.
[0267] The dedicated reporting configuration used for monitoring is linked to the inference reporting configuration.
[0268] FFS: A method to check connectivity between RSs within resource set(s) for monitoring and Set A beams.
[0269] The UE measures resource set(s) for monitoring
[0270] FFS: When to report monitoring results
[0271] Agreement
[0272] For the UE-sided model, at least for BM-Case 1, the beam information in the inference result report is the CRI / SSBRI of the resource in Set A.
[0273] Agreement
[0274] For the UE-sided model for inference, in both BM-Case 1 and BM-Case 2, when Set A and Set B are configured within the CSI report configuration,
[0275] Two CSI-ResourceConfig IDs (CSI-ResourceConfigId) are configured separately for Set A and Set B, respectively.
[0276] Agreement
[0277] For Monitoring Type 1 Option 2 of UE-side model monitoring (where applicable), support is provided for reusing the CSI framework in the configuration for monitoring result reports in L1 signaling:
[0278] The dedicated resource set(s) for monitoring and the report configuration for monitoring are configured within the dedicated CSI report configuration used for monitoring.
[0279] The ID of the inference report configuration is configured within the configuration for monitoring to link the inference report configuration and the monitoring report configuration.
[0280] How to identify connections between RSs within resource set(s) for monitoring and Set A beams will be reviewed later (FFS).
[0281] Future review (FFS) on whether to support all combinations of time domain behavior of reportConfigType for inference reports and reportConfigType for monitoring reports
[0282] Timing-related issues will be reviewed later (FFS)
[0283] The terminal (UE) measures the dedicated resource set(s) for monitoring.
[0284] Agreement
[0285] For the report content of the inference results for the UE-sided model, the largest RSRP value is quantized into a 7-bit value in the range of [-140, -44] dBm with a step size of 1 dB, and the differential RSRP is quantized into a 4-bit value with a step size of 2 dB.
[0286] Note: The model output is a UE implementation matter and does not need to be RSRP subject to dBm value.
[0287] Agreement
[0288] For the report content of the inference results of the UE-sided model for BM-Case 1, the RSRP of the predicted beam(s) in the report of inference results is the predicted RSRP, where the predicted RSRP is based on the AI / ML output.
[0289] Note: How to reflect this in the specification is a separate discussion.
[0290] Agreement
[0291] For UE-side AI / ML model inference and BM-Case 2, in the quantization of RSRP values of inference results in a report across multiple future time instances, the largest RSRP value based on the prediction of all time instances is the reference RSRP, and differential RSRPs in the report are computed relative to the reference RSRP.
[0292] Time instance information of the beam with the largest RSRP is additionally indicated in the report.
[0293] Agreement
[0294] Regarding inference, in the case of BM-Case 2 of the UE-side model,
[0295] The time gap between two consecutive future time instances is configured by the RRC, and the number N of future time instance(s) is configured by the RRC.
[0296] The time gap is [10ms, 20ms, 40ms, 80ms, 160ms]
[0297] N = [1, 2, 4, 8]
[0298] The reference time for the earliest time instance of the predicted results is based on the most recent occasion of the CSI-RS / SSB resource in Set B for measurement.
[0299] Here, the most recent occasion of the CSI-RS / SSB resource of Set B is the latest CSI-RS / SSB occasion no later than the corresponding CSI reference resource of the corresponding inference report.
[0300] Agreement
[0301] For the UE-sided model, the following is supported for the configuration of resources for data collection purposes:
[0302] CSI-ReportConfig can be used to configure resources for data collection purposes without CSI reporting.
[0303] One CSI-ResourceConfigId is configured for Set A.
[0304] One CSI-ResourceConfigId is configured for Set B.
[0305] Note: The UE performs measurement on all resources.
[0306] One or two associated IDs can be configured within the CSI-ReportConfig.
[0307] If Set B is identical to Set A or is a subset of Set A (i.e., if the NZP-CSI-RS-ResourceId / SSB-Index in the resource set for Set B is within the NZP-CSI-RS-ResourceId / SSB-Index in the resource set for Set A), one associated ID is constructed.
[0308] Otherwise, one associated ID is configured for Set A, and another one associated ID is configured for Set B.
[0309] Future Review (FFS): Whether to support aperioditic CSI RS and / or how to support
[0310] Note: This is not related to whether or how the delivery / transmission of the collected data for training the UE-sided model is supported.
[0311] Agreement
[0312] For the UE-sided model, in the CSI-ReportConfig for inference,
[0313] One or two associated IDs can be configured within the CSI-ReportConfig.
[0314] If Set B is identical to Set A or is a subset of Set A (i.e., if the NZP-CSI-RS-ResourceId / SSB-Index in the resource set for Set B is within the NZP-CSI-RS-ResourceId / SSB-Index in the resource set for Set A), one associated ID is constructed.
[0315] Otherwise, one associated ID is configured for Set A, and another one associated ID is configured for Set B.
[0316] Further Review (FFS): Applicability of aperioditic CSI RS for at least BM-Case 1
[0317] Agreement
[0318] For the UE-sided model, regarding resource types for data collection purposes, only always-on SSB and P / SP CSI-RS are supported.
[0319] Agreement
[0320] Regarding BM-Case1, regarding the Top K beam(s) report as inference results,
[0321] K is configured for the terminal (UE) in the inference report configuration.
[0322] Agreement
[0323] For BM-Case 2 of the UE-side model, one RRC parameter represents a time gap set between two consecutive future time instances, and also represents a time gap between a reference time for prediction and a first future time instance.
[0324] In the NR standard, when a terminal performs CSI-related reporting, the CSI-ResourceConfig for channel / interference measurement is associated / linked with the CSI-ReportConfig setting for CSI reporting, allowing the terminal to perform CSI / beam reporting based on DL RS measurements. In this case, depending on the terminal's parallel processing capability, the maximum number of CSI measurements / reports that can be performed simultaneously is N. cpuAs a value, it can be reported to the base station through UE capability reporting, and depending on what report value various CSI reports (i.e., CSI-ReportConfig) have (e.g., reportQuantity set respectively for CSI reporting or beam reporting), the amount of CSI processing unit (CPU) value occupied (i.e., O cpu It is defined by the standard regarding ). In addition, these specific CSI-related reports are O cpu The standard also defines a timeline regarding the period of occupation, specifically from which point in time to which point. The CSI processing criteria are explained in detail below.
[0325] 5.2.1.6 CSI Processing Criteria
[0326] The terminal (UE) is the number of simultaneous CSI calculations N supported by the parameter simultaneousCSI-ReportsPerCC or [simultaneousCSI-SubReportsPerCC-r18] on a single component carrier. CPUIndicates [simultaneousCSI-ReportsAllCC] or [simultaneousCSI-SubReportsAllCC-r18] across all component carriers. If, on one component carrier, the terminal (UE) is configured with at least one CSI report setting having a sub-configuration, the terminal (UE) must use the parameter [simultaneousCSI-SubReportsPerCC-r18] on that component carrier. Otherwise, the terminal (UE) must use simultaneousCSI-ReportsPerCC on that component carrier. If, on any one component carrier, the terminal (UE) is configured with at least one CSI reporting setting having a sub-configuration, the terminal (UE) must use [simultaneousCSI-SubReportsAllCC-r18]. Otherwise, the terminal (UE) must use simultaneousCSI-ReportsAllCC. If the terminal (UE) is N CPU If simultaneous CSI calculations are supported, the terminal (UE) is for processing N CSI reports. CPU It is said to have CSI processing units. If L CPUs are occupied for the calculation of CSI reports in a given OFDM symbol, the terminal (UE) has N CPU - It has L unoccupied CPUs. There are N CSI reports, N CPU-L CPUs start occupying their respective CPUs in the same unoccupied OFDM symbol, where each CSI report n=0,..,N-1 In the case corresponding to , the terminal (UE) is not required to update NM of requested CSI reports with the lowest priority (in accordance with Section 5.2.5). Here, 0≤M≤N, and M is It is the largest value that satisfies .
[0327] The terminal (UE) is N CPU It is not expected to consist of an aperioditic CSI trigger state containing more than a certain number of reporting settings. Processing of a CSI report occupies a certain number of CPUs for a certain number of symbols as follows.
