Method and device for communication in wireless communication system

Abstract

The present disclosure relates to a 5G or 6G communication system for providing higher data transmission rates than 4G communication systems such as LTE. A method performed by a terminal in a wireless communication system may comprise the steps in which the terminal: receives, from a base station, information indicating whether the terminal is to perform AI-based channel state information (CSI) prediction; determines whether to perform the AI-based CSI prediction on the basis of the information; when performing the AI-based CSI prediction, acquires channel state information on the basis of at least one reference signal and a first AI model; and transmits the channel state information.

Description

Method and device for performing communication in a wireless communication system

[0001] The present disclosure relates to a wireless communication system, and more particularly, to a method and device for performing communication based on AI (Artificial Intelligence).

[0002] Looking back at the evolution of wireless communication over successive generations, technologies have primarily been developed for human-facing services such as voice, multimedia, and data. With the commercialization of 5G (5th-generation) communication systems, an explosive increase in connected devices is expected to be connected to communication networks. Examples of networked objects include vehicles, robots, drones, home appliances, displays, smart sensors installed in various infrastructures, construction equipment, and factory equipment. Mobile devices are also expected to evolve into diverse form factors, such as augmented reality glasses, virtual reality headsets, and holographic devices. In the 6th-generation (6G) era, efforts are being made to develop improved 6G communication systems to connect hundreds of billions of devices and objects and provide diverse services. For this reason, 6G communication systems are often referred to as "beyond 5G."

[0003] The 6G communication system, expected to be realized around 2030, will have a maximum transmission speed of terabytes per second (1,000 gigabits per second) and a wireless latency of 100 microseconds (μsec). This means that compared to 5G, the transmission speed in a 6G communication system will be 50 times faster, while the wireless latency will be reduced to one-tenth.

[0004] To achieve these high data rates and ultra-low latency, 6G communication systems are being considered for implementation in the terahertz band (e.g., from 95 gigahertz (GHz) to 3 terahertz (THz)). Compared to the millimeter wave (mmWave) band introduced in 5G, the terahertz band is expected to experience more severe path loss and atmospheric absorption, making it more crucial to ensure signal reach, or coverage, in this band. Key technologies to ensure coverage include radio frequency (RF) components, antennas, new waveforms that offer better coverage than orthogonal frequency division multiplexing (OFDM), beamforming, and multiple antenna transmission technologies such as massive multiple-input and multiple-output (MIMO), full-dimensional MIMO (FD-MIMO), array antennas, and large-scale antennas. In addition, new technologies such as metamaterial-based lenses and antennas, high-dimensional spatial multiplexing using orbital angular momentum (OAM), and reconfigurable intelligent surfaces (RIS) are being discussed to improve the coverage of terahertz band signals.

[0005] In addition, in order to improve frequency efficiency and system network, 6G communication systems are developing full duplex technology that utilizes the same frequency resources for uplink and downlink at the same time; network technology that integrates satellites and high-altitude platform stations (HAPS); network structure innovation technology that supports mobile base stations and enables optimization and automation of network operation; dynamic spectrum sharing technology through collision avoidance based on spectrum usage prediction; AI-based communication technology that utilizes artificial intelligence (AI) from the design stage and internalizes end-to-end AI support functions to realize system optimization; and next-generation distributed computing technology that realizes services with complexity that exceeds the limits of terminal computing capabilities by utilizing ultra-high-performance communication and computing resources (mobile edge computing (MEC), cloud, etc.). In addition, efforts are being made to further strengthen connectivity between devices, further optimize networks, promote softwareization of network entities, and increase the openness of wireless communications through the design of new protocols to be used in 6G communication systems, the implementation of hardware-based security environments, the development of mechanisms for the safe use of data, and the development of technologies for maintaining privacy.

[0006] Research and development of these 6G communication systems are expected to enable a new level of hyper-connected experience (the next hyper-connected experience) through the hyper-connectivity of 6G communication systems, which encompass not only connections between things but also connections between people and things. Specifically, 6G communication systems are expected to enable services such as truly immersive extended reality (Truly Immersive XR), high-fidelity mobile holograms, and digital replicas. Furthermore, services such as remote surgery, industrial automation, and emergency response, which are provided through 6G communication systems through enhanced security and reliability, will find application in diverse fields such as industry, medicine, automobiles, and home appliances.

[0007] According to an embodiment of the present disclosure, a CSI reporting method and device can be provided using AI that compensates for a situation in which a channel changes between when a CSI-RS is transmitted and when the CSI is reported and actually used.

[0008] The technical problem to be solved by one embodiment of the present disclosure is not limited to the technical problem described above, and other technical problems can be inferred from the following examples.

[0009] According to an embodiment of the present disclosure, a method performed by a terminal in a wireless communication system is provided. The method may include the steps of: receiving information from a base station indicating whether the terminal will perform AI-based channel state information (CSI) prediction; determining whether to perform AI-based CSI prediction based on the information; obtaining channel state information based on at least one reference signal and a first AI model when performing AI-based CSI prediction; and transmitting the channel state information.

[0010] According to an embodiment of the present disclosure, a method performed by a base station in a wireless communication system is provided. The method may include the steps of transmitting information to a terminal indicating whether the terminal will perform AI-based CSI prediction; and receiving channel state information from the terminal. If the terminal performs AI-based CSI prediction, the channel state information may be obtained based on at least one reference signal and a first AI model.

[0011] According to an embodiment of the present disclosure, a terminal of a wireless communication system is provided. The terminal may include a transceiver; and at least one processor connected to the transceiver. The at least one processor may receive information from a base station indicating whether the terminal will perform AI-based channel state information (CSI) prediction; determine whether to perform AI-based CSI prediction based on the information; and, when performing AI-based CSI prediction, obtain channel state information based on at least one reference signal and a first AI model; and transmit the channel state information to the base station.

[0012] According to an embodiment of the present disclosure, a base station of a wireless communication system is provided. The base station may include a transceiver; and at least one processor connected to the transceiver. The at least one processor may transmit information to a terminal indicating whether the terminal will perform AI-based CSI prediction; and receive channel state information from the terminal. When the terminal performs AI-based CSI prediction, the channel state information may be acquired based on at least one reference signal and a first AI model.

[0013] Figure 1 is a diagram for explaining the AI / ML model.

[0014] FIG. 2 is a diagram for explaining reporting CSI according to one embodiment of the present disclosure.

[0015] FIG. 3 is a diagram for explaining AI-based CSI prediction according to one embodiment of the present disclosure.

[0016] FIG. 4 and FIG. 5 are diagrams for explaining an AI model according to one embodiment of the present disclosure.

[0017] FIG. 6 is a diagram for explaining a terminal performing AI-based CSI prediction according to one embodiment of the present disclosure.

[0018] FIG. 7 is a diagram illustrating a base station performing AI-based CSI prediction according to one embodiment of the present disclosure.

[0019] FIG. 8 is a flowchart of a method for performing CSI reporting according to one embodiment of the present disclosure.

[0020] FIG. 9 is a flowchart illustrating a process for transmitting status information regarding AI-based CSI prediction according to one embodiment of the present disclosure.

[0021] FIG. 10 is a flowchart of a method for performing CSI reporting according to one embodiment of the present disclosure.

[0022] FIG. 11 is a flowchart of a method for performing CSI reporting according to one embodiment of the present disclosure.

[0023] FIG. 12 is a block diagram schematically illustrating the configuration of a base station (1200) according to one embodiment of the present disclosure.

[0024] FIG. 13 is a block diagram schematically illustrating the configuration of a terminal (1300) according to one embodiment of the present disclosure.

[0025] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the attached drawings.

[0026] In describing the present disclosure, descriptions of technical details that are well known in the technical field to which the present disclosure pertains and are not directly related to the present disclosure will be omitted. This is to avoid obscuring the gist of the present disclosure by omitting unnecessary explanations and to convey the gist more clearly.

[0027] The terms described below are defined based on the functions of the present disclosure, and may vary depending on the intent or custom of the user or operator. Therefore, their definitions should be based on the contents throughout this specification.

[0028] Other technical features will become readily apparent to those skilled in the art from the following drawings, descriptions and claims.

[0029] Before embarking on the “Description of the Detailed Description of the Invention” below, it may be advantageous to note definitions of certain words and phrases used throughout this patent document.

[0030] The term "couple" and its derivatives refer to any direct or indirect communication between two or more elements, whether or not they are in physical contact with each other. The terms "transmit," "receive," and "communicate," as well as their derivatives, encompass both direct and indirect communication. The terms "include," "comprising," and "including," as well as their derivatives, encompass without limitation.

[0031] The term "or" is inclusive and means "and / or." The phrase "associated with" as well as its derivatives means includes, is included within, interconnects with, contains, is contained within, connects to or with, couples to or with, is communicable with, cooperates with, interleaves with, juxtaposes with, is proximate to, is associated with or is connected to, has, has the characteristics of, is related to or is related to, etc.

[0032] Additionally, in the present disclosure, when a component is referred to as being “connected” or “connected” to another component, it should be understood that the component may be directly connected or connected to the other component, but may also be connected or connected via another component in between, unless there is a specific description to the contrary.

[0033] The present disclosure is susceptible to various modifications and embodiments. Specific embodiments are illustrated in the drawings and described in detail herein. However, the various embodiments discussed below, as well as those used to illustrate the principles of the present disclosure in this specification, are merely illustrative and should not be construed as limiting the scope of the present disclosure in any way.

[0034] Those skilled in the art will appreciate that the principles of the present disclosure can be implemented in any appropriately arranged system or device. Furthermore, those skilled in the art will appreciate that the principles of the present disclosure can be implemented in any appropriately configured wireless communication system.

[0035] For the same reason, some components in the attached drawings are exaggerated, omitted, or schematically depicted. Furthermore, the dimensions of each component do not entirely reflect its actual size. Furthermore, the numbers used throughout the description (e.g., "First," "Second," etc.) are merely identifiers used to distinguish one component from another.

