Method and device for spatial-domain channel prediction based on artificial intelligence

Abstract

The present disclosure relates to a 5G or 6G communication system for supporting higher data transmission rates than 4G communication systems such as LTE. The present disclosure relates to a method performed by a terminal in a wireless communication system and a device performing same, the method comprising the steps of: receiving information related to a channel state information-reference signal (CSI-RS) from a base station; measuring a partial antenna port CSI-RS, transmitted from the base station, on the basis of the information related to the CSI-RS; estimating a channel state for a partial antenna port on the basis of the measurement of the partial antenna port CSI-RS; acquiring channel state information about a full antenna port on the basis of the channel state for the partial antenna port; and reporting the channel state information about the full antenna port to the base station.

Description

AI-based space-domain channel prediction method and device

[0001] The present disclosure relates to a method and device for channel prediction in a wireless communication system. More specifically, the present disclosure relates to an artificial intelligence (AI)-based spatial-domain channel prediction method and device for reducing the overhead of a channel state information-reference signal (CSI-RS) and a sound reference signal (SRS) in a massive multiple-input-multiple-output (MIMO) system. The present disclosure relates to a method and device for measuring (or predicting) a channel state using AI to reduce the overhead of a reference signal in a 6G MIMO system.

[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 expected to evolve into diverse form factors, including 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 terahertz bands (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 important 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 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 (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 be applied in various fields such as industry, medicine, automobiles, and home appliances.

[0007] The purpose of this disclosure is to propose a method for reducing CSI-RS and SRS overhead through spatial-domain channel prediction using artificial intelligence (AI).

[0008] The technical problems to be achieved in the present disclosure are not limited to the technical problems mentioned above, and other technical problems not mentioned will be clearly understood by a person having ordinary skill in the technical field to which the present invention pertains from the description below.

[0009] According to one embodiment of the present disclosure, a method performed by a terminal in a wireless communication system may be provided, the method including the steps of: receiving information related to a channel state information reference signal (CSI-RS) from a base station; measuring a partial antenna port channel state information reference signal (partial antenna port CSI-RS) transmitted from the base station based on the information related to the channel state information reference signal; estimating a channel state for a partial antenna port based on the measurement of the partial antenna port channel state information reference signal; obtaining channel state information for a full antenna port based on the channel state for the partial antenna port; and reporting the channel state information for the full antenna port to the base station.

[0010] According to one embodiment of the present disclosure, a method performed by a base station in a wireless communication system may be provided, the method comprising the steps of: transmitting information related to a channel state information reference signal (CSI-RS) to a terminal; transmitting a partial antenna port channel state information reference signal (partial antenna port CSI-RS) to the terminal based on the information related to the channel state information reference signal; and receiving channel state information for a full antenna port from the terminal, wherein the channel state information for the full antenna port is obtained based on a channel state for a partial antenna port measured based on measurement of the partial antenna port channel state information reference signal.

[0011] According to one embodiment of the present disclosure, in a wireless communication system, a terminal may be provided, including a transceiver and a control unit, wherein the control unit receives information related to a channel state information reference signal (CSI-RS) from a base station, measures a partial antenna port channel state information reference signal (partial antenna port CSI-RS) transmitted from the base station based on the information related to the channel state information reference signal, estimates a channel state for a partial antenna port based on the measurement of the partial antenna port channel state information reference signal, obtains channel state information for a full antenna port based on the channel state for the partial antenna port, and reports the channel state information for the full antenna port to the base station.

[0012] According to one embodiment of the present disclosure, in a wireless communication system, a base station in a wireless communication system includes a transceiver and a control unit, wherein the control unit transmits information related to a channel state information reference signal (CSI-RS) to a terminal, transmits a partial antenna port channel state information reference signal (partial antenna port CSI-RS) to the terminal based on the information related to the channel state information reference signal, and controls to receive channel state information for a full antenna port, and the channel state information for the full antenna port is obtained based on a channel state for a partial antenna port measured based on measurement of the partial antenna port channel state information reference signal.

[0013] According to one embodiment of the present disclosure, an improved channel prediction method and device can be provided in a wireless communication system.

[0014] According to one embodiment of the present disclosure, an AI-based spatial-domain channel prediction method and device for reducing overhead of CSI-RS or SRS in a multi-antenna technology system can be provided.

[0015] According to one embodiment of the present disclosure, spatial-domain channel prediction can be performed through port pattern configuration. This can reduce CSI-RS and SRS overhead because full port channels are estimated through channels estimated with a small number of ports. This can address issues arising from the increase in the number of antenna ports in 5G massive MIMO systems or future 6G X-MIMO systems.

[0016] The effects that can be obtained from the present disclosure are not limited to the effects mentioned in the embodiments, and other effects that are not mentioned can be clearly understood by a person having ordinary skill in the art to which the present disclosure belongs from the description below.

[0017] FIG. 1 is a diagram illustrating the concept of spatial-domain channel prediction using AI to reduce overhead of CSI-RS applied to various embodiments of the present disclosure.

[0018] FIG. 2 is a diagram illustrating an example of AI utilization in a one-sided AI model applied to various embodiments of the present disclosure.

[0019] FIG. 3 is a diagram illustrating an example of AI utilization in a two-sided AI model applied to various embodiments of the present disclosure.

[0020] FIG. 4 is a diagram for explaining the relationship between a base station and a terminal in an AI-based spatial domain channel estimation process according to one embodiment of the present disclosure.

[0021] FIG. 5 is a diagram for explaining a procedure of a terminal in the process of AI-based spatial domain channel estimation according to one embodiment of the present disclosure.

[0022] FIG. 6 is a diagram for explaining the procedure of a base station in the process of AI-based spatial domain channel estimation according to one embodiment of the present disclosure.

[0023] FIG. 7 is an example of a guideline for expressing a port pattern as a bitmap according to an embodiment of the present disclosure, and an example of a port pattern and bitmap expressed according to the guideline.

[0024] FIG. 8a is an example of a port pattern according to a port reduction ratio by port number according to an embodiment of the present disclosure.

[0025] FIG. 8b is an example of a port pattern according to a port reduction ratio by port number according to an embodiment of the present disclosure.

[0026] FIG. 8c is an example of a port pattern according to a port reduction ratio by port number according to an embodiment of the present disclosure.

[0027] FIG. 8d is an example of a port pattern according to a port reduction ratio by port number according to an embodiment of the present disclosure.

[0028] FIG. 8e is an example of a port pattern according to a port reduction ratio by port number according to an embodiment of the present disclosure.

[0029] FIG. 8f is an example of a port pattern according to a port reduction ratio by port number according to an embodiment of the present disclosure.

[0030] FIG. 8g is an example of a port pattern according to a port reduction ratio by port number according to an embodiment of the present disclosure.

[0031] FIG. 8h is an example of a port pattern according to a port reduction ratio by port number according to an embodiment of the present disclosure.

[0032] FIG. 9 is a diagram illustrating a CSI-RS-based spatial domain channel prediction method between a terminal and a base station according to one embodiment of the present disclosure.

[0033] FIG. 10 is a diagram illustrating a CSI-RS-based spatial domain channel prediction method of a terminal according to one embodiment of the present disclosure.

[0034] FIG. 11 is a diagram illustrating a CSI-RS-based spatial domain channel prediction method of a base station according to one embodiment of the present disclosure.