[0328] - O CPU For a CSI report having a CSI-ReportConfig with a CSI-RS-ResourceSet configured with the higher layer parameter trs-Info, where =0, the higher layer parameter reportQuantity is set to 'none'
[0329] - O CPUFor a CSI report (CSI report) with LTM-CSI-ReportConfig =1, or a CSI report (CSI report) with CSI-ReportConfig where the upper tier parameter reportQuantity is set to 'cri-RSRP', 'ssb-Index-RSRP', 'cri-SINR', 'ssb-Index-SINR', 'cri-RSRP-Index', 'ssb-Index-RSRP-Index', 'cri-SINR-Index', 'ssb-Index-SINR-Index' or 'none' (and CSI-RS-ResourceSet where the upper tier parameter trs-Info is not configured
[0330] - O CPU =(Y+1)·X, for a CSI report having a CSI-ReportConfig where the upper layer parameter reportQuantity is set to 'tdcp' and the number of delays Y configured by the upper layer parameter Y is given, where the value of X∈{1,2} is reported by the terminal capability (UE capability).
[0331] - For a CSI report (CSI report) having a CSI-ReportConfig where the upper-level parameter reportQuantity is set to 'cri-RI-PMI-CQI', 'cri-RI-i1', 'cri-RI-i1-CQI', 'cri-RI-CQI', or 'cri-RI-LI-PMI-CQI',
[0332] - If max{ μPDCCH, μCSI-RS, μUL} ≤ 3 and L=0 CPUs are occupied, if CSI corresponds to a single CSI and corresponds to wideband frequency-granularity, and corresponds to up to 4 CSI-RS ports within a single resource without a CRI report, and if codebookType is set to 'typeI-SinglePanel' or reportQuantity is set to 'cri-RI-CQI', and a CSI report is triggered non-periodically without transmitting a transport block or a PUSCH containing HARQ-ACK or both, O CPU =N CPU ,
[0333] - If CSI-ReportConfig is configured so that codebookType is set to 'typeI-SinglePanel', and the corresponding CSI-RS resource set for channel measurement consists of 2 resource groups and N resource pairs, O CPU =X·N+M, where X is the number of CPUs occupied by a pair of CMRs that are the target of mTRP-CSI-numCPU-r17, and M is defined in Section 5.2.1.4.2.
[0334] - If CSI-ReportConfig contains a list of L sub-configurations provided by the parent parameter csi-ReportSubConfigToAddModList,
[0335] - , in the case of periodic CSI reporting, here is the total number of CSI-RS resources corresponding to the i-th sub-configuration.
[0336] - , in the case of aperioditic and semi-persistent CSI reporting, here is the total number of CSI-RS resources corresponding to the i-th sub-configuration, where the i-th sub-configuration is from the specified N sub-configurations among the L sub-configurations included in CSI-ReportConfig, where N≤L and N≥1.
[0337] - CSI-ReportConfig is configured so that 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', and the corresponding NZP-CSI-RS-ResourceSet for channel measurement is 1 <N_TRP≤4개의 자원(resources)으로 구성되는 경우, O CPU =ceil(X·N TRP ), where X∈{1,1.5,2} is reported by the UE capability indication.
[0338] - If CSI-ReportConfig is configured such that the upper-level parameter reportQuantity is set to 'cri-RI-PMI-CQI' and codebookType is set to 'typeII-Doppler-r18' or 'typeII-Doppler-PortSelection-r18',
[0339] - If the corresponding CSI-RS resource set for channel measurement is aperioditic and consists of K CSI-RS resources, then when K=12, O CPU =8 and when K<12 O CPU =Y_1·K, where Y_1∈{1,2,3} is reported by the UE capability indication.
[0340] - If the corresponding CSI-RS resource set for channel measurement is periodic or semi-persistent and consists of a single CSI-RS resource, then when N_4=1, O CPU =4 and when N_4>1 O CPU =max(Y_2·N_4, 4), where the value of N_4 is configured by the upper layer parameter N4, and Y_2∈{1,2,3} is reported by the UE capability indication.
[0341] - In other cases (otherwise), O CPU =K s , here K s is the number of CSI-RS resources within the CSI-RS resource set for channel measurement.
[0342] For a CSI report with a CSI-ReportConfig where the upper layer parameter reportQuantity is not set to 'none', or for a CSI report with an LTM-CSI-ReportConfig, the CPU(s) are occupied for a certain number of OFDM symbols as follows.
[0343] - Periodic or semi-persistent CSI reports (excluding initial semi-persistent CSI reports on PUSCH following a PDCCH that triggers the report, and semi-persistent CSI reports on PUSCH configured so that the upper-level parameter codebookType is set to 'typeII-Doppler-r18' or 'typeII-Doppler-PortSelection-r18') are periodic CSI reports corresponding to CSI-ReportConfig containing a list of subconfigurations provided by csi-ReportSubConfigToAddModList, starting from the first symbol of the earliest of each CSI-RS / CSI-IM / SSB occasion not later than the corresponding CSI reference resource for channel or interference measurement. Starting from the first symbol of the earliest of each CSI-RS / CSI-IM resource associated with all configured sub-configurations for the report,Or, occupy CPU(s) from the first symbol of the earliest of each CSI-RS / CSI-IM resource associated with all activated / triggered sub-configurations for a semi-persistent CSI report corresponding to a CSI-ReportConfig containing a list of sub-configurations provided by csi-ReportSubConfigToAddModList, to the last symbol of the configured PUSCH / PUCCH carrying the report.
[0344] - An aperiodic CSI report occupies CPU(s) from the first symbol after the PDCCH that triggers the CSI report until the last symbol of the scheduled PUSCH carrying the report. If the PDCCH reception contains two PDCCH candidates from each of the two search space sets, as described in Section 10.1 of [6, TS 38.213], the PDCCH candidate that ends later in time is used for the purpose of determining the CPU occupation duration.
[0345] - An initial semi-persistent CSI report on PUSCH following a PDCCH trigger occupies CPU(s) from the first symbol after the PDCCH until the last symbol of the scheduled PUSCH carrying the report. If the PDCCH reception contains two PDCCH candidates from each of the two search space sets, as described in Section 10.1 of [6, TS 38.213], the PDCCH candidate that ends later in time is used for the purpose of determining the CPU occupation duration.
[0346] - A semi-persistent CSI report on PUSCH configured such that the upper-level parameter codebookType is set to 'typeII-Doppler-r18' or 'typeII-Doppler-PortSelection-r18' occupies CPU(s) from the first symbol of the KP-th latest among consecutive periodic / semi-persistent CSI-RS occasions that are not later than the CSI reference resource to the last symbol of PUSCH carrying the report. Here, K P∈ The value of {1,2,4} is determined by the terminal capability (UE capability).
[0347] For a CSI report with a CSI-ReportConfig having a CSI-RS-ResourceSet where the upper layer parameter reportQuantity is set to 'none' and the upper layer parameter trs-Info is not configured, the CPU(s) are occupied for a certain number of OFDM symbols as follows.
[0348] - Semi-persistent CSI reports (excluding initial semi-persistent CSI reports on PUSCH following the PDCCH that triggers the report) are from the first symbol of the earliest of each transmission occasion of the periodic or semi-persistent CSI-RS / SSB resources for channel measurement for L1-RSRP computation, to the last symbol of the latest of the CSI-RS / SSB resources for channel measurement for L1-RSRP computation in each transmission occasion. Occupies CPU(s) up to the symbol.
[0349] - An aperiodic CSI report is from the first symbol after the PDCCH that triggers the CSI report, the Z3 symbol after the first symbol after the PDCCH that triggers the CSI report, and after the last symbol of the latest of each CSI-RS / SSB resource. CPU(s) are occupied up to the last symbol among the symbols. Here, the aforementioned CSI-RS / SSB resources are for channel measurement for L1-RSRP calculation.
[0350] Here (Z3, ) is defined in Table 3.
[0351] In the Rel-19 AI / ML work item, discussions are underway regarding the possibility that when a terminal utilizes UE-side AI / ML to perform CSI reporting, operations similar to those of a CPU may be performed in AI / ML-based CSI reporting, as the terminal's AI / ML-related parallel computing power or computing resources (e.g., memory, hardware capability of GPUs for inference / actuation, etc.) may be limited. The relevant consensus points are as follows.