[0036] The advantages and features of the present disclosure, and methods for achieving them, will become clearer with reference to the embodiments described in detail below together with the accompanying drawings. However, the present disclosure is not limited to the embodiments disclosed below and may be implemented in various different forms. These embodiments are provided only to ensure that the disclosure of the present disclosure is complete and to fully inform those skilled in the art of the scope of the disclosure, and the present disclosure is defined only by the scope of the claims. Like reference numerals throughout the specification denote like elements or corresponding elements.

[0037] Hereinafter, a base station (BS) is an entity that performs resource allocation of a terminal, and may be at least one of a gNode B, an eNode B, a Node B (or xNode B (where x is an alphabet including g and e)), a wireless access unit, a base station controller, a satellite, an airborne, or a node on a network. In the present disclosure, a base station may be used interchangeably with a network.

[0038] A user equipment (UE) may include a mobile station (MS), a vehicle, a satellite, an airborne vehicle, a cellular phone, a smartphone, a computer, or a multimedia system capable of performing communication functions.

[0039] Additionally, in the present disclosure, a cell may represent an area covered by a single base station in wireless communications. Cells may be classified into mega cells, macro cells, micro cells, and pico cells based on their size, but these are merely examples, and the types of cells are not limited to the above.

[0040] In the present disclosure, downlink (DL) refers to a wireless transmission path for a signal transmitted from a base station to a terminal, and uplink (UL) refers to a wireless transmission path for a signal transmitted from a terminal to a base station. Additionally, a sidelink (SL) may exist, which refers to a wireless transmission path for a signal transmitted from a terminal to another terminal.

[0041] In the following description, terms referring to broadcast information, terms referring to control information, terms related to communication coverage, terms referring to state changes (e.g., events), terms referring to network entities, terms referring to messages, terms referring to device components, etc. are provided as examples for convenience of explanation.

[0042] In addition, although LTE, LTE-A, or 5G systems may be described below as examples, embodiments of the present disclosure may also be applied to other communication systems having similar technical backgrounds or channel types. For example, this may include 5G-Advance or NR-Advance developed after 5G mobile communication technology (or new radio, NR), or 6th generation mobile communication technology (6G or beyond 5G), and 5G below may also be a concept that includes existing LTE, LTE-A, and other similar services. In addition, the present disclosure may be applied to other communication systems with some modifications within a range that does not significantly deviate from the scope of the present disclosure, as determined by a person having skilled technical knowledge.

[0043] For convenience of explanation, this disclosure uses terms and names defined in the LTE (long term evolution) and NR (new radio) standards, the most recent standards defined by the 3rd Generation Partnership Project (3GPP) among the existing communication standards. However, this disclosure is not limited to the aforementioned terms and names, and can be equally applied to wireless communication systems conforming to other standards.

[0044] For example, although the embodiments of the present disclosure are described by taking the 6th generation wireless communication technology (6G, system) as an example, an embodiment of the present disclosure may be applied to other wireless communication systems having a similar technical background or channel type. According to another example, the embodiments of the present disclosure may be applied to NR, LTE, or LTE-A, which are wireless communication systems prior to NR, and further, the embodiments of the present disclosure may be applied to wireless communication systems developed after NR. Furthermore, the embodiments of the present disclosure may be applied to other wireless communication systems through some modifications within a range that does not significantly deviate from the scope of the present disclosure at the discretion of a person having skilled technical knowledge.

[0045] Therefore, the present disclosure is not limited to the terms described below, and other terms having equivalent technical meanings may be used. For the convenience of the following description, the present disclosure may use CSI-RS among reference signals, but the present disclosure is not limited to CSI-RS, and may also be applied to reference signals described in the 3GPP standard, such as DMRS (Demodulation reference signal), PTRS (Phase-tracking reference), PRS (Positioning reference signal), TRS (tracking reference signal), RIM RS (remote interference management reference signal), etc.

[0046] For convenience of explanation below, the present disclosure may be explained using CQI (Channel Quality Indicator) and PMI (precoding matrix indicator) among channel state information (CSI), but this is only an example, and the present disclosure may also be applied to CRI (CSI-RS resource indicator), SSBRI (SS / PBCH Block Resource indicator), LI (layer indicator), RI (rank indicator), L1-RSRP (Layer 1 reference signal received power), L1-SINR (signal-to-noise and interference ratio), or Capability Index.

[0047] For convenience of explanation below, the present disclosure may be described by way of example in which AI-based CSI prediction is performed in the time domain; however, this is merely an example, and embodiments of the present disclosure may be applied to at least one of the time domain, the frequency domain, and the spatial domain.

[0048] At this time, it will be understood that each block of the processing flow diagrams and combinations of the flow diagrams can be performed by computer program instructions. These computer program instructions can be installed on a processor of a general-purpose computer, special-purpose computer, or other programmable data processing equipment, so that the instructions, when executed by the processor of the computer or other programmable data processing equipment, create a means for performing the functions described in the flow diagram block(s).

[0049] These computer program instructions may also be stored in a computer-available or computer-readable memory that can direct a computer or other programmable data processing device to implement a function in a particular manner, so that the instructions stored in the computer-available or computer-readable memory can also produce an article of manufacture that includes instruction means for performing the function described in the flowchart block(s).

[0050] Since the computer program instructions may be installed on a computer or other programmable data processing device, a series of operational steps may be performed on the computer or other programmable data processing device to create a computer-executable process, and the instructions that cause the computer or other programmable data processing device to perform the steps for performing the functions described in the flowchart block(s) may also provide steps for performing the functions described in the flowchart block(s).

[0051] Additionally, each block may represent a module, segment, or portion of code that contains one or more executable instructions for performing a specific logical function(s). It should also be noted that in some alternative implementations, the functions described in the blocks may occur out of order. For example, two blocks depicted in succession may actually be executed substantially concurrently, or the blocks may sometimes be executed in reverse order, depending on their respective functions.

[0052] Here, the term '~ unit' used in this embodiment refers to software or hardware components such as FPGA (Field Programmable Gate Array) or ASIC (Application Specific Integrated Circuit), and '~ unit' performs certain roles. However, '~ unit' is not limited to software or hardware. '~ unit' may be configured to be on an addressable storage medium or may be configured to play one or more processors.

[0053] As an example, '~bu' includes components such as software components, object-oriented software components, class components, and task components, as well as processes, functions, attributes, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables.

[0054] The functionality provided within the components and '~sub-units' may be combined into a smaller number of components and '~sub-units' or further separated into additional components and '~sub-units'. Furthermore, the components and '~sub-units' may be implemented to activate one or more CPUs within the device or secure multimedia card. Furthermore, in an embodiment, the '~sub-unit' may include one or more processors.

[0055] Figure 1 is a diagram for explaining the AI / ML model.

[0056] AI / ML models are models used in next-generation wireless communication systems, such as 6G, that utilize artificial intelligence (AI) and machine learning (ML) to achieve higher performance and efficiency.

[0057] Wireless communication systems can use AI / ML models to improve network performance. Using AI / ML models, wireless communication systems can analyze channel characteristics and optimize signals. Using AI / ML models, wireless communication systems can analyze user behavior and preferences to provide personalized services and enhance the user experience.

[0058] Using AI / ML models, wireless communication systems can analyze network traffic and efficiently allocate network resources. Using AI / ML models, wireless communication systems can detect and recover from network failures.

[0059] Using AI / ML models, wireless communication systems can optimize dynamic resource allocation across multiple frequency bands and bandwidths, thereby improving spectral efficiency. AI / ML models enable wireless communication systems to utilize more frequency bands and bandwidths, enabling faster and more stable communications.

[0060] The AI / ML model may include a one-sided AI / ML model (101, 103, 107), a two-sided AI / ML model (105), etc. The one-sided AI / ML model (101, 103, 107) may be operated only at the terminal, only at the base station, or both at the terminal and the base station. The two-sided AI / ML model (105) is a pair of AI models on which joint inference is performed, and joint inference can be performed at the UE and the base station.

[0061] AI / ML models may include Bi-LSTM (Bidirectional long short-term memory) models, LSTM (long short-term memory) models, FCN (fully convolutional network), MLP-Mixer (Mixed Multi-Layer Perceptron) models, ResNet (Residual Network), etc., but these are just examples and are not limited to the examples mentioned.

[0062] Wireless communication systems can perform AI-based CSI prediction using AI / ML models. AI-based CSI prediction can be performed in the time domain, frequency domain, or spatial domain. AI-based CSI prediction can utilize unidirectional AI / ML models and can be operated on both the terminal and the network.

[0063] FIG. 2 is a diagram for explaining reporting CSI according to one embodiment of the present disclosure.

[0064] Services being considered for post-5G communication systems include enhanced mobile broadband (eMBB), massive machine type communication (mMTC), and ultra-reliable low-latency communication (URLLC). To satisfy these diverse services, beam management and various frequency bands may be supported.

[0065] In these situations, various channel conditions can occur across frequency bands or beams. The terminal can estimate or measure the channel state and report CSI for the estimated channel to the base station.

[0066] For example, a terminal may transmit CSI, such as RI, PMI, and CQI, to a base station. The process of reporting CSI to the base station can consume significant resources. According to an example of the present disclosure, an AI-based channel prediction method and device can be provided that reduces the overhead of CSI-RS in MIMO systems and the like.

[0067] The base station can transmit a reference signal (RS) to the terminal. The terminal estimates the channel state (channel estimation) through the received RS, and You can get it.

[0068] The terminal is a channel can be preprocessed to obtain a precoding matrix, for example, an eigenvector V. The preprocessing may include Eigen Value Decomposition (EVD) or Singular Value Decomposition (SVD). For example, the terminal V can be generated based on the relationship.

[0069] channel If the V value is known, the V value can be obtained. In codebook-based CSI feedback, the terminal can identify the PMI based on the V value and transmit the PMI to the base station. For example, the terminal can transmit the PMI corresponding to the index of the codebook that is most similar to the V value.