[0035] FIG. 12 is a diagram illustrating a transmit antenna switching (TAS) technology related to SRS according to one embodiment of the present disclosure.

[0036] FIG. 13 is a diagram illustrating an SRS-based spatial domain channel prediction method between a terminal and a base station according to one embodiment of the present disclosure.

[0037] FIG. 14 is a diagram illustrating a method for predicting a terminal's SRS-based spatial domain channel according to an embodiment of the present disclosure.

[0038] FIG. 15 is a diagram illustrating a method for predicting a spatial domain channel based on SRS of a base station according to an embodiment of the present disclosure.

[0039] FIG. 16 is a diagram illustrating the structure of a terminal according to an embodiment of the present disclosure.

[0040] FIG. 17 is a diagram illustrating the structure of a base station according to an embodiment of the present disclosure.

[0041] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the attached drawings.

[0042] In describing the embodiments, 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 convey the gist of the present disclosure more clearly without obscuring it by omitting unnecessary explanations.

[0043] 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. Identical or corresponding components in each drawing are assigned the same reference numbers.

[0044] The advantages and features of the present disclosure, and methods for achieving them, will become clearer with reference to the embodiments described below in detail 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 designate like elements throughout the specification. In addition, when describing the present disclosure, if a specific description of a related function or configuration is determined to unnecessarily obscure the gist of the present disclosure, a detailed description thereof will be omitted. In addition, the terms described below are terms defined in consideration of the functions of the present disclosure, and may vary depending on the intention or custom of the user or operator. Therefore, their definitions should be made based on the contents throughout the specification.

[0045] Hereinafter, the base station is an entity that performs resource allocation of the terminal, and may be at least one of a gNode B, an eNode B, a Node B, a BS (Base Station), a wireless access unit, a base station controller, or a node on a network. The terminal may include a UE (User Equipment), an MS (Mobile Station), a cellular phone, a smartphone, a computer, or a multimedia system capable of performing a communication function. In the present disclosure, downlink (DL) refers to a wireless transmission path of a signal transmitted from a base station to a terminal, and uplink (UL) refers to a wireless transmission path of a signal transmitted from a terminal to a base station. In addition, although the LTE or LTE-A system may be described below as an example, the embodiments of the present disclosure may also be applied to other communication systems having a similar technical background or channel type. For example, the 5th generation mobile communication technology (5G, new radio, NR) developed after LTE-A may be included here, and the 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 scope that does not significantly deviate from the scope of the present disclosure, as judged by a person having skilled technical knowledge.

[0046] At this time, it will be understood that each block of the processing flowchart drawings and combinations of the flowchart drawings can be performed by computer program instructions. These computer program instructions can be installed in a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing equipment, so that the instructions executed by the processor of the computer or other programmable data processing equipment create a means for performing the functions described in the flowchart block(s). These computer program instructions can also be stored in a computer-available or computer-readable memory that can direct a computer or other programmable data processing equipment to implement the functions in a specific manner, so that the instructions stored in the computer-available or computer-readable memory can also produce a manufactured item that includes an instruction means for performing the functions described in the flowchart block(s). 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).

[0047] 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 implementation examples, 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.

[0048] Also, it will be understood that singular expressions such as "a" and "the above" in one embodiment of the present disclosure include plural expressions unless they clearly indicate different content. Also, in one embodiment of the present disclosure, terms including ordinal numbers such as first, second, etc. may be used to describe various components, but the components are not limited by the terms. The terms are used only for the purpose of distinguishing one component from another. For example, without departing from the scope of the present disclosure, the first component may be referred to as the second component, and similarly, the second component may also be referred to as the first component. Also, in one embodiment of the present disclosure, the term "and / or" includes a combination of a plurality of related listed items or any item among a plurality of related listed items.

[0049] In addition, the terms used in the embodiments of the present disclosure are only used to describe specific embodiments and are not intended to limit the present disclosure. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this specification, it should be understood that the terms "comprise" or "have" are intended to specify the presence of a feature, number, step, operation, component, part, or combination thereof described in the specification, but do not exclude in advance the possibility of the presence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof. Additionally, the terms “associated with” and “associated therewith” and their derivatives used in one embodiment of the present disclosure may mean include, be included within, interconnect with, contain, be contained within, connect to or with, couple to or with, be communicated with, cooperate with, interleave, juxtapose, be proximate to, be bound to or with, have, have a property of, and the like. Additionally, the term “report” in the present disclosure may be used with the same meaning as “transmit.”

[0050] Here, the term '~ unit' used in the present embodiment means software or hardware components such as FPGA (Field Programmable Gate Array) or ASIC (Application Specific Integrated Circuit), and the '~ unit' performs certain roles. However, the '~ unit' is not limited to software or hardware. The '~ unit' may be configured to be on an addressable storage medium and may be configured to play one or more processors. Therefore, as an example, the '~ unit' includes components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables. The functions provided within the components and '~ units' may be combined into a smaller number of components and '~ units' or further separated into additional components and '~ units'. In addition, the components and '~parts' may be implemented to regenerate one or more CPUs within a device or a secure multimedia card. Furthermore, in an embodiment, '~parts' may include one or more processors. Hereinafter, a / b may be understood as at least one of a or b.

[0051] In one embodiment of the present disclosure, the term "antenna port" may mean something like "antenna port," "antenna port," "port," or "port," and may be used interchangeably. For example, a "partial" port may mean, but is not limited to, a "partial antenna port," "partial antenna port," or "partial antenna port." The term "estimation" used in the present disclosure may refer to a process of checking a channel state of a specific antenna port. In addition, the term "prediction" used in the present disclosure may refer to a process of deriving a full antenna port from a partial antenna port. However, "estimation" and "prediction" are terms that may be used interchangeably and are not limited to the above definitions. In addition, "channel estimation" in the present disclosure may be used in the same sense as "channel state estimation," and "channel prediction" may be used in the same sense as "channel state prediction."

[0052] In the present disclosure, CSI-RS is used as an example of a reference signal for downlink channel measurement in one embodiment of the present disclosure, and SRS is used as an example of a reference signal for uplink channel measurement. However, the scope of the present invention is not limited thereto, and it can also be applied to reference signals of other communication systems having a similar technical background or channel format to LTE, LTE-A, NR, or 6G.

[0053] While various embodiments of the present disclosure are described below using LTE, LTE-A, NR, or 6G-based systems as examples, the various embodiments of the present disclosure may also be applied to other communication systems with similar technical backgrounds or channel configurations. Furthermore, the various embodiments of the present disclosure may be applied to other communication systems with some modifications within the scope of the present disclosure, as determined by a person skilled in the art.

[0054] Wireless communication technologies have evolved in various ways to provide faster data rates, improved coverage, and more stable connections. In particular, multi-antenna technology has been actively developed at the physical layer to increase the number of transmit and receive antennas and achieve performance gains. Accordingly, the 5G New Radio (NR) system supports up to 32 antenna ports. Therefore, base stations and terminals estimate downlink (DL) and uplink (UL) channels by transmitting and receiving channel state information-reference signals (CSI-RS) and sounding reference signals (SRS) corresponding to up to 32 ports.