[0352] Agreement @ 9.1.1 Specification support for beam management
[0353] For the UE-side model, the existing CPU mechanism is used as a starting point for AI / ML-based CSI processing.
[0354] Future review (FFS) on whether to share overall CPU or count separately between legacy CSI reporting and AI / ML-based CSI reporting, and between AI / ML features / functionalities.
[0355] Further review (FFS) is required regarding whether it is fully applicable to BM-Case 1 and / or BM-Case 2.
[0356] Agreement @ 9.1.1 Specification support for beam management
[0357] For the UE-sided model, regarding the CSI report corresponding to the CSI-ReportConfig for Type 1 option 2 monitoring, O CPU =1.
[0358] Note: The occupation duration is a separate discussion.
[0359] Agreement @ 9.1.1 Specification support for beam management
[0360] For the UE-side model, regarding AI / ML-based beam management for BM-Case 1 and BM-Case 2, and in relation to the processing of CSI reports for inference, consider the following options for potential down selection:
[0361] Option 1: Only dedicated AI / ML PUs are occupied, and O APU =N1 is reported by the terminal (UE).
[0362] And O CPU =0.
[0363] Option 2: Only the legacy CPU is occupied, and O CPU =M is reported by the terminal (UE).
[0364] Option 3: Both the dedicated AI / ML PU and the legacy CPU are occupied, and O APU =N2 is reported by the terminal (UE).
[0365] And O CPU =1.
[0366] Note: If multiple options are supported, the supported option by the terminal (UE) is reported by the terminal capability (UE capability).
[0367] The total number of dedicated AI / ML PUs for AI / ML is reported by UE capability.
[0368] Note: The total number of use case-specific dedicated AI / ML PUs may be discussed separately.
[0369] Agreement @ 9.1.3 Specification support for CSI prediction
[0370] Regarding CSI prediction using a UE-side model, further study is conducted on the following matters concerning CSI processing criteria and timelines, at least with respect to inference.
[0371] - Whether to share CPUs or count them separately between legacy CSI reporting and AI / ML-based CSI reporting
[0372] - Whether to share processing units among AI / ML related features / functionalities or to count them separately
[0373] - Whether a new timeline is required for inference / whether it is updated, and whether a different timeline is required when functionality switches / activates
[0374] - Whether the legacy framework for active CSI-RS resources and port counting can be reused
[0375] Note: We strive to research CSI processing criteria that consider both BM cases and CSI cases, and use existing solutions as a starting point.
[0376] For example, an AI / ML based processing unit, an AI / ML related processing unit, or an APU refers to the maximum processing unit or / and maximum processing resource (cf. N) related to AI / ML that the terminal reports (via UE capability).cpu The value occupied / consumed by a specific terminal-side AI / ML functionality / model within the legacy NR) value (cf, O cpu It may mean in legacy NR). The APU value occupied / consumed by a specific functionality / model of such a terminal may be expressed as an integer value or a real value, as a multiple of a specific unit value, or as one or more resources / units.
[0377] For example, 'APU' refers to a CPU involved in the processing of AI / ML model-based CSI reports (e.g., CSI reports related to prediction or CSI reports related to performance monitoring (e.g., prediction accuracy)). In other words, 'APU' is merely a term to distinguish it from conventional CPUs, and is not intended to limit the technical concept according to the embodiments of this specification to 'APU'. As a specific example, the CPU involved in the processing of conventional CSI reports may be referred to as the 'first CPU (or second CPU)', and the CPU involved in the processing of the aforementioned AI / ML model-based CSI reports may be referred to as the 'second CPU (or first CPU)'. As a specific example, the number of first CPU(s) occupied for the processing of the CSI reports is O CPU Represented as, and the number of second CPU(s) occupied for processing the aforementioned AI / ML model-based CSI report is O CPU,x (e.g., O CPU,1 , O CPU,2 , O CPU,3 It can be expressed as .. etc.
[0378] Additionally, the standard defines the minimum processing time of the terminal in relation to the CSI report as follows. The values of Z and Z' may be defined in the standard or reported by the terminal UE capability report.
[0379] First, we examine the previously defined behavior regarding CSI processing time.
[0380] When a CSI request field on DCI triggers CSI report(s) on PUSCH, the UE must provide a valid CSI report for the nth triggered report in the following cases:
[0381] The first uplink symbol carrying the relevant CSI report(s), including the effect of the timing advance, is symbol Z ref If you do not start earlier, and
[0382] The first uplink symbol carrying the nth CSI report, including the effect of the timing advance, is symbol Z' ref If it does not start earlier than (n).
[0383] Here, Z ref is from the end of the last symbol of the PDCCH that triggers the CSI report(s). It is subsequently defined as the next uplink symbol where the CP begins. Z' ref (n) is from the end of the last symbol of the latest in time among the following when aperiodic CSI-RS is used for channel measurement for the nth triggered CSI report It is subsequently defined by the following uplink symbol where the CP begins: an aperiodic CSI-RS resource for channel measurement, an aperiodic CSI-IM used for interference measurement, and an aperiodic NZP CSI-RS for interference measurement for CSI-ReportConfig (or for all triggered subconfigurations if CSI-ReportConfig includes multiple subconfigurations). switch is defined in Section 6.4 and of Table 2 This applies only when applicable.
[0384] If the PUSCH directed by the DCI overlaps with another PUCCH or PUSCH, the CSI report(s) are multiplexed in accordance with the procedures of Section 9.2.5 of [6, TS 38.213] and, where applicable, Section 5.2.5; otherwise, the CSI report(s) are transmitted over the PUSCH directed by the DCI.
[0385] When a CSI request field on DCI triggers CSI report(s) on PUSCH, the first uplink symbol carrying said CSI report(s), including the effect of the timing advance, is symbol Z ref If you start earlier:
[0386] - If HARQ-ACK or transport blocks are not multiplexed on PUSCH, the UE can ignore the scheduling DCI.
[0387] When a CSI request field on DCI triggers a CSI report(s) on PUSCH, the first uplink symbol carrying the nth CSI report, including the effect of the timing advance, is symbol Z' ref If you start earlier than (n):
[0388] - If the number of triggered reports is one and HARQ-ACKs or transmission blocks are not multiplexed on PUSCH, the UE can ignore the scheduling DCI.
[0389] - Otherwise, the UE is not required to update the CSI for the nth triggered CSI report.
[0390] As described in Section 10.1 of [6, TS 38.213], if a PDCCH reception contains two PDCCH candidates from two respective sets of search spaces, the PDCCH candidate ending later in time is used to determine the last symbol of the PDCCH that triggers the CSI report(s).
[0391] Z, Z' and μ are defined as follows:
[0392] and and, where M is the number of CSI report(s) updated in accordance with Section 5.2.1.6, and corresponds to the m-th updated CSI report and is defined as follows:
[0393] - Table 2's , max{ μ PDCCH , μ CSI-RS , μ UL} ≤ 3, and when L=0 CPU is occupied (according to Section 5.2.1.6), a CSI is triggered without a PUSCH having a transmission block or HARQ-ACK or both, the CSI to be transmitted is a single CSI and corresponds to a wideband frequency-granularity corresponding to up to 4 CSI-RS ports within a single resource without CRI reporting, and CodebookType is set to 'typeI-SinglePanel' or reportQuantity is set to 'cri-RI-CQI'.
[0394] - Table 3 , if the CSI to be transmitted corresponds to broadband frequency granularity corresponding to up to 4 CSI-RS ports within a single resource without CRI reporting, and CodebookType is set to 'typeI-SinglePanel' or reportQuantity is set to 'cri-RI-CQI'.
[0395] - Table 3 , if the CSI to be transmitted corresponds to a broadband frequency granularity and reportQuantity is set to 'ssb-Index-SINR', 'cri-SINR', 'ssb-Index-SINR-Index', or 'cri-SINR-Index'.
[0396] - Table 3 , if reportQuantity is set to 'cri-RSRP', 'ssb-Index-RSRP', 'cri-RSRP-Index', or 'ssb-Index-RSRP-Index'. Here follows the capability beamReportTiming reported by the UE and KB l ...depends on the UE reported capability beamSwitchTiming defined in [13, TS 38.306], or when CSI reporting is configured with LTM-CSI-ReportConfig for L1-RSRP measurements.