[0070] A channel estimated from a reference signal, such as a CSI-RS, can operate ideally under the assumption that the channel remains unchanged during the time between the base station transmitting the CSI-RS and the terminal receiving the CSI-RS, the time between the terminal reporting the CSI to the base station, and the time the base station performs downlink scheduling. However, in an actual system, the channel may change over time due to factors such as the channel environment and the mobility of the terminal.

[0071] According to an embodiment of the present disclosure, a method and device for reporting AI-related CSI in the time domain can be provided that compensates for the situation in which a channel ages between when a CSI-RS is transmitted and when the CSI is reported and when it is actually used. The acquired eigenvector, ), channel (H), PMI, CQI, etc., can be utilized in an AI model to obtain CSI with scheduling delay, etc. compensated for.

[0072] FIG. 3 is a diagram for explaining AI-based CSI prediction according to one embodiment of the present disclosure.

[0073] When performing AI-based CSI prediction, the resolution of the results may vary depending on whether the node where the AI ​​model is operated is a terminal or a base station.

[0074] When an AI model is operated on a terminal, the terminal uses the eigenvector V and channel as inputs to the AI ​​model. It can be used to obtain high-resolution PMI, CQI, and other CSI as output.

[0075] When an AI model is operated at a base station, PMI quantized by a codebook or CQI quantized by SINR can be used as input to the AI ​​model, and at least one of PMI or CQI with a lower resolution than the terminal can be obtained as output.

[0076] AI-based CSI prediction can be performed using a unidirectional AI model. The unidirectional AI model can operate solely on the terminal, solely on the base station, or on both the terminal and the base station. When operated on both the terminal and the base station, similar to the bidirectional model, the terminal's AI model output can be used as input to the base station's AI model, or vice versa. Regardless of whether the terminal performs AI-based CSI prediction, the network may perform unintended AI-based CSI prediction. If the network expects the terminal to perform AI-based CSI prediction, the terminal may not perform AI-based CSI prediction.

[0077] Referring to Figure 3, a situation may arise where CSI is predicted for a slot other than the intended slot to compensate for scheduling delays, such as due to unexpected AI-based CSI predictions performed by the terminal or failure to perform expected AI-based CSI predictions. In this case, channel estimation failure may occur, resulting in problems such as reduced system throughput.

[0078] According to one embodiment of the present disclosure, in relation to performing AI-based CSI prediction, a network can control the operation of a terminal with respect to performing prediction of the terminal, and can reduce or prevent errors in channel estimation that may occur when performing AI-based CSI.

[0079] According to one embodiment of the present disclosure, in a wireless communication system, signaling between a base station and a terminal can be used to control the AI-based CSI prediction operation of the terminal. This can reduce channel estimation errors when performing dual prediction between the terminal and the base station, and prevent degradation of channel estimation performance due to the terminal's failure to perform CSI prediction.

[0080] According to one embodiment of the present disclosure, a terminal may perform AI-based CSI prediction for a time resource designated by a base station, and the network may perform AI-based CSI prediction for a period after the designated time resource. Scheduling delay may be compensated.

[0081] According to one embodiment of the present disclosure, a terminal can compensate for the time delay between a CSI-RS and a CSI report by performing AI-based CSI prediction. A network can compensate for the delay between a CSI report and scheduling by performing AI-based CSI prediction.

[0082] FIG. 4 and FIG. 5 are diagrams for explaining an AI model according to one embodiment of the present disclosure.

[0083] The artificial intelligence-related functions according to the present disclosure can be operated via a processor and memory. The processor may be comprised of one or more processors. In this case, one or more processors may be a general-purpose processor such as a CPU, AP, or DSP (Digital Signal Processor), a graphics-only processor such as a GPU or VPU (Vision Processing Unit), or an artificial intelligence-only processor such as an NPU.

[0084] One or more processors control the processing of input data according to predefined operating rules or AI models stored in memory. If one or more of these processors are dedicated AI processors, they may be designed with a hardware architecture specialized for processing a specific AI model.

[0085] A predefined behavioral rule or artificial intelligence model is characterized by being created through training. Here, "created through training" means that a basic artificial intelligence model (or deep learning model) is trained using a learning algorithm using a large amount of learning data, thereby creating a predefined behavioral rule or artificial intelligence model configured to perform a desired characteristic (or purpose).

[0086] This learning may be performed on the device itself, where the artificial intelligence according to the present disclosure is implemented, or through a separate server and / or system. Examples of learning algorithms include, but are not limited to, supervised learning, unsupervised learning, semi-supervised learning, or reinforcement learning.

[0087] An AI model can be composed of multiple neural network layers. Each of these neural network layers has multiple weight values, and neural network operations can be performed through calculations between the results of previous layers and the multiple weights.

[0088] The multiple weights of multiple neural network layers can be optimized based on the learning results of the AI ​​model. For example, during the learning process, the multiple weights can be updated to reduce or minimize the loss or cost values ​​obtained by the AI ​​model.

[0089] Device (1200, 1300) is a channel , V, PMI. AI-based CSI can be obtained based on at least one of CQI.

[0090] For example, referring to FIGS. 4 and 5, the device (1200, 1300) can apply at least one input data (401) to an AI model (450) to obtain output data (403) corresponding to a predetermined time resource. The AI ​​model (450) can be trained so that when at least one input data (401) is input, output data (403) corresponding to a predetermined time resource is output.

[0091] The input data (401) is , V, PMI, CQI, or at least one of other CSI. The output data (403) may correspond to PMI, CQI, or other CSI.

[0092] For example, a device may include at least one precoding matrix V( ) is applied to the AI ​​model (450) to obtain a precoding matrix corresponding to a given time resource ( ) can be obtained. The AI ​​model (450) can be trained to output a precoding matrix corresponding to a given time resource when at least one precoding matrix is ​​input.

[0093] The device can obtain at least one PMI corresponding to a given time resource by applying at least one V to the AI ​​model (450). The AI ​​model (450) can be trained to output at least one PMI corresponding to a given time resource when at least one V is input.

[0094] The device can obtain at least one CQI corresponding to a given time resource by applying at least one V to the AI ​​model (450). The AI ​​model (450) can be trained to output at least one CQI corresponding to a given time resource when at least one V is input.

[0095] The device has at least one channel By applying the AI ​​model (450), at least one precoding matrix corresponding to a given time resource can be obtained. The AI ​​model (450) can be configured to obtain at least one channel When input, it can be trained to output at least one precoding matrix corresponding to a given time resource.

[0096] The device has at least one channel By applying the AI ​​model (450), at least one PMI corresponding to a given time resource can be obtained. The AI ​​model (450) can be configured to have at least one channel When input, it can be trained to output at least one PMI corresponding to a given time resource.

[0097] The device has at least one channel By applying the AI ​​model (450), at least one CQI corresponding to a given time resource can be obtained. The AI ​​model (450) can obtain at least one channel When input, it can be trained to output at least one CQI corresponding to a given time resource.

[0098] The device can obtain at least one CQI corresponding to a given time resource by applying at least one CQI to the AI ​​model (450). The AI ​​model (450) can be trained to output at least one CQI corresponding to a given time resource when at least one CQI is input.

[0099] The device can obtain at least one precoding matrix corresponding to a given time resource by applying at least one PMI to the AI ​​model (450). The AI ​​model (450) can be trained to output at least one precoding matrix corresponding to a given time resource when at least one PMI is input.

[0100] The device (1200, 1300) can apply at least one PMI to the AI ​​model (450) to obtain at least one PMI corresponding to a given time resource. The AI ​​model (450) can be trained to output at least one PMI corresponding to a given time resource when at least one PMI is input.

[0101] FIG. 6 is a diagram for explaining a terminal performing AI-based CSI prediction according to one embodiment of the present disclosure.

[0102] A terminal can perform AI-based CSI prediction based on at least one reference signal and an AI model. By performing AI-based CSI prediction, the terminal can obtain CSI corresponding to a given time resource. The terminal can transmit the obtained CSI to the base station.

[0103] The given time resource is position information on the time axis and may include at least one of a time resource between the CSI RS reception time and the CSI reporting time or a time resource after the CSI reporting time.

[0104] Time resources can be identified by time units such as subframes, slots, mini-slots, and symbols, but these are only examples and the time units are not limited to those described above.

[0105] Referring to FIG. 6, the terminal transmits at least one channel based on at least one CSI-RS. , at least one V, at least one PMI, or at least one CQI can be obtained.

[0106] The terminal can receive at least one CSI-RS from the base station. The terminal estimates the channel state (channel estimation) through the received RS, and can be obtained. The terminal is based on methods such as EVD or SVD. The relationship can be obtained. The terminal is a channel Based on this, an eigenvector V associated with the precoding matrix can be generated. The terminal can generate a channel Based on this, you can obtain PMI or CQI.

[0107] The terminal is a channel , V, PMI, CQI, and can obtain AI-based CSI based on at least one of the following: For example, the terminal may obtain at least one precoding matrix V( ) is applied to the AI ​​model to obtain a precoding matrix corresponding to a given time resource ( ) can be obtained. The terminal can identify at least one PMI based on the obtained precoding matrix. The terminal can transmit CSI including the PMI to the base station.

[0108] The terminal can obtain at least one PMI corresponding to a given time resource by applying at least one V to the AI ​​model. The terminal can obtain at least one CQI corresponding to a given time resource by applying at least one V to the AI ​​model.

[0109] The terminal has at least one channel By applying the AI ​​model, at least one precoding matrix corresponding to a given time resource can be obtained. The terminal can obtain at least one channel By applying it to an AI model, at least one PMI corresponding to a given time resource can be obtained.

[0110] The terminal has at least one channel By applying the AI ​​model, at least one CQI corresponding to a given time resource can be obtained. The terminal can obtain at least one CQI corresponding to a given time resource by applying the at least one CQI to the AI ​​model.