[0055] When looking at downlink channel measurement via CSI-RS in a 5G system, the terminal can report to the base station the number of spatial dimensions (rank indicator, RI), the index of the precoding matrix (precoding matrix indicator, PMI), and channel state information (channel quality indicator, CQI) for the estimated channel through the CSI-RS transmitted by the base station. During this CSI reporting process, the terminal utilizes a channel corresponding to the CSI-RS port set by the base station among the number of CSI-RS ports supported by NR. The number of CSI-RS ports supported by NR can be 2, 4, 8, 16, or 32. In a 5G system, the purpose of uplink channel estimation via SRS is mainly for the base station to obtain the downlink channel state through the estimated uplink channel in a time division duplexing (TDD) system. In practice, the uplink transmission power of a terminal is insufficient compared to the downlink transmission power of a base station, and it is difficult to implement adding RF chains to each antenna of the terminal. Therefore, the terminal can transmit by switching the transmit antenna when transmitting SRS. This is called transmit antenna switching (TAS).

[0056] Meanwhile, discussions about the upper-mid band as a 6G frequency are actively underway. Examples of the upper-mid band, which has been actively discussed as a 6G frequency, include 10 GHz and 13 GHz. In general, as the frequency band increases, a greater number of antennas can be integrated into the same area, so it is expected that the upper-mid band will use more antennas than NR. In particular, the next-generation multi-antenna system called extreme MIMO (X-MIMO) is considering up to 256 ports and 2048 antenna elements.

[0057] In the case of channel estimation through CSI-RS discussed above, the overhead is determined by the number of CSI-RS ports, density in the frequency domain, and CSI-RS transmission period in the time domain. Specifically, the more ports there are, the higher the density, and the shorter the transmission period, the greater the overhead. In the X-MIMO system, since the number of ports increases up to 8 times compared to the conventional technology, the CSI-RS overhead is also expected to increase by 8 times. When the transmission period of the CSI-RS is 5, the CSI-RS overhead accounts for 30.48%, which has a significant impact on the reduction of throughput (see [Table 1]). In this case, the overhead can mean the number of REs of the reference signal among the total resource elements (REs). For example, CSI-RS overhead may mean the number of REs occupied by CSI-RS among the total REs, and SRS overhead may mean the number of REs occupied by SRS among the total REs.

[0058] Density = 1CSI-RS periodicity[slot]5102040overhead[%](overhead[%])32-port3.811.900.950.48128-port15.247.623.811.90256-port30.4815.247.623.81

[0059] When setting a long CSI-RS periodicity to reduce CSI-RS overhead, the trade-off can be significant channel aging, which can lead to reduced throughput. Furthermore, in the case of UL channel estimation via SRS, as the number of ports increases, the transmission power at the terminal decreases proportionally, which can lead to reduced system throughput and insufficient coverage due to degraded channel estimation performance.

[0060] The present invention relates to a method for reducing CSI-RS and SRS overhead through spatial domain channel prediction using artificial intelligence (AI). In particular, it aims to define signaling for spatial domain channel prediction between a base station and a terminal. Below, a detailed method is disclosed to address the increasing overhead of CSI-RS or SRS as the number of ports increases.

[0061] FIG. 1 is a diagram illustrating the concept of spatial-domain channel prediction using AI to reduce overhead of CSI-RS applied to various embodiments of the present disclosure.

[0062] Referring to FIG. 1, the base station can select a partial antenna port (120) from the full antenna ports (110) through a CSI-RS port selection (130) procedure. The CSI-RS port selection (130) may be a procedure for selecting a partial antenna port (120) from the full antenna ports (110). The full antenna port (110) is an antenna port from which the base station wishes to receive channel state information, and the partial antenna port (120) may be a partial antenna port of the full antenna ports. In addition, the CSI-RS port selection may be a procedure for selecting a partial antenna port (120) from the full antenna ports (110) using AI. Based on the CSI-RS port selection, the base station can transmit a CSI-RS corresponding to the partial antenna port (120) to the terminal (140). The base station can provide CSI-RS configuration information based on the CSI-RS port selection to the terminal, and transmit a CSI-RS corresponding to the partial antenna port to the terminal based on the CSI-RS configuration information.

[0063] The terminal can receive (or measure) the CSI-RS corresponding to the partial antenna port (120) from the base station and estimate the partial antenna port channel (150) based on the CSI-RS. Estimating the channel may mean estimating the channel state based on the measured reference signal. This is a concept applicable to various embodiments of the present disclosure. The terminal can predict the full antenna port channel (170) based on the estimated partial antenna port channel (150) (160). Predicting the channel may mean predicting the channel state. This is a concept applicable to various embodiments of the present disclosure. AI may be utilized in the procedure (160) in which the terminal predicts the full antenna port channel (170) based on the estimated partial antenna port channel (150).

[0064] As discussed in this disclosure, AI can be utilized solely at the terminal, or at both the terminal and base station. If AI is utilized solely at the terminal, this is referred to as a one-sided AI model. If AI is utilized at both the terminal and base station, this is referred to as a two-sided AI model.

[0065] FIG. 2 is a diagram illustrating an example of AI utilization in a one-sided AI model applied to various embodiments of the present disclosure.

[0066] Referring to FIG. 2, a one-sided AI model may be composed of an input (input, 210), an output (output, 230), a channel predictor (channel predictor, 220), etc. When performing spatial domain channel prediction using a one-sided AI model, selection of a partial antenna port may be made at the base station without using AI. In a one-sided AI model, input 210 may be, for example, a partial antenna port channel (a channel measured based on a partial antenna port) or may be. In this case, the partial antenna port may be a partial antenna port selected without using AI at the base station. In the unidirectional AI model, the output (230) may be obtained from the input (210) through the channel predictor (220). The channel predictor (220) may be an AI trained to predict the entire antenna port channel state from the partial antenna port channel. The output (230) of the unidirectional AI model may be, for example, the entire antenna port channel or It can be. The terminal can transmit a CSI (channel state information) report for the entire antenna port channel to the base station. The CSI report transmitted to the base station can include information about the entire antenna port channel state. For example, the information about the entire antenna port channel state can include at least one of the number of spatial dimensions (rank indicator, RI), the index of the precoding matrix (precoding matrix indicator, PMI), and channel state information (channel quality indicator, CQI).

[0067] FIG. 3 is a diagram illustrating an example of AI utilization in a two-sided AI model applied to various embodiments of the present disclosure.

[0068] Referring to Figure 3, the bidirectional AI model can be divided into the network side and the UE side. After training is complete, the AI ​​model on the network side can be used as a port pattern generator at the base station, and the AI ​​model on the UE side can be used as a spatial domain channel predictor at the UE.

[0069] The bidirectional AI model may be composed of an input (310), a pattern generator (320), a latent vector (330), a channel predictor (340), and an output (350). In the bidirectional AI model, the input (310) may be, for example, all antenna ports. In the bidirectional AI model, the latent vector (330) may be a partial antenna port or a partial antenna port channel. The latent vector (330) of the bidirectional AI model may be obtained from the input (310) through the pattern generator (320). In the bidirectional AI model, the pattern generator (320) may be an AI trained to select a partial antenna port from all antenna ports. The output (350) of the bidirectional AI model may be, for example, all antenna port channels. In addition, the output (350) of the bidirectional AI model may be obtained from the latent vector (330) through the channel predictor (340). The channel predictor (340) of the bidirectional AI model may be an AI trained to appropriately predict the entire antenna port channel from a partial antenna port channel. The terminal may transmit a CSI report for the entire antenna port channel to the base station. The CSI report transmitted to the base station may include information on the entire antenna port channel status obtained from the output (350). The information on the entire antenna port channel status may include at least one of the number of spatial dimensions (rank indicator, RI), the index of the precoding matrix (precoding matrix indicator, PMI), and channel status information (channel quality indicator, CQI).