[0397] - Table 3 or , depending on the UE reported capability, the codebookType is set to 'typeII-CJT-r18' or 'typeII-CJT-PortSelection-r18' and the corresponding NZP-CSI-RS-ResourceSet for channel measurements If composed of resources.
[0398] - Table 3 Using , CSI report It is configured as such, and the codebookType is set to 'typeII-Doppler-r18' or 'typeII-Doppler-PortSelection-r18', and the corresponding NZP-CSI-RS-ResourceSet for channel measurement is non-periodic with K CSI-RS resources.
[0399] - Table 3 Using , here =56.(K P -1) or 56.K P It is a symbol, and depends on the reported UE capabilities, The ∈ {1,2,4} value is indicated by the UE capability, and the CSI report is Configured as such, where codebookType is set to 'typeII-Doppler-r18' or 'typeII-Doppler-PortSelection-r18', and the corresponding NZP-CSI-RS-ResourceSet for channel measurement is periodic or semi-persistent having a single CSI-RS resource.
[0400] - Table 3 Using or , depending on UE reported capability, CSI reporting It is configured as such, and the codebookType is set to 'typeII-Doppler-r18', and the corresponding NZP-CSI-RS-ResourceSet for channel measurement is non-periodic with K CSI-RS resources.
[0401] - Table 3 Using or , depending on UE reported capability, CSI reporting Configured as such, with codebookType set to 'typeII-Doppler-r18', and the corresponding NZP-CSI-RS-ResourceSet for channel measurement having a single CSI-RS resource in the case of periodic or semi-static.
[0402] - In other cases, Table 3 .
[0403] - μ in Tables 2 and 3 is min (μ PDCCH , μ CSI-RS , μ UL Corresponds to ). Here, μ PDCCH corresponds to the subcarrier interval of the PDCCH transmitted by DCI, and μ UL corresponds to the subcarrier interval of the PUSCH to which the CSI report will be transmitted, and μ CSI-RS It corresponds to the minimum subcarrier interval of the non-periodic CSI-RS triggered by DCI.
[0404]
[0405] Table 2 shows CSI computation delay requirement 1.
[0406]
[0407] Table 3 shows CSI computation delay requirement 2.
[0408] Meanwhile, the following discussions are underway regarding the CPU and AI / ML-related processing units occupied by the inference result reporting settings (including BM-case1 and BM-case2) utilizing the UE-side mode of Rel-19 AI / ML BM and CSI prediction.
[0409] To summarize the discussion, since the CSI report configuration for performance monitoring reports performs operations similar to existing beam measurement / reports, O CPU It was agreed to occupy = 1, and in the case of CSI report settings for UE-side model data collection, O CPU Discussions are being held regarding the operation between base station terminals occupying 0. Additionally, discussions are underway regarding the simultaneous occupation of the AI / ML-related processing unit and CPU in the case of report settings for reporting inference results. At this time, the discussion regarding the CPU occupied by the report settings for reporting inference results and the timeline of the AI / ML PU is explained with reference to Fig. 7.
[0410] Figure 7 is a diagram illustrating the CPU occupancy time associated with BM-Case 2.
[0411] Specifically, FIG. 7 illustrates the CSI processing criteria in BM-Case 2. Referring to FIG. 7, in the CSI report settings for inference related to BM-case 2 as described above, the terminal performs observations (e.g., reception) for multiple transmission occasions related to Set B. Prediction can be performed by utilizing the result values of these multiple measurement occasions as inference inputs. Specifically, the terminal can report predicted Top-K beam(s) (e.g., K CRI(s) or K SSBRI(s)) and / or corresponding predicted RSRP values for one or more future instances. The CSI report settings [represent] the AI / ML-related PU value O from the last transmission occasion of Set B resource(s) prior to the slot corresponding to / corresponding to the CSI reference resource. APU Occupies the value.
[0412] For example, the occupancy may be released at the last symbol where the reporting of the corresponding CSI report setting ends. This is during the period when the terminal performs inference based on the results of the measurements O APU It can be assumed that APU(s) are occupied.
[0413] Occupying (legacy) O in the CSI report settings for inference related to inference BM-case2 CPU The timeline is discussed as shown in candidates 1 to 3 of Fig. 7. Candidate 1 continuously O from the time the observation is performed until the time the reporting of the setting ends. CPUIt is a method of occupying. Candidate 2 is the O mentioned above. APU O that has the same timeline as the value occupancy CPU This is the occupancy method. Candidate 3 is O only for terminal reporting, base station setup, and / or for a certain period of time defined by the specification, starting from each Set B measurement time point corresponding to the observation instance. CPU Possessing O APU O even while the value is occupied CPU This is the method by which the value is occupied.
[0414] Meanwhile, in addition to the Z / Z' value related to NR legacy CSI reporting, a new Z / Z' value may also be defined for CSI reporting related to AI / ML inference. Below, Z / Z' related to NR legacy CSI reporting and the new Z / Z' related to CSI reporting related to AI / ML inference will be referred to as the first Z / Z' and the second Z / Z' (or first Z / Z'). In defining the second Z / Z', the minimum time requirement for inference of the UE-side AI / ML model may need to be considered. In addition to the time required to input data into the UE-side AI / ML model and output the result to derive the inference result, the time required to load the UE-side AI / ML model into the actuator or retrieve it from memory (depending on whether it is already loaded into an actuator such as a CPU / GPU) may also need to be considered. The aforementioned problem is referred to as Problem 1 below.
[0415] Additionally, CSI reports related to UE-side model inference of the terminal may occupy both the CPU and the APU. In such cases, if the transmission time of the CSI report satisfies both the first Z / Z' and the second Z / Z', the CSI report may be performed based on the output based on UE-side model inference (in other words, the terminal may provide / transmit a valid CSI report to the base station). If the transmission time of the CSI report does not satisfy both the first Z / Z' and the second Z / Z', it may operate as follows. For example, the terminal may ignore the scheduling DCI (of the aperioditic / semi-persistent CSI report). For example, the terminal may report a non-updated CSI (if more than 1 CSI report is triggered or multiplexed). Specifically, the terminal may not update the n-th CSI report at the time of the report and may perform the (n-1)-th inference result report as is.
[0416] However, if the transmission timing of the CSI report related to UE-side model inference set / instructed from the base station satisfies the first Z / Z' but does not satisfy the second Z / Z', a problem (Problem 2) occurs in which the terminal reporting operation becomes ambiguous.
[0417] Meanwhile, regarding NR legacy CSI reporting, since the UE buffer / memory for measured values is limited when many DL RS are measured and reported simultaneously, standards for CSI processing units (CPUs) and active resource counting have been established to limit such simultaneous CSI measurement and reporting. In Rel-18 MIMO CSI prediction, CSI reporting is performed based on one or more CSI-RS measurement instances; therefore, when reporting related to Rel-18 CSI prediction, active resource counting during the period when CSI measurement and reporting are performed is K as follows. P Standardization proceeded through the occupation of ships.
[0418] For periodic or semi-persistent CSI-RS resources within a CSI-RS resource set for channel measurement linked to a CSI-ReportConfig where the higher layer parameter codebookType is set to 'typeII-Doppler-r18' or 'typeII-Doppler-PortSelection-r18', the CSI-RS resource and the CSI-RS ports within the said CSI-RS resource are K P counted K P times) is counted, where K P The value of ∈{1,2,4} is indicated by the terminal capability (UE capability).
[0419] Similarly, since multiple measurement instances are required for reporting inference results in the BM-case 2 and CSI prediction operations of Rel-19 AI / ML beam prediction, active resource counting for such inference-related CSI reports needs to be redefined. This is referred to as Problem 3.
[0420] Note: The period of active resource occupancy according to the P / SP / AP CSI report is as follows.
[0421] Time duration for active P / SP CSI-RS
[0422] - Legacy for P CSI-RS: Starts when periodic CSI-RS is configured by higher layer signaling, and terminates when the periodic CSI-RS configuration is released.
[0423] - Legacy for SP CSI-RS: Starts at the end point when the activation command is applied and ends at the end point when the deactivation command is applied.
[0424] Time duration for active AP CSI-RS
[0425] - Legacy: Starts at the end of the PDCCH containing the request and ends at the end of the scheduled PUSCH containing the report related to this aperiodic CSI-RS.