[0111] The terminal can obtain at least one precoding matrix corresponding to a given time resource by applying at least one PMI to the AI ​​model. The terminal can obtain at least one PMI corresponding to a given time resource by applying at least one PMI to the AI ​​model.

[0112] FIG. 7 is a diagram illustrating a base station performing AI-based CSI prediction according to one embodiment of the present disclosure.

[0113] A base station can transmit at least one reference signal (RS) to a terminal. The terminal estimates the channel state (channel estimation) through the received RS, and can be obtained. The terminal is a channel Based on this, the eigenvector V associated with the precoding matrix can be generated. The terminal can generate the eigenvector V based on a method such as EVD or SVD. The relationship can be obtained. The terminal is a channel Based on this, CSI including PMI or CQI can be obtained. The terminal can transmit the CSI to the base station.

[0114] A base station can perform AI-based CSI prediction based on at least one CSI and AI model. By performing AI-based CSI prediction, the base station can obtain CSI corresponding to a given time resource. The base station can perform communications, such as scheduling terminals, based on the CSI.

[0115] The given time resource is position information on the time axis and may include at least one of a time resource between CSI reporting times or a time resource after the CSI reporting.

[0116] A given time resource may be identified by a time unit such as a subframe, slot, mini-slot, symbol, etc., but this is only an example and the time unit is not limited to the above.

[0117] The base station can obtain AI-based CSI based on at least one of at least one PMI and at least one CQI. For example, referring to FIG. 7, the base station can obtain at least one PMI ( ) can be applied to the AI ​​model to obtain at least one PMI corresponding to a given time resource.

[0118] The base station can obtain at least one precoding matrix corresponding to a given time resource by applying at least one PMI to the AI ​​model. The base station can obtain a CQI corresponding to a given time resource by applying at least one CQI to the AI ​​model.

[0119] FIG. 8 is a flowchart of a method for performing CSI reporting according to one embodiment of the present disclosure.

[0120] In step S810, the base station transmits information to the terminal indicating whether to perform AI (Artificial Intelligence)-based CSI prediction.

[0121] The terminal may receive information from the base station indicating whether to perform AI-based CSI prediction.

[0122] In the present disclosure, information indicating whether to perform AI-based CSI prediction or reporting may be referred to as UE prediction state information or UE measurement state information, but this is only an example and is not limited to the examples described above.

[0123] The terminal prediction status information may indicate at least one of whether the terminal performs CSI prediction using an AI model, whether the terminal performs CSI measurement without using an AI model, or whether the terminal reports CSI acquired using an AI model.

[0124] The terminal prediction status information may include 1 bit of information indicating whether the terminal will perform AI-based CSI prediction or reporting.

[0125] For example, if 1 bit of information is set to '1', it may indicate that the terminal will perform AI-based CSI prediction or reporting (prediction on). If 1 bit of information is set to '0', it may indicate that the terminal will not perform AI-based CSI prediction or reporting (prediction off).

[0126] For example, if 1 bit of information is set to '0', it may indicate that the terminal will perform AI-based CSI prediction or reporting (prediction on). If 1 bit of information is set to '1', it may indicate that the terminal will not perform AI-based CSI prediction or reporting (prediction off). However, this is merely an example and is not limited to the aforementioned method.

[0127] If the terminal prediction status information is configured to enable the terminal to perform AI-based CSI prediction or reporting, the terminal prediction status information may further include one bit of information indicating that the terminal will determine a time resource for performing prediction.

[0128] Terminal prediction status information may be included in a DL message. The terminal prediction status information may be delivered in the form of at least one of an RRC message, downlink control information (DCI), or MAC CE, but these are only examples and are not limited to the examples mentioned.

[0129] When terminal prediction status information is included in an RRC message, the terminal prediction status information may be included in a measurement configuration. The terminal prediction status information may be included in at least one of measurement object information, CSI report configuration, or CSI resource configuration, but this is only an example and is not limited to the examples mentioned. The CSI report configuration may include a codebook configuration.

[0130] In relation to AI-based CSI prediction or reporting, the terminal may be configured with at least one CSI report configuration, at least one CSI resource configuration, or at least one trigger state list via upper layer signaling (e.g., an RRC message).

[0131] AI-based CSI prediction or AI-based CSI reporting based on RRC messages can be configured to be periodic, aperiodic, or semi-persistent.

[0132] When terminal prediction status information is transmitted in the form of MAC CE or DCI, the terminal may trigger or activate / deactivate AI-based CSI prediction or reporting.

[0133] Trigger states may include aperiodic trigger states or semi-permanent trigger states. Each trigger state may include associated CSI reporting settings. Based on at least one of the MAC CE and DCI, the terminal may activate or deactivate corresponding AI-based CSI prediction or reporting.

[0134] When UE mobility is higher than a certain level, when channel variation due to delay is greater than a certain level, or when high-accuracy CSI is required, the UE prediction status information may indicate that AI-based CSI prediction or reporting will be performed.

[0135] When the required reliability level is high and the required latency constraint increases, the terminal prediction status information may indicate that the terminal will perform AI-based CSI prediction or reporting.

[0136] When the number of AI operation terminals allocated to the network increases, when use cases utilizing AI increase, when the network's available AI resources are insufficient, when the network's AI resources need to be reduced, when the network cannot operate AI, or when the terminal needs to operate AI alone, the terminal prediction status information may indicate that the terminal will perform AI-based CSI prediction or reporting.

[0137] If UE mobility is lower than a certain level, if the time correlation of the channel is higher than a certain level, if a precoding matrix with an accuracy lower than a certain level is available, if AI operation of the UE is unnecessary, or if the network intends to operate AI on its own, the UE prediction status information may indicate that the UE will not perform AI-based CSI prediction or reporting.

[0138] If the required reliability is low, the required CSI accuracy is low, the latency is low, or the sensitivity to retransmission due to CSI prediction failure is low, the terminal prediction status information may indicate that the terminal will not perform AI-based CSI prediction or reporting.

[0139] If AI operation on the terminal is not possible, if power consumption of the terminal needs to be reduced, if computing resources of the terminal need to be reduced, if other AI needs to be used on the terminal, or if the network has sufficient resources for AI operation, the terminal prediction status information may indicate that the terminal will not perform AI-based CSI prediction or reporting.

[0140] In one embodiment, the terminal may transmit a response message to the base station regarding whether the terminal will perform AI-based CSI prediction.

[0141] In step S820, the terminal can determine whether to perform AI-based CSI prediction.

[0142] Based on terminal status information, the terminal can determine whether to perform AI-based CSI prediction or reporting.

[0143] If the terminal status information indicates that AI-based CSI prediction or reporting will be performed, the terminal may determine to acquire CSI using the AI ​​model. If the terminal status information indicates that AI-based CSI prediction or reporting will not be performed, the terminal may determine to perform CSI measurement without using the AI ​​model.

[0144] In step S830, when performing AI-based CSI prediction, the terminal can obtain channel state information based on at least one reference signal and an AI model.

[0145] The terminal can obtain at least one reference signal. For example, the terminal can receive a reference signal from a base station.

[0146] The terminal can perform AI-based CSI prediction based on at least one reference signal and an AI model. By performing AI-based CSI prediction, the terminal can obtain CSI corresponding to a given time resource.

[0147] The terminal may transmit at least one channel based on at least one CSI-RS. , at least one V, at least one PMI, or at least one CQI can be obtained.

[0148] The terminal is a channel , V, PMI. AI-based CSI can be acquired based on at least one of CQI. The content of the terminal acquiring AI-based CSI can correspond to the content of Fig. 6, and some overlapping parts can be omitted. The content of the terminal acquiring CSI without using AI can correspond to the content of Fig. 2, and some overlapping parts can be omitted.

[0149] In step S840, the terminal can transmit channel state information to the base station.

[0150] Channel State Information (CSI) may include at least one of AI-based CSI and CSI acquired without AI. If a terminal performs AI-based CSI prediction or reporting, the terminal may transmit AI-based CSI to the base station.

[0151] A base station can receive CSI from a terminal. The base station can determine whether to perform AI-based CSI prediction. If performing AI-based CSI prediction, the base station can acquire at least one CSI corresponding to a predetermined time based on at least one CSI including the received CSI and an AI model. The details of how the base station acquires AI-based CSI may correspond to the details of FIG. 7, and some overlapping portions may be omitted.

[0152] FIG. 9 is a flowchart illustrating a process for transmitting status information regarding AI-based CSI prediction according to one embodiment of the present disclosure.

[0153] In step S910, the base station may transmit a request message to the terminal for status check regarding AI-based CSI prediction performance.

[0154] The terminal may receive a request message for status confirmation regarding AI-based CSI prediction performance from the base station, and may include the request message in a DL message.

[0155] In the present disclosure, a request message for checking the status of a terminal for performing AI-based CSI prediction may be referred to as an AI preference request message, an AI prediction preference request message, an AI measurement preference request message, an AI capability request message, an AI prediction capability request message, an AI measurement capability request message, or one of the first messages, but is not limited to the examples mentioned.

[0156] The AI ​​preference request message may include a request message to check at least one of whether the terminal prefers AI-based CSI prediction or reporting, whether the terminal is capable of or supports AI operation, the resource status related to AI operation of the terminal, or the battery status of the terminal.

[0157] The AI ​​preference request message may include a 1-bit information indicating whether or not the AI ​​operation is preferred. For example, a "1" in the 1-bit information may indicate a preference, while a "0" may indicate a non-preference. For example, a "0" in the 1-bit information may indicate a preference, while a "1" may indicate a non-preference. However, this is merely an example and is not limited to the aforementioned method.

[0158] The AI ​​preference request message may be delivered in the form of at least one of an RRC message, DCI, or MAC CE, but this is only an example and is not limited to the examples mentioned.