[0070] When performing spatial domain channel prediction using unidirectional and bidirectional AI models, port patterns between the base station and the terminal may need to be aligned. These port patterns may be based on base station designation, terminal capability information, and other factors. Furthermore, port patterns for the transmit antenna must also be aligned in SRS.

[0071] FIG. 4 is a diagram for explaining the relationship between a base station and a terminal in an AI-based spatial domain channel estimation process according to one embodiment of the present disclosure.

[0072] Referring to FIG. 4, in step 410, the base station (402) may transmit a CSI-RS to the terminal (401). The CSI-RS may be a CSI-RS based on a partial antenna port. The CSI-RS based on the partial antenna port may be referred to as a partial antenna port channel state information reference signal. The base station (402) may select a partial antenna port from all antenna ports. The number of partial antenna ports is smaller than the number of total antenna ports, and some of the total antenna ports may be selected as the partial antenna ports. The full antenna ports are antenna ports from which the base station (402) wishes to receive channel state information, and the partial antenna ports may be some antenna ports of the total antenna ports. The base station (402) may select a partial antenna port from all antenna ports and transmit a CSI-RS corresponding to the partial antenna port to the terminal (401). The base station (402) may or may not use AI in the process of selecting a partial antenna port from all antenna ports. When the base station (402) uses AI in the process of selecting a partial antenna port from all antenna ports, the AI ​​may be, for example, a pattern generator (320), but is not limited thereto. The number of total antenna ports may be the number of total antenna ports used for transmitting a reference signal or a smaller number. For example, when the base station (402) can use N antenna ports for transmitting a reference signal, the number of total antenna ports may be less than or equal to N. The total antenna ports may be the total antenna ports for estimating channel state information. For example, when it is desired to obtain channel state information for M antenna ports among N antenna ports, an antenna port composed of M (M <= N) antenna ports may be the total antenna ports.Partial antenna ports may be some of the full antenna ports for estimating channel state information of all antenna ports. For example, partial antenna ports for estimating channel state information for M full antenna ports may be composed of L (M > L) antenna ports. Full antenna ports may be defined as a first antenna port set, a first antenna port group, etc. In addition, partial antenna ports may be defined as a second antenna port set, a second antenna port group, etc. These various definitions and concepts of full antenna ports and partial antenna ports may be similarly applied not only to the embodiment of FIG. 4 but also to various embodiments of the present disclosure.

[0073] Meanwhile, before step 410, the base station (402) may provide the terminal (401) with information related to CSI-RS. The CSI-RS information may include configuration information for a partial antenna port CSI-RS and resource information of the partial antenna port CSI-RS. In addition, the information related to the CSI-RS may include information on all antenna ports to be predicted through the resource information of the partial antenna port CSI-RS. The information on all antenna ports may include at least one of the number of antenna ports of all antenna ports, CSI-RS resources of all antenna ports, antenna pattern information, or the relationship between the partial antenna ports and all antenna ports. The above information may be provided to the terminal (401) in one control information, or may be provided to the terminal (401) through different control information.

[0074] At step 420, the terminal (401) can predict (or estimate) a channel state. The channel state may be a spatial-domain channel state. The terminal (401) can predict the channel state for all antenna ports based on the channel state measurement for the partial antenna port. The terminal (401) can measure the CSI-RS for the partial antenna port transmitted from the base station (402). The terminal (401) can estimate the channel state for the partial antenna port based on the CSI-RS measurement for the partial antenna port. Thereafter, the terminal (401) can predict the channel state for all antenna ports based on the estimated channel state for the partial antenna port. For example, the terminal (401) can predict the channel state for all M antenna ports based on the channel state obtained based on the CSI-RS measurements for L partial antenna ports. The process of predicting the channel state for all antenna ports can be defined as a spatial-domain channel prediction process. AI can be used in the channel prediction process for all antenna ports. The AI ​​used here may be, for example, a channel predictor (220, 340), but is not limited thereto. Channel state information for all antenna ports may be obtained based on the AI ​​training results and the channel states for partial antenna ports.

[0075] In step 430, the terminal (401) may report channel state information (CSI) to the base station (402). The channel state information may be channel state information for all antenna ports. The channel state information for all antenna ports may be channel state information predicted (or estimated) from the estimated channel states for partial antenna ports. The channel state information for all antenna ports may include at least one of a spatial dimension number (rank indicator, RI), a precoding matrix index (precoding matrix indicator, PMI), and channel state information (channel quality indicator, CQI). According to an embodiment, the terminal (401) may transmit the channel state information for all antenna ports to the base station (402) when there is a request from the base station (402) or when a preset reporting condition is satisfied.

[0076] FIG. 5 is a diagram for explaining a procedure of a terminal in the process of AI-based spatial domain channel estimation according to one embodiment of the present disclosure.

[0077] The content of Fig. 5 may be related to the description of the relationship between the terminal and the base station in the spatial domain channel estimation process of Fig. 4. In other words, the content of Fig. 5 may be related to the content of Fig. 4. Operations or content omitted in Fig. 5 may refer to the corresponding configuration of Fig. 4.

[0078] Referring to FIG. 5, at step 510, the terminal can measure the CSI-RS transmitted from the base station. The CSI-RS may be a CSI-RS corresponding to a partial antenna port. Prior to step 510, the terminal can receive information related to the CSI-RS from the base station. Based on the information related to the CSI-RS, the terminal can perform CSI-RS measurement for the partial antenna port and channel prediction operation for all antenna ports, which will be described later.

[0079] In step 520, the terminal can predict (or estimate) a channel state. The channel state may be a spatial-domain channel state. The terminal can predict the channel state for all antenna ports based on the channel state measurement for the partial antenna port. The terminal can measure the CSI-RS for the partial antenna port transmitted from the base station. The terminal can estimate the channel state for the partial antenna port based on the CSI-RS measurement for the partial antenna port. Thereafter, the terminal can predict the channel state for all antenna ports based on the estimated channel state for the partial antenna port. For example, the terminal can predict the channel state for all M antenna ports based on the channel state obtained based on the CSI-RS measurements for L partial antenna ports. The process of predicting the channel state for all antenna ports can be defined as a spatial-domain channel prediction process. An AI can be used in the channel prediction process for all antenna ports. The AI ​​used at this time can be, for example, a channel predictor (220, 340), but is not limited thereto.

[0080] In step 530, the terminal may report CSI to the base station. The channel state information may be channel state information for all antenna ports. The channel state information for all antenna ports may be predicted channel state information from the estimated channel states for partial antenna ports. The channel state information for all antenna ports may include at least one of a spatial dimension number (rank indicator, RI), a precoding matrix index (precoding matrix indicator, PMI), and channel state information (channel quality indicator, CQI). According to an embodiment, the terminal may transmit the channel state information for all antenna ports to the base station when the base station requests it or when a preset reporting condition is satisfied.