[0426] Based on this background, the present specification proposes a method for reporting an inference result by considering the minimum processing time related to AI or / and non-AI when a terminal performs DL Tx beam prediction of BM-case1 / BM-case2 using UE-sided AI / ML, and proposes a subsequent terminal operation.
[0427] In this specification, ' / ' may be interpreted as 'and', 'or', or 'and / or' depending on the context.
[0428] Proposal 1
[0429] (Including the solution to problem 1) The first Z / Z' and second Z / Z' of a specific terminal may be determined based on at least one of the following methods (based on base station settings / instructions or / and specification definitions).
[0430] Method 1)
[0431] For CSI reports related to AI / ML inference result reporting (e.g., beam prediction for BM-case1 / 2, CSI prediction), the minimum processing time for the CSI report may need to be defined by considering the time required to derive the inference output in addition to the minimum processing time 1 Z / Z' required for the NR legacy CSI report. 2 Z / Z' can be defined as [2 Z / Z' = 1 Z / Z' + new Z / Z' (or offset value delta_Z / delta_Z')].
[0432] Method 2)
[0433] In the NR legacy CSI report, since the existing definition of Z already includes a minimum processing time corresponding to Z', a new value can be defined for the second Z'. The second Z can be defined as i) [second Z = first Z + max(second Z' - first Z', 0)] or / and ii) [second Z = first Z + newly defined second Z']. For example, regarding i), parallel processing between the legacy CSI-related terminal operation and the AI / ML inference operation can be assumed. For example, regarding ii), serial processing between the legacy CSI-related terminal operation and the AI / ML inference operation can be assumed.
[0434] Method 3)
[0435] As described in Problem 1, the time required to perform AI / ML inference and derive an inference output may vary depending on whether the UE-side AI / ML model is loaded into the actuator or / and the AI engine (e.g., AI / ML model / functionality active state) or not (e.g., AI / ML model / functionality inactive state). Accordingly, the second Z / Z' values may be defined differently based on whether the CSI-ReportConfig related to the AI / ML inference result report of a specific terminal is in an active state or an inactive state. Additionally, information regarding the active / inactive state of the AI / ML model / functionality (and related CSI-ReportConfig) of such a specific terminal may be reported from the terminal to the base station, or the base station may perform activation / inactivation settings / instructions to the terminal regarding the specific AI / ML model / functionality (and related CSI-ReportConfig) of the terminal. However, in such cases, the base station may need to determine whether the UE-side AI / ML model / functionality of each terminal is in an active / inactive state for each terminal, and add up for each terminal to consider different second Z / Z' values for each terminal's specific AI / ML model / functionality (and related CSI-ReportConfig) to set / instruct the timing of CSI reporting related to the inference result report of each terminal, which increases the complexity of base station implementation.
[0436] Based on this problem, the base station / terminal may agree / define / consider the periodic or semi-persistent CSI reporting settings related to the inference result reporting of a specific terminal to always be in an active state (e.g., a state in which the relevant UE-side AI / ML model / functionality is being loaded into the actuator or / and AI engine), and thus a second Z / Z' may be applied without considering the additional delay that occurs when the terminal's specific AI / ML model / functionality is in an inactive state.
[0437] More specifically, in the case of periodic CSI reporting above, it can always be considered to be in an active state from the time of RRC configuration / re-configuration until the time of release, and in the case of semi-persistent CSI reporting, it can always be considered to be in an active state from the time of activation after receiving the activation MAC CE (e.g., from 3 ms after the terminal transmits A / N after receiving the activation MAC CE until the time of deactivation). On the other hand, regarding the aperiodic CSI reporting settings related to the inference result reporting of a specific terminal, the base station / terminal may always be agreed / defined / considered to be in an inactive state (e.g., a state in which the relevant UE-side AI / ML model / functionality is not loaded into the actuator or / and AI engine), and a second Z / Z' that takes into account the additional delay that occurs when the terminal's specific AI / ML model / functionality is in an inactive state may be applied.
[0438] As another example of Method 3, if the processing unit occupied by a specific AI / ML model / functionality (corresponding to CSI-ReportConfig) of the terminal does not include an AI / ML processing unit but only a CSI processing unit, there is no need to consider whether the AI / ML model / functionality is in an active / inactive state, so a second Z / Z' that does not consider the additional delay that occurs when the AI / ML model / functionality is in an inactive state can always be applied.
[0439] For example, information related to the above proposal 1 may be reported to a base station by a terminal.
[0440] As a specific example, the terminal may transmit capability information containing information related to the above proposal 1 to the base station.
[0441] As a specific example, the terminal may transmit an applicability report containing information related to the above proposal 1 to the base station. In other words, the information related to the above proposal 1 may be transmitted based on UAI reporting (signaling of viaOtherConfig or / and RRCReconfigurationComplete) for the applicability report of the UE-side AI / ML model / functionality.
[0442] As a specific example, the information related to the above proposal 1 may include at least one of the following: a second Z / Z' value of the terminal, new Z / Z', an offset value delta_Z / delta_Z' value from the first Z / Z' value for the second Z / Z', a second Z / Z' value according to the active / inactive state of the AI / ML model / functionality, or / and active / inactive state information of the terminal-specific AI / ML model / functionality.
[0443] Effect of Proposal 1) Methods 1 and 2 of Proposal 1 propose a method for defining the second Z / Z', and have the effect of enabling the terminal to derive / report inference results within the second Z / Z' by considering the processing time required for inference of the UE-side AI / ML model / functionality. In the case of Method 3 of Proposal 1, it has the effect of reducing the complexity of operations between the base station and the terminal by defining the complex problem, where the minimum processing time required for the terminal's inference-related CSI reporting may differ depending on whether the UE-side AI / ML model / functionality is in an active / inactive state, according to the time domain behavior of the CSI report.
[0444] Proposal 2
[0445] If the transmission timing of a CSI report related to UE-side AI / ML model inference set / instructed by the base station, as in Problem 2 above, satisfies the first Z / Z' (e.g., timing condition based on the first number of symbols satisfied) but does not satisfy the second Z / Z' (e.g., timing condition based on the second number of symbols not satisfied), the terminal may perform the CSI report based on at least one of the following embodiments.
[0446] Example 1)
[0447] For example, the terminal can ignore the scheduling DCI that triggered / activated the relevant CSI report (for aperiodic / semi-persistent CSI reports).
[0448] For example, (when more than 1 CSI report is triggered or multiplexed) the terminal may report a non-updated CSI. Specifically, the terminal may not update the n-th CSI report at the time of reporting and may perform the (n-1)-th inference result reporting as is.
[0449] Example 2)
[0450] The terminal can perform legacy CSI reporting instead of performing inference result reporting utilizing UE-side AI / ML models / functionality.
[0451] Specifically, (since the second Z / Z' for inference is not satisfied) the terminal does not perform an inference result report utilizing a UE-side AI / ML model / functionality for the relevant CSI report, and the terminal may perform a CSI report based on a CSI-RS measurement similar to a legacy CSI / beam report. More specifically, the terminal may perform the report based on at least one of the following methods.
[0452] Example 2-1)
[0453] (In the case of a CSI prediction report) At least one of the CQI / RI / PMI related to DL CSI may be reported based on one or more CSI-RS (resource set) measurement instance(s) prior to the reporting time of the CSI report (prior to the slot corresponding to / corresponding to the CSI reference resource of the CSI report). For example, the terminal may report at least one of the CQI / RI / PMI related to DL CSI based on the first and / or last CSI-RS (resource set) measurement instance(s) among the one or more CSI-RS (resource set) measurement instance(s). For example, the base station may set / instruct the terminal to perform a CSI report based on which instance(s) among the one or more CSI-RS (resource set) measurement instance(s). For example, the terminal may report at least one of the CQI / RI / PMI related to DL CSI by taking the average of the one or more CSI-RS (resource set) measurement instance(s).
[0454] Example 2-2)
[0455] (In the case of beam prediction BM-case1) K resource IDs or / and corresponding L1-RSRPs (set by the base station) may be reported based on one CSI-RS / SSB (resource set) measurement instance prior to the reporting time of the relevant CSI report (prior to the slot corresponding to / corresponding to the CSI reference resource of the relevant CSI report).