[0159] For example, when included in an RRC message, the AI ​​preference request message may be included in an RRC reconfiguration message or UE capability enquiry information.

[0160] When UE mobility is higher than a certain level, when channel variation due to delay is greater than a certain level, or when high-accuracy CSI is required, the base station can transmit an AI preference request message to the UE.

[0161] When the required reliability level is high and the required latency constraint increases, the base station can send an AI preference request message to the terminal.

[0162] When the number of AI operation terminals allocated to the network increases, when the use cases of the network utilizing AI increase, when the available AI resources of the network are insufficient or below a certain level, when the AI ​​resources of the network need to be reduced, when the network cannot operate AI, or when the terminal needs to operate AI alone, the base station can transmit an AI preference request message to the terminal.

[0163] In step S920, the terminal can transmit status information regarding AI-based prediction performance to the base station.

[0164] The terminal can transmit status information regarding AI-based prediction performance to the base station. This status information regarding AI-based prediction performance can be included in an UL message.

[0165] Status information regarding AI-based prediction performance may be referred to as one of AI preference information, AI prediction preference information, AI measurement preference information, AI capability information, AI prediction capability information, and AI measurement capability information, but this is only an example and is not limited to the examples mentioned.

[0166] Status information regarding AI-based prediction performance may include information regarding at least one of whether the terminal prefers AI-based CSI prediction or reporting, whether the terminal is capable of or supports AI operation, the resource status regarding AI operation of the terminal, or the battery status of the terminal.

[0167] Status information regarding AI-based prediction performance may be conveyed in the form of at least one of an RRC message, uplink control information (UCI), or MAC CE, but this is only an example and is not limited to the examples mentioned.

[0168] For example, when included in an RRC message, status information regarding AI-based prediction performance may be included in UE assistance information or UE capability information.

[0169] When a base station transmits an AI preference request message to a terminal, the terminal can transmit status information regarding AI-based prediction performance to the base station as a response message.

[0170] In one embodiment, step S910 may be omitted. Regardless of whether an AI preference request message is received, the terminal may transmit status information regarding AI-based prediction performance to the base station. For example, if the terminal periodically performs UL transmission, status information regarding AI-based prediction performance may be transmitted to the base station even if an AI preference request message is not received.

[0171] If the terminal desires a higher KPI (key performance indicator) while not performing AI-based CSI prediction, or if the terminal can secure a threshold amount of AI-capable resources, or if the terminal's battery is above a certain level, the terminal can transmit status information regarding the performance of AI-based CSI prediction to the base station.

[0172] In this case, status information regarding AI-based prediction performance may indicate that the terminal prefers AI-based CSI prediction or reporting, or that AI-based CSI prediction performance or reporting is possible.

[0173] When a terminal desires a lower KPI (key performance indicator) while performing AI-based CSI prediction, when the terminal's AI-operable resource capacity is below a threshold, when the terminal's battery is below a certain level, when the terminal has difficulty in continuing to perform AI-based CSI prediction, or when the terminal requests a change of AI operation entity to the base station, the terminal may transmit status information regarding the performance of AI-based CSI prediction to the base station.

[0174] In this case, the status information regarding AI-based prediction performance may indicate that the terminal does not prefer AI-based CSI prediction or reporting, or that the terminal is not capable of AI-based CSI prediction or reporting.

[0175] FIG. 10 is a flowchart of a method for performing CSI reporting according to one embodiment of the present disclosure.

[0176] In step S1010, the terminal can transmit status information regarding AI-based prediction performance to the base station.

[0177] The terminal can transmit status information regarding AI-based prediction performance to the base station. This status information regarding AI-based prediction performance can be included in an UL message.

[0178] Status information regarding AI-based prediction performance may be referred to as one of AI preference information, AI prediction preference information, AI measurement preference information, AI capability information, AI prediction capability information, and AI measurement capability information, but is not limited to the examples mentioned.

[0179] Status information regarding AI-based prediction performance may include information regarding at least one of whether the terminal prefers AI-based CSI prediction or reporting, whether the terminal is capable of or supports AI operation, the resource status regarding AI operation of the terminal, or the battery status of the terminal.

[0180] Status information regarding AI-based prediction performance may be conveyed in the form of at least one of an RRC message, DCI, or MAC CE, but this is only an example and is not limited to the examples mentioned.

[0181] For example, when included in an RRC message, status information regarding AI-based prediction performance may be included in UE assistance information or UE capability information.

[0182] When a base station transmits an AI preference request message to a terminal, the terminal may transmit status information regarding AI-based prediction performance to the base station as a response message. The step of the terminal receiving the AI ​​preference request message from the base station may further include. The content of the AI ​​preference request message may correspond to the content of step S910, and some overlapping portions may be omitted.

[0183] Regardless of whether an AI preference request message is received, the terminal may transmit status information regarding AI-based prediction performance to the base station.

[0184] For example, if a terminal periodically performs UL transmission, it can transmit status information regarding AI-based prediction performance to the base station even if an AI preference request message is not received.

[0185] If the terminal desires a higher KPI (key performance indicator) while not performing AI-based CSI prediction, or if the terminal can secure a threshold amount of AI-capable resources, or if the terminal's battery is above a certain level, the terminal can transmit status information regarding the performance of AI-based CSI prediction to the base station.

[0186] In this case, status information regarding AI-based prediction performance may indicate that the terminal prefers AI-based CSI prediction or reporting, or that AI-based CSI prediction performance or reporting is possible.

[0187] When a terminal desires a lower KPI (key performance indicator) while performing AI-based CSI prediction, when the terminal's AI-operable resource capacity is below a threshold, when the terminal's battery is below a certain level, when the terminal has difficulty in continuing to perform AI-based CSI prediction, or when the terminal requests a change of AI operation entity to the base station, the terminal may transmit status information regarding the performance of AI-based CSI prediction to the base station.

[0188] In this case, the status information regarding AI-based prediction performance may indicate that the terminal does not prefer AI-based CSI prediction or reporting, or that the terminal is not capable of AI-based CSI prediction or reporting.

[0189] In step S1020, the base station transmits information to the terminal indicating whether to perform AI-based CSI prediction.

[0190] The terminal may receive information from the base station indicating whether to perform AI-based CSI prediction.

[0191] In the present disclosure, information indicating whether to perform AI-based CSI prediction or reporting may be referred to as UE prediction state information or UE measurement state information, but this is only an example and is not limited to the examples described above.

[0192] The terminal prediction status information may indicate at least one of whether the terminal performs CSI prediction using an AI model, whether the terminal performs CSI measurement without using an AI model, or whether the terminal reports CSI acquired using an AI model.

[0193] The terminal prediction status information may include 1 bit of information indicating whether the terminal will perform AI-based CSI prediction or reporting.

[0194] For example, if 1 bit of information is set to '1', it may indicate that the terminal will perform AI-based CSI prediction or reporting (prediction on). If 1 bit of information is set to '0', it may indicate that the terminal will not perform AI-based CSI prediction or reporting (prediction off).

[0195] If the terminal prediction status information is configured to enable the terminal to perform AI-based CSI prediction or reporting, the terminal prediction status information may further include one bit of information indicating that the terminal will determine a time resource for performing prediction.

[0196] Terminal prediction status information may be included in a DL message. The terminal prediction status information may be delivered in at least one of an RRC message, DCI, or MAC CE, but these are only examples and are not limited to the examples mentioned.

[0197] When terminal prediction status information is included in an RRC message, the terminal prediction status information may be included in a measurement configuration. The terminal prediction status information may be included in at least one of measurement object information, CSI report configuration, or CSI resource configuration, but this is only an example and is not limited to the examples mentioned. The CSI report configuration may include a codebook configuration.

[0198] In relation to AI-based CSI prediction or reporting, the terminal may be configured with at least one CSI report configuration, at least one CSI resource configuration, or at least one trigger state list via upper layer signaling (e.g., an RRC message).

[0199] AI-based CSI prediction or AI-based CSI reporting based on RRC messages can be configured to be periodic, aperiodic, or semi-persistent.

[0200] When terminal prediction status information is transmitted in the form of MAC CE or DCI, the terminal may trigger or activate / deactivate AI-based CSI prediction or reporting.

[0201] Trigger states may include aperiodic trigger states or semi-permanent trigger states. Each trigger state may include associated CSI reporting settings. Based on at least one of the MAC CE and DCI, the terminal may activate or deactivate corresponding AI-based CSI prediction or reporting.

[0202] Based on status information regarding AI-based prediction performance, the base station can transmit terminal prediction status information to the terminal.

[0203] When UE mobility is higher than a certain level, when channel variation due to delay is greater than a certain level, or when high-accuracy CSI is required, the UE prediction status information may indicate that AI-based CSI prediction or reporting will be performed.

[0204] When the required reliability level is high and the required latency constraint increases, the terminal prediction status information may indicate that the terminal will perform AI-based CSI prediction or reporting.

[0205] When the number of AI operation terminals allocated to the network increases, when use cases utilizing AI increase, when the network's available AI resources are insufficient, when the network's AI resources need to be reduced, when the network cannot operate AI, or when the terminal needs to operate AI alone, the terminal prediction status information may indicate that the terminal will perform AI-based CSI prediction or reporting.

[0206] If UE mobility is lower than a certain level, if the time correlation of the channel is higher than a certain level, if a precoding matrix with an accuracy lower than a certain level is available, if AI operation of the UE is unnecessary, or if the network intends to operate AI on its own, the UE prediction status information may indicate that the UE will not perform AI-based CSI prediction or reporting.

[0207] If the required reliability is low, the required CSI accuracy is low, the latency is low, or the sensitivity to retransmission due to CSI prediction failure is low, the terminal prediction status information may indicate that the terminal will not perform AI-based CSI prediction or reporting.

[0208] If AI operation on the terminal is not possible, if power consumption of the terminal needs to be reduced, if computing resources of the terminal need to be reduced, if other AI needs to be used on the terminal, or if the network has sufficient resources for AI operation, the terminal prediction status information may indicate that the terminal will not perform AI-based CSI prediction or reporting.