[0081] FIG. 6 is a diagram for explaining the procedure of a base station in the process of AI-based spatial domain channel estimation according to one embodiment of the present disclosure.

[0082] The content of Fig. 6 may be related to the description of the relationship between the terminal and the base station in the spatial domain channel estimation process of Fig. 4. In other words, the content of Fig. 6 may be related to the content of Fig. 4. Operations or content omitted in Fig. 6 may refer to the corresponding configuration of Fig. 4.

[0083] In step 610, the base station may transmit a CSI-RS to the terminal. The CSI-RS may be a CSI-RS based on a partial antenna port. The base station may select a partial antenna port from all antenna ports. The number of partial antenna ports may be smaller than the number of total antenna ports, and some of the partial antenna ports may be selected as partial antenna ports. The base station may select a partial antenna port from all antenna ports and transmit a CSI-RS corresponding to the partial antenna port to the terminal. The base station may or may not use AI in the process of selecting a partial antenna port from all antenna ports. If the base station uses AI in the process of selecting a partial antenna port from all antenna ports, the AI ​​may be, for example, a pattern generator (320), but is not limited thereto. The number of total antenna ports may be the number of total antenna ports used for transmitting a reference signal or a smaller number. For example, if the base station can use N antenna ports for transmitting a reference signal, the number of total antenna ports may be less than or equal to N. The full antenna ports may be the full antenna ports for estimating channel state information. For example, when channel state information for M antenna ports among N antenna ports is to be obtained, an antenna port composed of M (M <= N) antenna ports may be the full antenna ports. The partial antenna ports may be some of the full antenna ports for estimating channel state information for the full antenna ports. For example, the partial antenna ports for estimating channel state information for the M full antenna ports may be composed of L (M > L) antenna ports. The full antenna ports may be defined as a first antenna port set, a first antenna port group, etc. In addition, the partial antenna ports may be defined as a second antenna port set, a second antenna port group, etc.These various definitions and concepts of full antenna ports and partial antenna ports can be similarly applied to various embodiments of the present disclosure as well as the embodiment of FIG. 6.

[0084] Meanwhile, before step 610, the base station may provide (or transmit) information related to CSI-RS to the terminal. The CSI-RS information may include configuration information for a partial antenna port CSI-RS and resource information of the partial antenna port CSI-RS. In addition, the information related to the CSI-RS may include information on all antenna ports to be predicted through the resource information of the partial antenna port CSI-RS. The information on all antenna ports may include at least one of the number of antenna ports of all antenna ports, CSI-RS resources of all antenna ports, antenna pattern information, or the relationship between the partial antenna ports and all antenna ports. The above information may be provided to the terminal in one control information, or may be provided to the terminal through different control information.

[0085] In step 620, the base station can receive channel state information (CSI) from the terminal. The channel state information may be channel state information for all antenna ports. The channel state information for all antenna ports may be channel state information predicted from a channel state estimated for a partial antenna port. In addition, the channel state information for all antenna ports may be obtained based on a channel state for a partial antenna port measured based on measurement of a reference signal for the partial antenna port channel state information. The channel state information for all antenna ports may include at least one of a rank indicator (RI), an index of a precoding matrix (PMI), and channel state information (channel quality indicator (CQI).

[0086] FIG. 7 is an example of a guideline for expressing a port pattern as a bitmap according to one embodiment of the present disclosure, and an example of a port pattern and bitmap expressed according to the guideline.

[0087] 710 is an example of a guideline for representing a port pattern as a bitmap. 701 and 702 are configurations that visualize the port pattern. Configuration 701 can be defined as a first pattern, and configuration 702 can be defined as a second pattern. This is an arbitrary definition of the pattern, and is only used to distinguish the pattern, and the name of the pattern is not limited to this. Drawing reference numerals 711, 712, 713, 714, 715, 716, 717, and 718 are areas where antenna ports can be configured. At least one pattern of the first pattern (701) or the second pattern (702) can be configured in each area, and neither the first pattern (701) nor the second pattern (702) may be configured. The order indicated by the arrows in drawing reference numeral 710 describes the order in which each area is read when representing the port pattern as a bitmap. For example, the port pattern configured in each area can be checked and a bitmap can be configured in the order of 711 -> 712 -> 713 -> 714 -> 715 -> 716 -> 717 -> 718. Meanwhile, an example of such an order is only for explaining a method of configuring a bitmap in one embodiment of the present disclosure, and the method of configuring a bitmap in various embodiments of the present disclosure is not limited thereto, and the order of reading each area may vary.

[0088] According to the 710 guideline, an entity generating a bitmap can generate a bitmap by checking for the presence of a first pattern (701) and / or a second pattern (702) in each area according to a predefined order. If the first pattern (701) is present in each area according to the predefined order, 1 can be written, and if the first pattern (701) is not present, 0 can be written. After checking for the first pattern (701), the presence of the second pattern (702) can be checked again according to a predefined order, and if the second pattern (702) is present in each area, 1 can be written, and if the second pattern (702) is not present, 0 can be written. Meanwhile, in the embodiment of FIG. 7, it has been described that the bitmap is configured by first checking the first pattern (701) and then checking the second pattern (702) according to a predefined order, but this is for the convenience of explanation, and the bitmap configuration method in various embodiments of the present disclosure is not limited thereto. For example, the order in which the areas are read may be changed, the order in which the first pattern (701) and the second pattern (702) are read may be changed, and the bitmap may be configured by proceeding to the next area after checking both the first pattern (701) and the second pattern (702) in each area.

[0089] 720 is an example of expressing port pattern 0 as a bitmap (725) according to the 710 guideline. According to the 710 guideline, if the first pattern (701) exists in the order of 721 -> 722 -> 723 -> 724 areas, 1 can be written, and if the first pattern does not exist, 0 can be written. Next, if the second pattern (702) exists in the order of 721 -> 722 -> 723 -> 724 areas, 1 can be written, and if the second pattern (702) does not exist, 0 can be written. Port pattern 0 (720) has the first pattern (701) in all of the 721, 722, 723, and 724 areas, and the second pattern (702) does not exist in all of the 721, 722, 723, and 724 areas. Accordingly, when port pattern 0 (720) is expressed as a bitmap, 1111 may be described according to the first pattern (701), and 0000 may be described according to the second pattern (702). That is, port pattern 0 (720) may be expressed as a 11110000 bitmap (725).

[0090] 730 is an example of expressing port pattern 1 as a bitmap (735) according to the 710 guideline. According to the 710 guideline, if the first pattern (701) exists in the order of 721 -> 722 -> 723 -> 724 areas, 1 can be written, and if the first pattern does not exist, 0 can be written. Next, if the second pattern (702) exists in the order of 721 -> 722 -> 723 -> 724 areas, 1 can be written, and if the second pattern (702) does not exist, 0 can be written. Port pattern 1 (730) does not have the first pattern (701) in any of the areas 721, 722, 723, and 724, and the second pattern (702) exists in any of the areas 721, 722, 723, and 724. Accordingly, when port pattern 1 (730) is expressed as a bitmap, 0000 may be written according to the first pattern (701), and 1111 may be written according to the second pattern (702). That is, port pattern 1 (730) may be expressed as a 00001111 bitmap (735).