[0456] Example 2-3)
[0457] (In the case of beam prediction BM-case2) K resource IDs or / and corresponding L1-RSRPs (set by the base station) may be reported for each measurement instance based on one or more CSI-RS / SSB (resource set) measurement instances prior to the reporting time of the relevant CSI report (prior to the slot corresponding to / corresponding to the CSI reference resource of the relevant CSI report). For example, the terminal may report K resource IDs or / and corresponding L1-RSRPs based on the first or / and last CSI-RS / SSB (resource set) measurement instance among the one or more CSI-RS / SSB (resource set) measurement instances. For example, the base station may set / instruct which instance(s) among the terminal's one or more CSI-RS / SSB (resource set) measurement instances will be used to perform the CSI report. For example, the terminal may report K resource IDs or / and corresponding L1-RSRPs obtained by taking the average of the one or more CSI-RS / SSB (resource set) measurement instance(s).
[0458] In the above, CSI-RS / SSB (resource set) measurement instance(s) may refer to CSI-RS / SSB (resource set) transmission occasion(s) or CSI-RS / SSB (resource set) measurement occasion(s).
[0459] In order to inform the base station that the terminal performs an operation based on at least one embodiment of Proposal 2, an indicator may exist or be included in the terminal's CSI report regarding whether or not the report content of at least one embodiment is included. Or / and, since the base station can know that the terminal's CSI report time satisfies the first Z / Z' but not the second Z / Z', the base station may know in advance that the CSI report content of Proposal 2 will be reported from the terminal (e.g., the base station knows the reporting payload size of Proposal 2 and the reporting payload size when the situation of Proposal 2 is not present, and there may be no problem with decoding the CSI report).
[0460] Effect of Proposal 2) In the case of AI / ML-based CSI prediction or beam prediction BM-case2, if CSI measurements are performed at intervals of tens of milliseconds at least once, but the inference result report is dropped because the first Z / Z' condition is satisfied but the second Z / Z' condition is not, the past operation in which the terminal performed the measurement becomes meaningless. Considering this problem, Proposal 2 is intended to enable reporting on past CSI / beam measurements (even if AI / ML-related inference reports cannot be performed). This has the advantage that the base station can obtain the terminal's current or past CSI / beam information compared to the terminal not reporting anything in such a situation. Additionally, the base station has the advantage of being able to predict future CSI / beams (based on NW-side AI / ML inference) based on the terminal's CSI / beam reports based on multiple past instances.
[0461] Proposal 3
[0462] The terminal can perform active resource counting based on at least one of the following embodiments for beam prediction BM-case2 or / and CSI prediction reports that require multiple measurement instances in CSI reports related to UE-side AI / ML model inference set / instructed from the base station as in Problem 3 above.
[0463] Example 1)
[0464] The terminal may assume / operate to occupy a resource corresponding to a multiple of the number of measurement instances (N) for CSI reporting (related to inference result reporting) that requires one or more multiple measurement instances as described above. Here, N is a natural number. For example, the CSI-RS resource and the CSI-RS ports within the CSI-RS resource are counted N times.
[0465] In particular, for CSI reports related to BM-case2, since the number of RS resources performing measurements per measurement instance may be large, it can be assumed / operated that a scale factor X (where X is a natural number) is applied to the above N to occupy resources corresponding to multiples of N / X. For example, the above X value may be a value reported from the terminal via the UE capability / applicability report. For example, the above X value may be a value set / instructed to the terminal by the base station. For example, the above X value may be a predefined value.
[0466] Example 2)
[0467] The terminal K for CSI reports (related to inference result reporting) that require one or more multiple measurement instances as described above P It can be assumed / operated to occupy resources corresponding to multiples of ∈1,2,4,8,
[0016] ,...}. For example, CSI-RS resources and CSI-RS ports within those CSI-RS resources are K P The rounds are counted. Here, K p The value of ∈ {1,2,4,8,
[0016] ,...} is indicated by the UE capability (the CSI-RS resource and the CSI-RS ports within the CSI-RS resource are counted times, where the value of K p ∈ {1,2,4,8,
[0016] ,...} is indicated by UE capability).
[0468] For example, the above K P The value may be a value reported from the terminal via the UE capability / applicability report. For example, the above K P The value may be a value set / instructed to the terminal by the base station. As an example, the above K P The value can be a predefined value.
[0469] In particular, in the case of CSI reports related to BM-case2, since the number of RS resources performing measurements per measurement instance may be large, the above K P From, the scale factor X (X is a natural number) is applied to K PIt can be assumed / operated by occupying a resource corresponding to a multiple of / X. For example, the value of X may be a value reported from the terminal via a UE capability / applicability report. For example, the value of X may be a value set / instructed to the terminal by the base station. For example, the value of X may be a predefined value.
[0470] Effect of Proposal 3) In the case of AI / ML-related CSI reporting where a terminal performs multiple measurement instances and reports the inference result after inference, the terminal may need to buffer a larger number of measurement results compared to a single measurement instance. According to Proposal 3, by counting a larger number of active resources, there is an effect of preventing CSI measurements for CSI reporting that exceed the terminal's max buffer / memory size from being performed.
[0471] The above embodiments may be operated by a combination of one or more embodiments.
[0472] An example of a terminal (or base station) operation based on at least one of the aforementioned embodiments (e.g., at least one of the embodiments of Proposal 1 to Proposal 3) is as follows.
[0473] 1) The terminal (base station) receives (transmits) settings related to CSI / beam measurement / reporting.
[0474] The above settings may include reporting settings related to Set A and Set B for beam prediction or / and reporting settings related to CSI-RS for CSI prediction.
[0475] The reporting settings may include inference result reporting settings related to BM-case1 / 2 or / and CSI prediction.
[0476] 2) The terminal (base station) receives (transmits) a message scheduling the transmission of the CSI / beam measurement report.
[0477] The transmission of reports scheduled by the base station mentioned above can have periodic / semi-persistent / aperiodic time domain behavior.
[0478] The above report may be an inference result report utilizing UE-sided AI / ML.
[0479] 3) The terminal (base station) transmits (receives) a beam measurement report based on the above message.
[0480] The above report may be performed based on the embodiments of proposals 1 to 3.
[0481] The above terminal / base station operation is merely an example, and each operation (or step) is not necessarily essential; depending on the terminal / base station implementation method, the beam measurement / reporting operation of the terminal according to the aforementioned embodiments may be omitted or added.
[0482] In terms of implementation, operations of a base station / terminal according to the embodiments described above (e.g., operations based on at least one of proposal 1, proposal 2 and / or proposal 3) can be processed by the device of FIG. 10 (e.g., the processor (110, 210) of FIG. 10).
[0483] In addition, the operations of the base station / terminal according to the above-described embodiment (e.g., operations based on at least one of proposal 1, proposal 2 and / or proposal 3) may be stored in memory (e.g., 140, 240 of FIG. 10) in the form of instructions / programs (e.g., instruction, executable code) for driving at least one processor (e.g., 110, 210 of FIG. 10).
[0484] The embodiments described above will be explained in detail below with reference to FIGS. 8 and FIG. 9 regarding the operation of the terminal and base station. The methods described below are distinguished only for convenience of explanation, and it is understood that a part of one method may be substituted with a part of another method or combined with one another and applied.
[0485] FIG. 8 is a flowchart illustrating a method according to one embodiment of the present specification.
[0486] Referring to FIG. 8, a method according to one embodiment of the present specification includes a step of receiving a report setting related to CSI (S810) and a step of transmitting a CSI report related to inference of an AI / ML model (S820).
[0487] In S810, the terminal receives reporting settings related to Channel State Information (CSI) from the base station.
[0488] For example, the above report setting may refer to one of the report setting(s) included in the configuration information related to CSI.
[0489] In other words, S810 can be rephrased as follows. The terminal receives configuration information related to Channel State Information (CSI) from the base station.
[0490] For example, the above configuration information may include information based on at least one of the above-described CSI-related operations and proposals 1 to 2. As a specific example, the above configuration information may include i) one or more reporting settings (e.g., N≥1 CSI-ReportConfig reporting setting) and / or ii) one or more resource settings (e.g., M≥1 CSI-ResourceConfig resource setting).
[0491] Each of the above 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.
[0492] For example, the above setting information may include at least one of i) a first reporting setting related to prediction and / or ii) a second reporting setting related to prediction accuracy. Specifically, the one or more reporting settings may include at least one of i) a first reporting setting related to prediction and / or ii) a second reporting setting related to prediction accuracy.