[0209] In step S1030, the terminal can determine whether to perform AI-based CSI prediction.

[0210] Based on terminal status information, the terminal can determine whether to perform AI-based CSI prediction or reporting.

[0211] If the terminal status information indicates that AI-based CSI prediction or reporting will be performed, the terminal may determine to acquire CSI using the AI ​​model. If the terminal status information indicates that AI-based CSI prediction or reporting will not be performed, the terminal may determine to perform CSI measurement without using the AI ​​model.

[0212] In one embodiment, the terminal may transmit a response message to the base station regarding whether the terminal will perform AI-based CSI prediction.

[0213] The content of the terminal acquiring AI-based CSI may correspond to the content of FIG. 6 and step S830, and some overlapping portions may be omitted. The content of the terminal acquiring CSI without using AI may correspond to the content of FIG. 2, and some overlapping portions may be omitted.

[0214] In step S1040, the terminal can transmit status information regarding AI-based prediction performance to the base station.

[0215] The content of step S1040 may correspond to the content of step S1010, and some overlapping parts may be omitted.

[0216] In step S1050, the terminal can transmit channel state information to the base station.

[0217] Channel State Information (CSI) may include at least one of AI-based CSI and CSI acquired without AI. If a terminal performs AI-based CSI prediction or reporting, the terminal may transmit AI-based CSI to the base station.

[0218] A base station can receive CSI from a terminal. The base station can determine whether to perform AI-based CSI prediction. If performing AI-based CSI prediction, the base station can acquire at least one CSI corresponding to a predetermined time based on at least one CSI including the received CSI and an AI model. The details of how the base station acquires AI-based CSI may correspond to the details of FIG. 7, and some overlapping portions may be omitted.

[0219] In step S1060, the terminal can transmit status information regarding AI-based prediction performance to the base station.

[0220] The content of step S1060 may correspond to the content of step S1010, and some overlapping parts may be omitted.

[0221] In one embodiment of the present disclosure, at least one of steps S1010, S1040, or S1060 may be omitted. The network may determine whether the terminal is capable of performing AI operations and whether AI operations are preferred. If participation in AI operations is required while the terminal is not operating AI, the network may check whether the terminal is capable of AI operations.

[0222] FIG. 11 is a flowchart of a method for performing CSI reporting according to one embodiment of the present disclosure.

[0223] In step S1110, the base station transmits information to the terminal indicating whether to perform AI-based CSI prediction.

[0224] The terminal may receive information from the base station indicating whether to perform AI-based CSI prediction.

[0225] In the present disclosure, information indicating whether to perform AI-based CSI prediction or reporting may be referred to as UE prediction state information or UE measurement state information, but this is only an example and is not limited to the examples described above.

[0226] The terminal prediction status information may indicate at least one of whether the terminal performs CSI prediction using an AI model, whether the terminal performs CSI measurement without using an AI model, or whether the terminal reports CSI acquired using an AI model.

[0227] The terminal prediction status information may include 1 bit of information indicating whether the terminal will perform AI-based CSI prediction or reporting.

[0228] For example, if 1 bit of information is set to '1', it may indicate that the terminal will perform AI-based CSI prediction or reporting (prediction on). If 1 bit of information is set to '0', it may indicate that the terminal will not perform AI-based CSI prediction or reporting (prediction off).

[0229] If the terminal prediction status information is configured to enable the terminal to perform AI-based CSI prediction or reporting, the terminal prediction status information may further include one bit of information indicating that the terminal will determine a time resource for performing prediction.

[0230] Terminal prediction status information may be included in a DL message. The terminal prediction status information may be delivered in at least one of an RRC message, DCI, or MAC CE, but these are only examples and are not limited to the examples mentioned.

[0231] When terminal prediction status information is included in an RRC message, the terminal prediction status information may be included in a measurement configuration. The terminal prediction status information may be included in at least one of measurement object information, CSI report configuration, or CSI resource configuration, but this is only an example and is not limited to the examples mentioned. The CSI report configuration may include a codebook configuration.

[0232] In relation to AI-based CSI prediction or reporting, the terminal may be configured with at least one CSI report configuration, at least one CSI resource configuration, or at least one trigger state list via upper layer signaling (e.g., an RRC message).

[0233] AI-based CSI prediction or AI-based CSI reporting based on RRC messages can be configured to be periodic, aperiodic, or semi-persistent.

[0234] When terminal prediction status information is transmitted in the form of MAC CE or DCI, the terminal may trigger or activate / deactivate AI-based CSI prediction or reporting.

[0235] Trigger states may include aperiodic trigger states or semi-permanent trigger states. Each trigger state may include associated CSI reporting settings. Based on at least one of the MAC CE and DCI, the terminal may activate or deactivate corresponding AI-based CSI prediction or reporting.

[0236] When UE mobility is higher than a certain level, when channel variation due to delay is greater than a certain level, or when high-accuracy CSI is required, the UE prediction status information may indicate that AI-based CSI prediction or reporting will be performed.

[0237] When the required reliability level is high and the required latency constraint increases, the terminal prediction status information may indicate that the terminal will perform AI-based CSI prediction or reporting.

[0238] When the number of AI operation terminals allocated to the network increases, when use cases utilizing AI increase, when the network's available AI resources are insufficient, when the network's AI resources need to be reduced, when the network cannot operate AI, or when the terminal needs to operate AI alone, the terminal prediction status information may indicate that the terminal will perform AI-based CSI prediction or reporting.

[0239] If UE mobility is lower than a certain level, if the time correlation of the channel is higher than a certain level, if a precoding matrix with an accuracy lower than a certain level is available, if AI operation of the UE is unnecessary, or if the network intends to operate AI on its own, the UE prediction status information may indicate that the UE will not perform AI-based CSI prediction or reporting.

[0240] If the required reliability is low, the required CSI accuracy is low, the latency is low, or the sensitivity to retransmission due to CSI prediction failure is low, the terminal prediction status information may indicate that the terminal will not perform AI-based CSI prediction or reporting.

[0241] If AI operation on the terminal is not possible, if power consumption of the terminal needs to be reduced, if computing resources of the terminal need to be reduced, if other AI needs to be used on the terminal, or if the network has sufficient resources for AI operation, the terminal prediction status information may indicate that the terminal will not perform AI-based CSI prediction or reporting.

[0242] Based on status information regarding AI-based prediction performance, the base station can transmit terminal prediction status information to the terminal. The base station can receive status information regarding AI-based CSI prediction performance from the terminal. The content of receiving status information regarding AI-based CSI prediction performance may correspond to the content of S920 or S1010, and some overlapping portions may be omitted.

[0243] In step S1120, the base station transmits information indicating to the terminal the time resources that require AI-based CSI prediction.

[0244] A terminal may receive information from a base station indicating the time resources required for CSI prediction. The information indicating the time resources may be included in a DL message.

[0245] Information indicating a time resource may indicate location information on the time axis corresponding to the CSI to be acquired. The information indicating a time resource may include information regarding at least one of a starting time resource index, the number of consecutive time resources, an interval, a configuration duration period, a pattern, a format, or a mapping type. For example, a base station may transmit information indicating a time resource, including a slot index or a slot format index, to a terminal.

[0246] Information indicating time resources may be conveyed in the form of at least one of an RRC message, DCI, or MAC CE, but these are only examples and are not limited to the examples mentioned.

[0247] When information indicating time resources is transmitted via RRC messages, the terminal can be allocated time resources semi-statically. When information indicating time resources is transmitted using DCI, the terminal can be allocated time resources dynamically.

[0248] In one embodiment of the present disclosure, the base station may omit the step of transmitting information indicating a time resource or transmit information indicating a time resource set by default.

[0249] For example, if the CSI report is short-period, if the correlation between slots is high, if the terminal determines the time resource, or if the base station determines that slot index designation is unnecessary, the base station may omit the step of transmitting information indicating the time resource or transmit information indicating the time resource with the slot index set to 0.

[0250] When the CSI report is a long period, or the correlation between slots is low, or when the base station wants to newly set or change the time resource, the base station can transmit information indicating the time resource to the terminal.

[0251] For the convenience of explanation, slots are used as examples, but this is only an example, and time units such as subframes, mini-slots, and symbols may be used, and are not limited to the examples described above.

[0252] In one embodiment, the terminal transmits information to the base station indicating the time resources that require AI-based CSI prediction.

[0253] The time resource that requires AI-based CSI prediction is the time resource corresponding to the CSI acquired or to be acquired using the AI ​​model. The terminal can determine the time resource. The terminal can transmit information indicating the determined time resource to the base station.

[0254] Information indicating a time resource may indicate location information on the time axis corresponding to the CSI to be acquired. Information indicating a time resource may include information regarding at least one of a starting time resource index, the number of consecutive time resources, an interval, a period during which the configuration lasts, a pattern, a format, or a mapping type.

[0255] For example, the base station may transmit to the terminal information indicating a time resource including a slot index or a slot format index.

[0256] The terminal can transmit information to the base station indicating the time resources required for CSI prediction. The information indicating the time resources can be included in an UL message.

[0257] Information indicating time resources may be conveyed in the form of at least one of an RRC message, UCI, or MAC CE, but these are only examples and are not limited to the examples mentioned.

[0258] When information indicating time resources is transmitted via RRC messages, the terminal can configure the time resources semi-statically. When information indicating time resources is transmitted using DCI, the terminal can configure the time resources dynamically.

[0259] In one embodiment of the present disclosure, the terminal may omit the step of transmitting information indicating a time resource or transmit information indicating a time resource set by default.

[0260] For example, if the CSI report is short-period, if the correlation between slots is high, or if the terminal determines that slot index designation is unnecessary, the terminal may omit the step of transmitting information indicating a time resource or transmit information indicating a time resource with the slot index set to 0.