[0091] FIG. 8a, FIG. 8b, FIG. 8c, FIG. 8d, FIG. 8e, FIG. 8f, FIG. 8g, and FIG. 8h are examples of port patterns according to port reduction ratios by number of ports according to one embodiment of the present disclosure.

[0092] Num. of ports in Fig. 8a is the number of ports. Num. of ports can mean the number of ports of the entire antenna port. For example, Num. of ports 32 can mean that the number of ports of the entire antenna port is 32. N1 in Fig. 8a means the size of one side in the virtual antenna port shape, and N2 means the size of the other side in the virtual antenna port shape. The size 1 by 1 of the side means one area where the antenna pattern described in Fig. 7 can be configured. For example, in the case of 8 by 2 configuration as in Fig. 8a, 16 port pattern areas can be configured, and the specific shape refers to the port pattern (port pattern, 816). It can be described as a case where N1 is greater than or equal to N2, but it can also be described as a case where N2 is greater than or equal to N1. The port reduction ratio (813) in Fig. 8a is the number of partial antenna ports relative to the number of total antenna ports. 1 / 2 means that the number of total antenna ports is 32 and the number of partial antenna ports is 16, and when operating according to the embodiment of the present disclosure, channels for 32 total antenna ports can be predicted (estimated) with 16 partial antenna ports. The port pattern index (port pattern index, 814) of FIG. 8A is an index indicating a port pattern (port pattern, 816). Different port patterns can be indicated depending on the index. Depending on the port pattern index, a corresponding port pattern bitmap (port pattern (bitmap), 815) or port pattern (port pattern, 816) can be identified. A corresponding port pattern can be defined depending on the port pattern index (814). The port pattern bitmap (Port pattern (bitmap), 815) of FIG. 8A is a bitmap representation of a port pattern according to the guideline of FIG. 8A.The port pattern of Fig. 8a may be an example of shaping the pattern of a virtual antenna port in a standard. The description of Fig. 8a is also applicable to Figs. 8b, 8c, 8d, 8e, 8f, 8g, and 8h.

[0093] FIG. 9 is a diagram illustrating a CSI-RS-based spatial domain channel prediction method between a terminal and a base station according to one embodiment of the present disclosure.

[0094] Referring to FIG. 9, in step 910, a terminal (901) may transmit capability information to a base station (902). Capability information may be defined as performance information. The terminal (901) may transmit capability information to the base station (902) using a capability report. The capability report may include a UE capability information report. The capability report may include capability information related to spatial domain channel prediction. The terminal (901) may report the UE capability information when there is a report request from the base station (902). The base station (902) may select a partial antenna port from all antenna ports based on the capability report received from the terminal (901). This selection process may be defined as a port selection process. The base station (902) configures partial antenna ports through the port selection process, and the port pattern may vary depending on the type of the configured partial antenna port. In addition, the terminal (901) can perform a spatial domain channel prediction process for a given port pattern to predict the channel of the full antenna port channel state information reference signal (full antenna port CSI-RS). The full antenna port is an antenna port through which the base station (902) wishes to receive channel state information, and the partial antenna port may be a part of the full antenna port.

[0095] The information in the capability report that the terminal (901) transmits to the base station (902) may vary depending on whether the terminal (901) uses its own AI model for spatial domain channel prediction or whether the base station (902) transmits an AI model for spatial domain channel prediction to the terminal (901).

[0096] a) When the terminal (901) uses an AI model for spatial domain channel prediction on its own, the terminal (901) may transmit information on a port pattern that it has trained on its own to the base station (902) during the process of transmitting a capability report to the base station (902) before performing spatial domain channel prediction. The information on the port pattern may include information for identifying the port pattern, and may include, for example, a port pattern index trained by the terminal (901), but is not limited thereto. The information transmitted from the terminal (901) to the base station (902) may be defined as UE-assisted information. In general, the greater the number of measured ports, the easier it is to predict the channel state of all antenna ports, and thus channel prediction is possible with a low AI model complexity (e.g., the number of model parameters, the number of FLOPs). In an embodiment of the present disclosure, a partial antenna port may be a measured port. The complexity of an AI model may be related to the number of model parameters and / or FLOPs (floating point operations). Typically, each terminal has different AI-related complexity capabilities, so the minimum number of measured ports required for full antenna port channel prediction may vary. Because the minimum required measured ports, i.e., partial antenna ports, vary, the port patterns internally trained by the terminal (901) may differ. Therefore, when using a terminal-trained AI model, a procedure for reporting the port patterns internally trained by the terminal (901) may be required. The format for reporting port patterns may vary depending on the presence or absence of a standard definition.

[0097] For example, if a port pattern table such as FIG. 8a, FIG. 8b, FIG. 8c, FIG. 8d, FIG. 8e, FIG. 8f, FIG. 8g, or FIG. 8h is defined in the standard, the terminal (901) can transmit to the base station (902) the indexes of the port patterns trained by the terminal (901) for each CSI-RS port. The terminal (901) can transmit to the base station (902) the indexes of the port patterns trained by the terminal (901) in the number of ports configured.

[0098] The spatial domain channel prediction performance is greatly influenced by the port pattern, which may vary depending on the site's channel environment (e.g., scatter, reflector, etc.) and the vendor's antenna configuration implementation. Therefore, if there is a high degree of freedom in the port pattern design, the vendor can operate the port pattern in a site-specific or site-common manner. In this case, the terminal (901) can transmit a terminal-specific port pattern (UE-specific port pattern) to the base station (902). The UE-specific port pattern can be interpreted as vendor-specific or site-specific. When using a UE-specific port pattern, the terminal (901) can report the port pattern in the form of a bitmap, as in the example of FIG. 7.

[0099] b) When the base station (902) transfers an AI model for spatial domain channel prediction to the terminal (901), the terminal (901) can perform spatial domain channel prediction using the AI ​​model transferred from the base station (902). This performing method can be defined as performing channel prediction in the form of AI model inference based on model transfer. When performing channel prediction in the form of AI model inference based on model transfer, the AI ​​model itself is held by the base station (902). Therefore, the terminal (901) can report capability information for operating the AI ​​model to the base station (902). The capability information transmitted by the terminal (901) may be AI complexity capability information. The AI ​​complexity capability information of the terminal (901) may include at least one or more of maximum allowable memory, storage, and computation capability information. For example, the terminal (901) can transmit to the base station (902) the allowable complexity for each terminal from the categorized AI-related terminal capability information. The base station (902) can determine an AI model suitable for the capability information received from the terminal (901) and transmit the determined AI model to the terminal (901). The terminal (901) can report a port reduction ratio based on its capability information.

[0100] In step 920, the base station (902) may transmit information about a port pattern to the terminal (901). The information about the port pattern may include information about at least one port pattern. The information about the port pattern may be information related to a port pattern candidate group. The information about the port pattern candidate group may include a plurality of port patterns determined based on channel prediction-related capability information of the terminal (901). The information about the port pattern may be defined as candidate port pattern configuration information, but is not limited thereto. The candidate port pattern configuration information may be transmitted to the terminal (901) via an RRC (radio resource control) message. This process may be defined as a process in which the terminal (901) receives information about the candidate port pattern group via an RRC message. When configuring the port pattern candidate group, the base station (902) may consider the terminal capability information of step 910. Additionally, when setting up a port pattern candidate group, the base station (902) may use a standard defined table or a UE specific (vendor-specific or site-specific) port pattern.