[0493] As a specific example, the above-mentioned configuration information may include a reporting configuration related to a prediction (e.g., the first reporting configuration). The above-mentioned reporting configuration (e.g., the first reporting configuration) may include at least one of i) information related to a first resource configuration for measurement (e.g., an ID representing the first resource configuration, more specifically, a higher-level parameter resourcesForChannelMeasurement representing CSI-ResourceConfigId), ii) information related to a second resource configuration for prediction (e.g., an ID representing the second resource configuration (CSI-ResourceConfigId), more specifically, a higher-level parameter resourcesForChannelPrediction-r19 representing CSI-ResourceConfigId), and / or iii) information regarding the number of one or more time instances (e.g., a higher-level parameter nrofTimeInstance-r19).
[0494] More specifically, the above reporting setting (e.g., the above first reporting setting) may be related to BM-case1. Specifically, the above reporting setting (e.g., the above first reporting setting) may include i) information related to a first resource setting for measurement and ii) information related to a second resource setting for prediction.
[0495] More specifically, the above reporting setting (e.g., the above first reporting setting) may be related to BM-case2. Specifically, the above reporting setting (e.g., the above first reporting setting) may include i) information related to a first resource setting for measurement, ii) information related to a second resource setting for prediction, and iii) information regarding the number of one or more time instances.
[0496] For example, the report quantity of the first report setting above may be set to p-cri, p-cri-RSRP, p-ssb-index, or p-ssb-index-RSRP. p-cri represents the predicted CSI-RS Resource Indicator (P-CRI). p-ssb-index represents the predicted SSB Resource Indicator (P-SSBRI). In p-cri-RSRP or p-ssb-index-RSRP, RSRP represents the predicted Layer1-Reference Signal Received Power (P-L1-RSRP).
[0497] For example, the above prediction may be performed based on a measurement. The measurement may be performed based on a first resource configuration (e.g., CSI-ResourceConfig) associated with the first reporting configuration. For example, the first reporting configuration may be linked to a first resource configuration associated with the measurement and a second resource configuration associated with the prediction. The first reporting configuration may include the ID of the first resource configuration and the ID of the second resource configuration. Each ID may be based on the CSI-ResourceConfigId.
[0498] For example, the above measurements may include Layer1-Reference Signal Received Power (L1-RSRP) measurements. Based on the L1-RSRP measurements, i) at least one predicted CSI-RS Resource Indicator (P-CRI), ii) at least one predicted SSB Resource Indicator (P-SSBRI), and / or iii) at least one predicted Layer1-Reference Signal Received Power (P-L1-RSRP) may be determined.
[0499] More specifically, predictions for CSI-RS resources or SSB resources associated with the second resource configuration may be performed based on the L1-RSRP measurements. Specifically, predicted L1-RSRPs of CSI-RS resources or SSB resources associated with the second resource configuration may be determined. For example, best CRI(s) or best SSBRI(s) may be determined based on the order or ranking of the predicted L1-RSRPs. For example, at least one P-CRI or at least one P-SSBRI included in the CSI report described below may be based on the best CRI(s) or best SSBRI(s).
[0500] For example, the report quantity of the second report setting above can be set to pai (or rs-pai).
[0501] The above one or more resource settings may include i) a first resource setting and a second resource setting related to the first reporting setting and ii) a third resource setting related to the second reporting setting.
[0502] Each resource configuration (e.g., CSI-ResourceConfig) may include information about a resource set (e.g., csi-SSB-ResourceSetList or nzp-CSI-RS-ResourceSetList based on csi-RS-ResourceSetList). For example, the resources within the resource set may be SSB resources or CSI-RS resources based on csi-RS-ResourceSetList. csi-SSB-ResourceSetList may include information for referencing the SSB resources (e.g., CSI-SSB-ResourceSetIds). nzp-CSI-RS-ResourceSetList may include information for referencing the CSI-RS resources (e.g., NZP-CSI-RS-ResourceSetIds).
[0503] For example, the first resource setting may include information about a first resource set for measurement (e.g., Set B described above). As a specific example, the first resource setting may include a list of SSB resources or CSI-RS resources for measurement.
[0504] For example, the second resource setting may include information regarding a second resource set for the prediction (e.g., Set A described above). As a specific example, the second resource setting may include a list of SSB resources or CSI-RS resources for the prediction.
[0505] For example, the third resource setting may include information regarding a third resource set(s) for measurement (e.g., the resource set(s) described above for monitoring). As a specific example, the third resource setting may include a list of SSB resources or CSI-RS resources for measurement.
[0506] In S820, the terminal transmits a CSI report related to the inference of an AI / ML (Artificial Intelligence / Machine Learning) model to the base station. For example, the CSI report may be related to BM-case1 or BM-case2.
[0507] As a specific example of BM-case1, the terminal can transmit a CSI report containing predicted information based on the second resource setting to the base station.
[0508] As a specific example of BM-case2, the terminal may transmit to the base station a CSI report containing predicted information based on the second resource setting for each of the one or more time instances.
[0509] For example, the predicted information may include predicted CSI parameter(s) per time instance (e.g., P-CRI(s) / P-SSBRI(s) / P-L1-RSRP(s) per time instance). Specifically, the predicted CSI parameter(s) may include predicted CSI parameter(s) based on the report quantity of the first reporting setting (e.g., P-CRI(s), P-SSBRI(s), and / or P-L1-RSRP(s)). To explain in more detail, the predicted information may include at least one of i) at least one predicted CSI-RS Resource Indicator (P-CRI), ii) at least one predicted SSB Resource Indicator (P-SSBRI), and / or iii) at least one predicted Layer1-Reference Signal Received Power (P-L1-RSRP).
[0510] According to one embodiment, the CSI report may be transmitted as a valid CSI report based on the satisfaction of a timing condition based on a second number of symbols defined separately from a first number of symbols. The first number of symbols (e.g., a first Z / Z') may be related to a delay requirement for non-AI / ML-based CSI. The second number of symbols (e.g., a second Z / Z') may be related to a delay requirement for AI / ML-based CSI. This embodiment may be based on Proposal 1.
[0511] According to one embodiment, based on the fact that a timing condition related to the first number of symbols is satisfied and a timing condition related to the second number of symbols is not satisfied: the CSI report containing defined information may be transmitted. This embodiment may be based on Embodiment 2) of Proposal 2. It will be described in detail below.
[0512] According to one embodiment, the defined information may include at least one CSI parameter determined based on a measurement prior to the reporting time associated with the CSI report. This embodiment may be based on at least one of embodiments 2-1) to 2-3) of Proposal 2.
[0513] According to one embodiment, the at least one CSI parameter may include at least one of i) a Channel Quality Indicator (CQI), ii) a Rank Indicator (RI), and / or iii) a Precoding Matrix Indicator (PMI). This embodiment may be based on Example 2-1) of Proposal 2.
[0514] According to one embodiment, the at least one CSI parameter may include at least one of i) a resource indicator and / or ii) L1 (Layer 1)-RSRP (Reference Signal Received Power). This embodiment may be based on Example 2-2) of Proposal 2.
[0515] According to one embodiment, the at least one CSI parameter may include at least one of i) a resource indicator (e.g., a resource indicator representing a resource ID) based on each of the one or more transmission opportunities for the measurement and / or ii) an L1 (Layer 1)-RSRP (Reference Signal Received Power) based on each of the one or more transmission opportunities. This embodiment may be based on Example 2-3 of Proposal 2. For example, the one or more transmission opportunities for the measurement may be based on one or more CSI-RS / SSB (resource set) measurement instance(s). In other words, the one or more transmission opportunities may mean one or more CSI-RS transmission opportunities (or one or more SSB transmission opportunities).
[0516] According to one embodiment, the second number of symbols (e.g., the second Z / Z') may be defined as the sum of i) the first number of symbols (e.g., the first Z / Z') and ii) the offset for the first number of symbols. This embodiment may be based on method 1) of Proposal 1.
[0517] According to one embodiment, the timing condition may include i) a first condition and ii) a second condition. The number of first symbols may include i) a first Z associated with the first condition and ii) a first Z' associated with the second condition. The number of second symbols may include i) a second Z associated with the first condition and ii) a second Z' associated with the second condition. The second Z may be defined based on the value obtained by subtracting the first Z' from the second Z' or based on the value obtained by adding the second Z' to the first Z. This embodiment may be based on method 2) of Proposal 1.