[0261] If the CSI report is a long period, if the correlation between slots is low, if the base station does not have capability information about the AI ​​model of the terminal, if the base station does not know which slot index the terminal can predict CSI for, or if the terminal wants to newly set or change the time resource, the terminal may transmit information indicating the time resource.

[0262] For the convenience of explanation, slots are used as examples, but this is only an example, and time units such as subframes, mini-slots, and symbols may be used, and are not limited to the examples described above.

[0263] In step S1130, the terminal can determine whether to perform AI-based CSI prediction.

[0264] Based on terminal status information, the terminal can determine whether to perform AI-based CSI prediction or reporting.

[0265] If the terminal status information indicates that AI-based CSI prediction or reporting will be performed, the terminal may determine to acquire CSI using the AI ​​model. If the terminal status information indicates that AI-based CSI prediction or reporting will not be performed, the terminal may determine to perform CSI measurement without using the AI ​​model.

[0266] In one embodiment, the terminal may transmit a response message to the base station regarding whether the terminal will perform AI-based CSI prediction.

[0267] In step S1140, when performing AI-based CSI prediction, the terminal can obtain channel state information based on at least one reference signal and an AI model.

[0268] The terminal can obtain at least one reference signal. The terminal can receive the reference signal from the base station.

[0269] The terminal can perform AI-based CSI prediction based on at least one reference signal and an AI model. By performing AI-based CSI prediction, the terminal can obtain CSI corresponding to a given time resource.

[0270] The terminal may transmit at least one channel based on at least one CSI-RS. , at least one V, at least one PMI, or at least one CQI can be obtained.

[0271] The terminal is a channel , V, PMI. AI-based CSI can be acquired based on at least one of CQI. The content of the terminal acquiring AI-based CSI can correspond to the content of Fig. 6, and some overlapping parts can be omitted. The content of the terminal acquiring CSI without using AI can correspond to the content of Fig. 2, and some overlapping parts can be omitted.

[0272] In step S1150, the terminal can transmit channel state information to the base station.

[0273] Channel State Information (CSI) may include at least one of AI-based CSI and CSI acquired without AI. If a terminal performs AI-based CSI prediction or reporting, the terminal may transmit AI-based CSI to the base station.

[0274] A base station can receive CSI from a terminal. The base station can determine whether to perform AI-based CSI prediction. If performing AI-based CSI prediction, the base station can acquire at least one CSI corresponding to a predetermined time based on at least one CSI including the received CSI and an AI model. The details of how the base station acquires AI-based CSI may correspond to the details of FIG. 7, and some overlapping portions may be omitted.

[0275] FIG. 12 is a block diagram schematically illustrating the configuration of a base station (1200) according to one embodiment of the present disclosure.

[0276] Referring to FIG. 12, the base station (1200) may be composed of a transceiver (1210), a processor (1230), and a memory (1220). Depending on the communication method of the base station (1200) described above, the transceiver (1210), the processor (1230), and the memory (1220) of the base station (1200) may operate. However, the components of the base station (1200) are not limited to the examples described above.

[0277] For example, the base station (1200) may include more or fewer components than the aforementioned components. In one embodiment, the transceiver (1210), processor (1230), and memory (1220) may be implemented in the form of a single chip. In addition, the processor (1230) may include one or more processors.

[0278] The transceiver (1210) is a general term for the receiver of the base station (1200) and the transmitter of the base station (1200), and can transmit and receive signals with a terminal or a network entity. The signals transmitted and received with the terminal or network entity may include control information and data. To this end, the transceiver (1210) may be configured with an RF transmitter that up-converts and amplifies the frequency of a transmitted signal, and an RF receiver that low-noise amplifies and frequency-downconverts a received signal. However, this is only one embodiment of the transceiver (1210), and the components of the transceiver (1210) are not limited to the RF transmitter and RF receiver.

[0279] Additionally, the transceiver (1210) can perform functions for transmitting and receiving signals via a wireless channel. For example, the transceiver (1210) can receive a signal via a wireless channel, output it to the processor (1230), and transmit the signal output from the processor (1230) via the wireless channel.

[0280] The memory (1220) can store programs and data required for the operation of the base station (1200). In addition, the memory (1220) can store control information or data included in a signal acquired from the base station. The memory (1220) can be configured as a storage medium or a combination of storage media, such as a ROM, a RAM, a hard disk, a CD-ROM, and a DVD. In addition, the memory (1220) may not exist separately but may be configured as included in the processor (1230). The memory (1220) can be configured as a volatile memory, a nonvolatile memory, or a combination of volatile and nonvolatile memories. In addition, the memory (1220) can provide stored data upon request of the processor (1230).

[0281] The processor (1230) may control a series of processes so that the base station (1200) can operate according to the above-described embodiment of the present disclosure. For example, the processor (1230) may receive control signals and data signals through the transceiver (1210) and process the received control signals and data signals. The processor (1230) may transmit the processed control signals and data signals through the transceiver (1210). In addition, the processor (1230) may write or read data to or from the memory (1220). The processor (1230) may perform functions of a protocol stack required by a communication standard. To this end, the processor (1230) may include at least one processor or microprocessor. In one embodiment, a portion of the transceiver (1210) or the processor (1230) may be referred to as a communication processor (CP).

[0282] The processor (1230) may be composed of one or more processors. In this case, one or more processors may be a general-purpose processor, such as a CPU, AP, or DSP (Digital Signal Processor), a graphics-only processor, such as a GPU or VPU (Vision Processing Unit), or an AI-only processor, such as an NPU. For example, if one or more processors are AI-only processors, the AI-only processor may be designed with a hardware structure specialized for processing a specific AI model.

[0283] According to an embodiment of the present disclosure, the processor (1230) may transmit information to the terminal indicating whether the terminal will perform AI-based CSI prediction. The processor (1230) may receive channel state information from the terminal. If the terminal performs AI-based CSI prediction, the channel state information may be acquired based on at least one reference signal and a first AI model.

[0284] According to an embodiment of the present disclosure, the processor (1230) can obtain AI-based channel state information based on at least one channel state information including channel state information and a second AI model.

[0285] According to an embodiment of the present disclosure, the processor (1230) can receive status information regarding AI-based CSI prediction performance from the terminal.

[0286] According to an embodiment of the present disclosure, the processor (1230) may transmit a request message to the terminal for status confirmation regarding AI-based CSI prediction performance.

[0287] According to an embodiment of the present disclosure, the processor (1230) may transmit information indicating time resources that require AI-based CSI prediction to the terminal.

[0288] According to an embodiment of the present disclosure, the processor (1230) may receive information indicating a prediction resource requiring AI-based CSI prediction from a terminal.

[0289] According to an embodiment of the present disclosure, the processor (1230) may receive a response message from the terminal regarding whether the terminal will perform AI-based CSI prediction.

[0290] FIG. 13 is a block diagram schematically illustrating the configuration of a terminal (1300) according to one embodiment of the present disclosure.

[0291] Referring to FIG. 13, the terminal (1300) may be composed of a transceiver (1310), a processor (1330), and a memory (1320). Depending on the communication method of the terminal (1300) described above, the transceiver (1310), the processor (1330), and the memory (1320) of the terminal (1300) may operate. However, the components of the terminal (1300) are not limited to the examples described above.

[0292] For example, the terminal (1300) may include more or fewer components than the aforementioned components. In one embodiment, the transceiver (1310), processor (1330), and memory (1320) may be implemented in the form of a single chip. Additionally, the processor (1330) may include one or more processors.

[0293] The transceiver (1310) is a general term for the receiver of the terminal (1300) and the transmitter of the terminal (1300), and can transmit and receive signals with at least one of another terminal, a base station, or a network entity. The signals transmitted and received with the terminal or network entity may include control information and data. To this end, the transceiver (1310) may be configured with an RF transmitter that up-converts and amplifies the frequency of a transmitted signal, and an RF receiver that low-noise amplifies and frequency-downconverts a received signal. However, this is only one embodiment of the transceiver (1310), and the components of the transceiver (1310) are not limited to the RF transmitter and the RF receiver.

[0294] Additionally, the transceiver (1310) can perform functions for transmitting and receiving signals via a wireless channel. For example, the transceiver (1310) can receive a signal via a wireless channel, output it to the processor (1330), and transmit the signal output from the processor (1330) via the wireless channel.

[0295] The memory (1320) can store programs and data required for the operation of the terminal (1300). In addition, the memory (1320) can store control information or data included in a signal acquired from the base station. The memory (1320) can be configured as a storage medium or a combination of storage media, such as a ROM, a RAM, a hard disk, a CD-ROM, and a DVD. In addition, the memory (1320) may not exist separately but may be configured as included in the processor (1330). The memory (1320) can be configured as a volatile memory, a nonvolatile memory, or a combination of volatile and nonvolatile memories. In addition, the memory (1320) can provide stored data upon request of the processor (1330).

[0296] The processor (1330) may control a series of processes so that the terminal (1300) can operate according to the above-described embodiment of the present disclosure. For example, the processor (1330) may receive control signals and data signals through the transceiver (1310) and process the received control signals and data signals. The processor (1330) may transmit the processed control signals and data signals through the transceiver (1310). In addition, the processor (1330) may write or read data to or from the memory (1320). The processor (1330) may perform functions of a protocol stack required by a communication standard. To this end, the processor (1330) may include at least one processor or microprocessor. In one embodiment, a portion of the transceiver (1310) or the processor (1330) may be referred to as a communication processor (CP).

[0297] The processor (1330) may be comprised of one or more processors. In this case, one or more processors may be a general-purpose processor, such as a CPU, AP, or DSP (Digital Signal Processor), a graphics-only processor, such as a GPU or VPU (Vision Processing Unit), or an AI-only processor, such as an NPU. For example, if one or more processors are AI-only processors, the AI-only processor may be designed with a hardware structure specialized for processing a specific AI model.