[0101] When the base station (902) configures among the standard defined port patterns, the base station (902) may have received the index of the port pattern from the terminal (901) in step 910. Accordingly, the base station (902) may set the index of patterns to be selectively used among the entire port patterns and transmit the configuration information to the terminal (901). If all port patterns corresponding to a specific number of ports are to be used, the index corresponding to the use of the entire pattern may be transmitted to the terminal (901). For example, when only the number of ports is indicated and not the index of a specific port pattern, it can be understood that the terminal (901) and the base station (902) use the entire port pattern that supports the corresponding number of ports.

[0102] The spatial domain channel prediction performance is greatly affected by the port pattern, which may vary depending on the site's channel environment (e.g., scatter, reflector, etc.) and the vendor's antenna configuration implementation. Therefore, if there is a high degree of freedom in the port pattern design, the vendor and / or site can operate the port pattern in a site-specific or site-common manner. If each vendor and / or site uses its own specific pattern, the base station (902) can set the port pattern it wants to operate in the form of a bitmap and transmit the port pattern candidate configuration information to the terminal (901). The base station (902) can transmit to the terminal (901) not only the candidate port pattern but also an AI model trained with the candidate port pattern. The base station (902) may provide the terminal (901) with information about the AI ​​model in advance, or may provide the terminal (901) with information about the AI ​​model together with the port pattern candidate configuration.

[0103] In step 930, the base station (902) may transmit port pattern configuration-related information to the terminal (901). The port pattern configuration-related information may be information indicating a port pattern to be measured by the terminal (901) among a plurality of port patterns. The port pattern-related information may be transmitted to the terminal (901) in the form of downlink control information (DCI), medium access control (MAC-CE), etc. This process may be defined as the terminal (901) receiving port pattern configuration-related information as DCI or MAC-CE. For example, among a plurality of port patterns configured based on the port pattern configuration information, a port pattern to be measured by the terminal (901) may be indicated through the port pattern configuration-related information. The terminal (901) may check the port pattern according to the configured port pattern and perform AI-related inference. In step 930, one or more port patterns may be indicated. When using one port pattern, the corresponding index can be set, and when using multiple port patterns, port pattern configuration can be performed by setting all of the corresponding indices. Multiple port patterns can also be specified by specifying a specific group, a specific number of ports, or a specific ratio.

[0104] When setting a port pattern, additional parameters can be set. That is, when setting multiple port patterns, the port patterns can be used in a mixed manner based on the CSI-RS transmission cycle and / or the performance monitoring results of the base station (902). When using a mixed manner of port patterns, the terminal (901) can use the port channel measured before the port pattern change to predict the new port pattern. Therefore, using a mixed manner of port patterns by additionally setting parameters can achieve performance gains in terms of implementation.

[0105] For example, if different port patterns are used for odd-numbered and even-numbered CSI-RS transmissions, the situation may be as shown in [Table 2] below. The numbers in the second row of [Table 2] indicate the port pattern index. (1: Port pattern 1, 2: Port pattern 2)

[0106] First->Second->Third->Fourth->Fifth->Sixth...121212...

[0107] Another example is when each port pattern has a different impact on AI inference performance. A dominant pattern, which has a significant impact on performance, may be used more frequently than other patterns, as shown in Table 3. The numbers in the second row of Table 3 represent the port pattern numbers (1: port pattern 1, 2: port pattern 2).

[0108] [Table 3] is an example of a case where port pattern 1 is a dominant pattern that has a greater impact on performance than port pattern 2.

[0109] First->Second->Third->Fourth->Fifth->Sixth...111121...

[0110] The base station (902) can set the port pattern to change when a specific event occurs. In addition, the base station (902) can set the index of the port pattern to change when a specific event occurs. For example, the base station (902) can set the port pattern to change when a performance metric deteriorates during AI / ML-related life-cycle-management (LCM) execution, and can set the related port pattern index. As another example, when the base station (902) monitors cell throughput as a performance metric, the base station (902) can set the port pattern to change according to a decrease in throughput, and can set the related port pattern index. As yet another example, the base station (902) can set the port pattern to change when the performance of the mean squared error (MSE) between the predicted (or reconstructed) full antenna port channel and the ground-truth full antenna port channel exceeds a specific threshold value, and can set the related port pattern index. As another example, when changing the port pattern for each specific location range within a cell (e.g., zone 1, zone 2, ...), the port pattern can be set to change each time the location of the terminal (901) goes out of the specific location range (e.g., zone 1-> zone 2), and the related port pattern index can be set.

[0111] At step 940, the base station (902) can transmit a CSI-RS to the terminal (901). The base station (902) can transmit the CSI-RS to the terminal (901) based on a port pattern configuration. The base station (902) can transmit the CSI-RS to the terminal (901) based on CSI-RS configuration-related information and the port pattern configuration. The CSI-RS may be a CSI-RS based on a partial antenna port. The CSI-RS based on the partial antenna port may be referred to as a partial antenna port channel state information reference signal. The relationship between the full antenna port and the partial antenna port is described in the embodiment of FIG. 4.

[0112] At step 950, the terminal (901) can predict (or estimate) a channel state. The channel state may be a spatial-domain channel state. The terminal (901) can predict the channel state for all antenna ports based on the channel state measurement for the partial antenna port. The terminal (901) can measure the CSI-RS for the partial antenna port transmitted from the base station (902). The terminal (901) can estimate the channel state for the partial antenna port based on the CSI-RS measurement for the partial antenna port. Thereafter, the terminal (901) can predict the channel state for all antenna ports based on the estimated channel state for the partial antenna port. For example, the terminal (901) can predict the channel state for all M antenna ports based on the channel state obtained based on the CSI-RS measurements for L partial antenna ports. The process of predicting the channel state for all antenna ports can be defined as a spatial-domain channel prediction process. AI can be used in the channel prediction process for all antenna ports. The AI ​​used at this time may be, for example, a channel predictor (220, 340), but is not limited thereto. Channel state information for all antenna ports may be obtained based on the results of AI training performed on a group of port pattern candidates and the channel state for a subset of antenna ports.

[0113] In step 960, the terminal (901) can report (transmit) channel state information (CSI) to the base station (902). The channel state information may be channel state information for all antenna ports. The channel state information for all antenna ports may be channel state information predicted (or acquired) from a channel state estimated for a partial antenna port. In addition, the channel state information for all antenna ports may be acquired based on a channel state for a partial antenna port measured based on measurement of a reference signal for the partial antenna port channel state information. The channel state information for all antenna ports may include at least one of a rank indicator (RI), an index of a precoding matrix (PMI), and channel state information (channel quality indicator (CQI).

[0114] At step 970, the terminal (901) can transmit a message requesting port pattern switching to the base station (902). The terminal (901) can measure the entire antenna port CSI-RS transmitted periodically from the base station (902). The channel identified based on the measured entire antenna port CSI-RS is called a ground-truth channel, and the entire antenna port channel predicted from the partial antenna port channel can be called a restored entire antenna port channel. Channel measurement based on the entire antenna port CSI-RS can be performed periodically, or can be performed aperiodically or periodically according to the request of the terminal (901).