[0518] According to one embodiment, the number of second symbols may be defined differently for each of the active state and inactive state of the AI / ML model or AI / ML functionality associated with the reporting setting. This embodiment may be based on method 3) of Proposal 1.
[0519] According to one embodiment, regarding the counting of activated CSI-RS resources and activated CSI-RS ports within a time duration associated with the CSI report, the CSI-RS resources and the CSI-RS ports within the CSI-RS resources may be counted a defined number of times. This embodiment may be based on Proposal 3.
[0520] For example, the number defined above may be determined based on i) a first value and / or ii) a scale factor.
[0521] For example, the first value may be i) the number of measurement instances associated with the CSI report, ii) a value reported by the terminal, or iii) a value set on the terminal by the base station. In the example described above, i) may mean N (a natural number) in Example 1) of Proposal 3. In the example described above, ii) and iii) are K in Example 2) of Proposal 3. p It can mean.
[0522] Whether the timing condition related to the number of the first and second symbols described above is satisfied can be determined as follows.
[0523] The timing condition related to the first symbol count mentioned above is satisfied in that the first uplink symbol carrying the CSI report is a first reference uplink symbol (e.g., Z) determined based on the first symbol count mentioned above. ref,1 , Z' ref,1 This may mean that it does not start earlier than ). In other words, the timing condition related to the number of the first symbol is not satisfied may mean that the first uplink symbol starts earlier than the first reference uplink symbol.
[0524] The fact that the timing condition related to the number of the second symbol is satisfied means that the first uplink symbol is a second reference uplink symbol (e.g., Z) determined based on the number of the second symbol. ref,2 , Z' ref,2 This may mean that it does not start earlier than ). In other words, the timing condition related to the number of the second symbol is not satisfied may mean that the first uplink symbol starts earlier than the second reference uplink symbol.
[0525] Accordingly, the expression "based on the timing condition related to the first number of symbols being satisfied and the timing condition related to the second number of symbols not being satisfied:" may be replaced with "based on the first uplink symbol carrying the CSI report not starting earlier than the first reference uplink symbol determined based on the first number of symbols, and the first uplink symbol starting earlier than the second reference uplink symbol determined based on the second number of symbols:".
[0526] Operations based on S810 to S820 described above can be implemented by the device of FIG. 10. For example, referring to FIG. 10, the terminal (200) can control one or more transceivers (230) and / or one or more memories (240) to perform operations based on S810 to S820.
[0527] The embodiments described above will be explained in detail below in terms of base station operation.
[0528] S910 to S920 described below correspond to operations based on S810 to S820 described in FIG. 8. 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. 8 corresponding to the operation.
[0529] FIG. 9 is a flowchart illustrating a method according to another embodiment of the present specification.
[0530] Referring to FIG. 9, a method according to another embodiment of the present specification includes a step of transmitting a report setting related to CSI (S910) and a step of receiving a CSI report related to inference of an AI / ML model (S920).
[0531] In S910, the base station transmits report settings related to Channel State Information (CSI) to the terminal.
[0532] In S920, the base station receives a CSI report related to the inference of an AI / ML (Artificial Intelligence / Machine Learning) model from the terminal.
[0533] According to one embodiment, the CSI report may be received as a valid CSI report based on the satisfaction of a timing condition based on a second number of symbols defined separately from a first number of symbols. The first number of symbols (e.g., a first Z / Z') may be related to a delay requirement for non-AI / ML-based CSI. The second number of symbols (e.g., a second Z / Z') may be related to a delay requirement for AI / ML-based CSI. This embodiment may be based on Proposal 1.
[0534] According to one embodiment, based on the fact that a timing condition related to the first number of symbols is satisfied and the timing condition related to the second number of symbols is not satisfied: the CSI report containing defined information may be received. This embodiment may be based on embodiment 2) of Proposal 2.
[0535] 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. 10.
[0536] FIG. 10 is a drawing showing the configuration of a first device and a second device according to an embodiment of the present specification.
[0537] The first device (100) may include a processor (110), an antenna unit (120), a transceiver (130), and a memory (140).
[0538] 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).
[0539] 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.
[0540] 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.
[0541] The second device (200) may include a processor (210), an antenna unit (220), a transceiver (230), and a memory (240).
[0542] 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).
[0543] 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.
[0544] 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.
[0545] 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.
[0546] 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.
[0547] 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.
[0548] 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 a report setting related to Channel State Information (CSI) from a base station by a terminal; and The method includes the step of transmitting a CSI report related to the inference of an AI / ML (Artificial Intelligence / Machine Learning) model to the base station by the terminal, Based on the satisfaction of a timing condition based on a second number of symbols defined separately from the first number of symbols, the above CSI report is transmitted as a valid CSI report, and The first number of symbols mentioned above is related to a delay requirement for non-AI / ML-based CSI, and the second number of symbols mentioned above is related to a delay requirement for AI / ML-based CSI, and Based on the fact that the timing condition related to the first number of symbols is satisfied, and the timing condition related to the second number of symbols is not satisfied: A method characterized by transmitting the above CSI report containing defined information.
2. In Paragraph 1, A method characterized by the above-defined information including at least one CSI parameter determined based on measurements prior to the reporting time associated with the above CSI report.
3. In Paragraph 2, A method characterized in that the above-mentioned at least one CSI parameter includes at least one of i) a Channel Quality Indicator (CQI), ii) a Rank Indicator (RI), and / or iii) a Precoding Matrix Indicator (PMI).
4. In Paragraph 2, A method characterized in that the above at least one CSI parameter includes at least one of i) a resource indicator and / or ii) L1 (Layer 1)-RSRP (Reference Signal Received Power).
5. In Paragraph 2, A method characterized in that the above at least one CSI parameter comprises at least one of i) a resource indicator based on each of one or more transmission opportunities for the measurement and / or ii) an L1 (Layer 1)-RSRP (Reference Signal Received Power) based on each of the one or more transmission opportunities.
6. In Paragraph 1, A method characterized in that the number of the second symbols is defined as the sum of i) the number of the first symbols and ii) the offset for the number of the first symbols.
7. In Paragraph 1, The above timing conditions include i) a first condition and ii) a second condition, and The number of the first symbols above includes i) a first Z related to the first condition and ii) a first Z' related to the second condition, and The number of the second symbols above includes i) a second Z related to the first condition and ii) a second Z' related to the second condition, and A method characterized in that the second Z is defined based on the value obtained by subtracting the first Z' from the second Z' or by adding the second Z' to the first Z.
8. In Paragraph 1, A method characterized in that the number of the second symbols is defined differently for each of the active state and inactive state of the AI / ML model or AI / ML functionality associated with the reporting settings.
9. In Paragraph 1, With respect to the counting of active CSI-RS resources and active CSI-RS ports within the time duration associated with the above CSI report, the CSI-RS resources and the CSI-RS ports within the said CSI-RS resources are counted a defined number of times, and A method characterized in that the number defined above is determined based on i) a first value and / or ii) a scale factor.
10. In Paragraph 9, A method characterized in that the first value is i) the number of measurement instances associated with the CSI report, ii) a value reported by the terminal, or iii) a value set on the terminal by the base station.
11. In 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 enabling the terminal to perform all steps of the method according to any one of claims 1 to 10, based on execution by the one or more processors.
12. 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 10, based on execution by the above one or more processors.
13. 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 to 10.
14. Regarding the method, A step of transmitting reporting settings related to Channel State Information (CSI) to a terminal by a base station; and The method includes the step of receiving a CSI report related to the inference of an AI / ML (Artificial Intelligence / Machine Learning) model from the terminal via the base station, Based on the satisfaction of a timing condition based on a second number of symbols defined separately from the first number of symbols, the above CSI report is received as a valid CSI report, and The first number of symbols mentioned above is related to a delay requirement for non-AI / ML-based CSI, and the second number of symbols mentioned above is related to a delay requirement for AI / ML-based CSI, and Based on the fact that the timing condition related to the first number of symbols is satisfied, and the timing condition related to the second number of symbols is not satisfied: A method characterized by receiving the above CSI report containing defined information.
15. 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 14.