[0298] According to an embodiment of the present disclosure, a processor (1330) may receive information from a base station indicating whether a terminal performs AI-based channel state information (CSI) prediction. At least one processor may determine whether to perform AI-based CSI prediction based on the information; and, when performing AI-based CSI prediction, may obtain channel state information based on at least one reference signal and a first AI model. At least one processor may transmit the channel state information.

[0299] According to an embodiment of the present disclosure, the processor (1330) can transmit status information regarding AI-based CSI prediction performance to the base station.

[0300] According to an embodiment of the present disclosure, the processor (1330) may receive a request message for status confirmation regarding AI-based CSI prediction performance from a base station.

[0301] According to an embodiment of the present disclosure, the processor (1330) may receive information from a base station indicating a time resource for which a terminal requires AI-based CSI prediction.

[0302] According to an embodiment of the present disclosure, the processor (1330) can determine time resources that require AI-based CSI prediction. The processor (1330) can transmit information indicating the time resources to the base station.

[0303] According to an embodiment of the present disclosure, the processor (1330) may transmit a response message to the base station regarding whether the terminal will perform AI-based CSI prediction.

[0304] According to an embodiment of the present disclosure, a method performed by a terminal in a wireless communication system is provided. The method may include receiving information from a base station indicating whether the terminal will perform AI-based channel state information (CSI) prediction. The method may include determining whether to perform AI-based CSI prediction based on the information. When performing AI-based CSI prediction, the method may include obtaining channel state information based on at least one reference signal and a first AI model. The method may include transmitting the channel state information.

[0305] According to an embodiment of the present disclosure, the method may include a step of transmitting status information regarding performance of AI-based CSI prediction to a base station.

[0306] According to an embodiment of the present disclosure, the method may include receiving a request message for status confirmation regarding AI-based CSI prediction performance from a base station.

[0307] According to an embodiment of the present disclosure, the method may include a step of receiving information from a base station indicating a time resource for which AI-based CSI prediction is required by a terminal.

[0308] According to an embodiment of the present disclosure, a method may include a step of determining a time resource requiring AI-based CSI prediction. According to an embodiment of the present disclosure, the method may include a step of transmitting information indicating the time resource to a base station.

[0309] According to an embodiment of the present disclosure, the method may include a step of transmitting a response message to the base station regarding whether the terminal will perform AI-based CSI prediction.

[0310] According to an embodiment of the present disclosure, a method performed by a base station in a wireless communication system is provided. The method may include transmitting information to a terminal indicating whether the terminal will perform AI-based CSI prediction. The method may include receiving channel state information from the terminal. If the terminal performs AI-based CSI prediction, the channel state information may be obtained based on at least one reference signal and a first AI model.

[0311] According to an embodiment of the present disclosure, the method may include obtaining AI-based channel state information based on at least one channel state information including channel state information and a second AI model.

[0312] According to an embodiment of the present disclosure, the method may include a step of receiving, from a terminal, status information regarding performance of AI-based CSI prediction.

[0313] According to an embodiment of the present disclosure, the method may include a step of transmitting a request message to a terminal for checking a status regarding AI-based CSI prediction performance.

[0314] According to an embodiment of the present disclosure, the method may include a step of transmitting information indicating to the terminal a time resource requiring AI-based CSI prediction.

[0315] According to an embodiment of the present disclosure, the method may include receiving information indicating a prediction resource requiring AI-based CSI prediction from a terminal.

[0316] According to an embodiment of the present disclosure, the method may include a step of receiving a response message from the terminal regarding whether the terminal will perform AI-based CSI prediction.

[0317] Embodiments of the present disclosure may be implemented or supported by one or more computer programs, and the computer programs may be formed from computer-readable program code and embodied in a computer-readable medium.

[0318] In embodiments of the present disclosure, "application" and "program" may refer to one or more computer programs, software components, sets of instructions, procedures, functions, objects, classes, instances, associated data, or portions thereof suitable for implementation in computer-readable program code. "Computer-readable program code" may include various types of computer code, including source code, object code, and executable code.

[0319] "Computer-readable media" may include various types of media that can be accessed by a computer, such as read only memory (ROM), random access memory (RAM), a hard disk drive (HDD), a compact disc (CD), a digital video disc (DVD), or various types of memory.

[0320] Additionally, the device-readable storage medium may be provided in the form of a non-transitory storage medium. Here, a "non-transitory storage medium" refers to a tangible device and may exclude wired, wireless, optical, or other communication links that transmit temporary electrical or other signals. Furthermore, this "non-transitory storage medium" does not distinguish between cases where data is permanently stored in the storage medium and cases where it is temporarily stored.

[0321] For example, a 'non-transitory storage medium' may include a buffer in which data is temporarily stored. A computer-readable medium may be any available medium that can be accessed by a computer, and may include both volatile and non-volatile media, removable and non-removable media. A computer-readable medium includes a medium on which data can be permanently stored and a medium on which data can be stored and later overwritten, such as a rewritable optical disk or an erasable memory device.

[0322] According to one embodiment, the methods according to the various embodiments disclosed in this document may be provided as a computer program product. The computer program product may be traded as a product between a seller and a buyer.

[0323] The computer program product may be distributed in the form of a machine-readable storage medium (e.g., compact disc read only memory (CD-ROM)), or may be distributed online (e.g., by download or upload) via an application store or directly between two user devices (e.g., smartphones).

[0324] In the case of online distribution, at least a portion of the computer program product (e.g., a downloadable app) may be temporarily stored or temporarily created in a device-readable storage medium, such as the memory of a manufacturer's server, an application store's server, or an intermediary server.

[0325] According to an embodiment of the present disclosure, a CSI reporting method and device using AI that compensates for a situation in which a channel changes between when a CSI-RS is transmitted and when the CSI is reported and actually used can be provided.

[0326] The effects of the present invention are not limited to the effects mentioned above, and other effects not mentioned will be clearly understood by those skilled in the art from the detailed description and the description of the claims.

[0327] The above description of the present disclosure is for illustrative purposes only, and those skilled in the art will appreciate that the present disclosure can be readily modified into other specific forms without altering the technical spirit or essential characteristics of the present disclosure. For example, appropriate results can be achieved even if the described techniques are performed in a different order than the described method, and / or components of the described systems, structures, devices, circuits, etc. are combined or combined in a different form than the described method, or are replaced or substituted with other components or equivalents.

[0328] Therefore, the embodiments described above should be understood as illustrative in all respects and not restrictive. For example, each component described as a single entity may be implemented in a distributed manner, and similarly, components described as distributed may be implemented in a combined manner. For example, some of the steps in FIGS. 8 through 11 may be implemented in a combined manner or in a distributed manner.

[0329] The scope of the present disclosure is indicated by the claims described below rather than the detailed description above, and all changes or modifications derived from the meaning and scope of the claims and their equivalent concepts should be interpreted as being included in the scope of the present disclosure.

Claims

1. In a method performed by a terminal in a wireless communication system, A step of receiving information from a base station indicating whether the terminal will perform AI (Artificial Intelligence)-based channel state information (CSI) prediction; A step of determining whether to perform AI-based CSI prediction based on the above information; When performing the AI-based CSI prediction, a step of obtaining channel state information based on at least one reference signal and a first AI model; and A method comprising the step of transmitting the channel state information to a base station.

2. In paragraph 1, A method further comprising the step of transmitting status information regarding the performance of the AI-based CSI prediction to the base station.

3. In paragraph 2, A method further comprising the step of receiving a request message for checking the status of performing the AI-based CSI prediction from the base station.

4. In paragraph 1, A method further comprising the step of receiving, from the base station, information indicating a time resource for which the terminal requires AI-based CSI prediction.

5. In paragraph 1, A step for determining time resources required for the AI-based CSI prediction; and A method comprising the step of transmitting information indicating the time resource to the base station.

6. In paragraph 1, A method further comprising the step of transmitting a response message to the base station regarding whether the terminal will perform the AI-based CSI prediction.

7. In a method performed by a base station in a wireless communication system, A step of transmitting information to a terminal indicating whether the terminal will perform AI (Artificial Intelligence)-based CSI (channel state information) prediction; and A step of receiving channel status information from the terminal is included, A method in which, when the terminal performs the AI-based CSI prediction, the channel state information is obtained based on at least one reference signal and a first AI model.

8. In paragraph 7, A method further comprising the step of obtaining AI-based channel state information based on at least one channel state information including the channel state information and a second AI model.

9. In paragraph 7, A method further comprising a step of receiving status information regarding the performance of the AI-based CSI prediction from the terminal.

10. In paragraph 9, A method further comprising the step of transmitting a request message for checking the status of performing the AI-based CSI prediction to the terminal.

11. In paragraph 7, A method further comprising the step of transmitting information indicating to the terminal a time resource requiring the AI-based CSI prediction.

12. In paragraph 7, A method further comprising the step of receiving information indicating a prediction resource requiring the AI-based CSI prediction from the terminal.

13. In paragraph 7, A method further comprising the step of receiving a response message from the terminal regarding whether the terminal will perform the AI-based CSI prediction.

14. The terminal of the wireless communication system is, Transmitter and receiver; and At least one processor connected to the transceiver, At least one processor of the above, Receive information from a base station indicating whether the terminal will perform AI (Artificial Intelligence)-based channel state information (CSI) prediction; Based on the above information, determine whether to perform the AI-based CSI prediction; When performing AI-based CSI prediction, channel state information is obtained based on at least one reference signal and a first AI model; and A terminal that transmits the above channel status information to the base station.

15. The base station of the wireless communication system is, Transmitter and receiver; and At least one processor connected to the transceiver, At least one processor of the above, Transmitting information to the terminal indicating whether the terminal will perform AI (Artificial Intelligence)-based CSI (channel state information) prediction; and Receive channel status information from the above terminal, A base station, wherein when the terminal performs the AI-based CSI prediction, the channel state information is obtained based on at least one reference signal and the first AI model.

Images