[0115] In step 970, the terminal (901) can transmit a request signal requesting port pattern transition to the base station (902) when a specific event occurs. The request signal transmission step of the terminal (901) is an optional step and can be performed as needed. Since the terminal (901) is the entity that can actually measure the raw channel, there is an advantage in measuring the prediction performance at the terminal (901). The terminal (901) typically measures the partial antenna port channel state information reference signal (partial antenna port CSI-RS) transmitted from the base station (902) and predicts the entire antenna port channel based on this. In addition, the terminal (901) can measure the entire antenna port CSI-RS transmitted by the base station (902) at a specific period. When measuring the entire antenna port CSI-RS, the terminal (901) can measure the prediction performance through the AI ​​performance metric. AI performance metrics may include the square of the generalized cosine similarity (SGCS) or the mean squared error (MSE). The SGCS or MSE can be used to set a threshold for prediction performance in advance, and the threshold of the SGC or MSE can be defined as a preset standard. Additionally, the terminal (901) can determine the channel relationship in the form of a channel covariance matrix each time it measures all antenna port CSI-RSs, thereby enabling performance comparison and analysis.

[0116] For example, the terminal (901) may request the base station (902) to use only port patterns that satisfy an MSE lower than or equal to a certain threshold value by comparing the MSE between the restored entire antenna port channel and the ground-truth channel for port patterns that are being used at a certain interval. As another example, the terminal (901) may request the base station (902) to use only port patterns that satisfy an SGCS lower than or equal to a certain threshold value by comparing the SGCS between the restored entire antenna port channel and the ground-truth channel for port patterns that are being used at a certain interval. An MSE or SGCS lower than or equal to a certain threshold value may be defined as a preset criterion. In addition, step 970 may be defined as a process in which the terminal (901) transmits to the base station (902) a message requesting the use of a port pattern that satisfies a preset criterion.

[0117] Meanwhile, step 970 may be performed before step 960, or may be performed after step 960.

[0118] FIG. 10 is a diagram illustrating a CSI-RS-based spatial domain channel prediction method of a terminal according to one embodiment of the present disclosure.

[0119] The content of Fig. 10 may be related to the description of the mutual relationship between the terminal and the base station in the spatial domain channel estimation process of Fig. 9. That is, the content of Fig. 10 may be related to the content of Fig. 9. Operations or content omitted in Fig. 10 may refer to the corresponding configuration of Fig. 9.

[0120] Referring to Figure 10, at step 1010, a terminal may transmit capability information to a base station. Capability information may be defined as performance information. The terminal may transmit capability information to the base station using a capability report. The capability report may include a UE capability information report. The capability report may include capability information related to spatial domain channel prediction. UE capability information may be reported upon request from the base station.

[0121] a) When a terminal uses an AI model for spatial domain channel prediction on its own, the terminal may transmit information about a port pattern trained on its own to the base station during the process of transmitting a capability report to the base station before performing spatial domain channel prediction. The information about the port pattern may include information for identifying the port pattern, and may include, for example, a port pattern index trained by the terminal, but is not limited thereto. The information transmitted from the terminal to the base station may be referred to as UE-assisted information. In general, the greater the number of measured ports, the easier it is to predict a full port channel, and thus channel prediction is possible with a low AI model complexity (e.g., the number of model parameters, the number of FLOPs). In an embodiment of the present disclosure, a partial antenna port may be a measured port. The AI ​​model complexity may be related to the number of model parameters and / or FLOPs (floating point operations). Typically, since AI-related complexity capability information is different for each terminal, the minimum number of measured ports required for full port channel prediction may vary. Because the minimum required measured port, or partial antenna port, varies, the terminal's internally trained port pattern may differ. Therefore, when using a terminal-trained AI model, a procedure for reporting the terminal's internally trained port pattern may be required. The format for reporting port patterns varies depending on the presence or absence of a standard definition.

[0122] For example, if a port pattern table such as FIG. 8a, FIG. 8b, FIG. 8c, FIG. 8d, FIG. 8e, FIG. 8f, FIG. 8g, or FIG. 8h is defined in the standard, the terminal can transmit to the base station the indexes of the port patterns trained by the terminal for each CSI-RS port. The terminal can transmit to the base station the indexes of the port patterns trained by the terminal in the number of ports configured by the base station.

[0123] b) When the base station transmits an AI model for spatial domain channel prediction to the terminal, the terminal can perform spatial domain channel prediction using the AI ​​model received from the base station. This method can be defined as prediction performance in the form of AI model inference based on model transfer. When prediction performance in the form of AI model inference based on model transfer is performed, the AI ​​model itself may be held by the base station. Therefore, the terminal can report capability information for operating the AI ​​model to the base station. The capability information transmitted by the terminal may be AI complexity capability information. The AI ​​complexity capability information of the terminal may include at least one or more of maximum allowable memory, storage, and computation capability information. For example, the terminal can transmit the allowable complexity of each terminal to the base station from categorized AI-related terminal capability information. The base station can determine an AI model appropriate for the capability information received from the terminal and transmit the determined AI model to the terminal. The terminal can report a port reduction ratio based on its capability information.

[0124] In step 1020, the terminal may receive information about a port pattern from the base station. The information about the port pattern may include information about at least one port pattern. The information about the port pattern may be defined as a candidate port pattern configuration, but is not limited thereto. The terminal may receive the candidate port pattern configuration information via an RRC (radio resource control) message.

[0125] In step 1030, the terminal can receive port pattern configuration related information from the base station. The port pattern configuration related information can be received by the terminal in the form of DCI, MAC CE, etc. For example, among multiple port patterns configured based on the port pattern candidate group configuration, the port pattern to be measured by the terminal can be indicated through the port pattern configuration related information. In step 1030, one or multiple port patterns can be indicated. When using one port pattern, the corresponding index may have been set, and when using multiple port patterns, the port pattern configuration may have been performed by setting all of the corresponding indices.

[0126] When a port pattern is configured, additional parameters can be configured. Specifically, when multiple port patterns are configured, the port patterns can be used in a mixed manner based on the CSI-RS transmission cycle and / or the performance monitoring results of the base station. When port patterns are used in a mixed manner, the terminal can use the port channels measured before the port pattern change to predict the new port pattern. Therefore, by additionally setting parameters and using a mixed manner of port patterns, performance gains can be achieved from an implementation perspective.

[0127] For example, different port patterns may be used for odd-numbered and even-numbered CSI-RS transmissions. Another example is when each port pattern has a different impact on AI inference performance. A dominant pattern with a greater impact on performance may be used more frequently than other patterns.

[0128] In step 1040, the terminal can receive a CSI-RS from the base station. The terminal can receive the CSI-RS from the base station based on the port pattern configuration. The terminal can receive the CSI-RS from the base station based on the CSI-RS configuration-related information and the port pattern configuration. The CSI-RS may be a CSI-RS based on a partial antenna port. The relationship between the full antenna port and the partial antenna port is described in the embodiment of FIG. 4.

[0129] At step 1050, the terminal can predict (or estimate) a channel state. The channel state may be a spatial-domain channel state. The terminal can predict the channel state for all antenna ports based on the channel state measurement for the partial antenna port. The terminal can measure the CSI-RS for the partial antenna port transmitted from the base station. The terminal can estimate the channel state for the partial antenna port based on the CSI-RS measurement for the partial antenna port. Thereafter, the terminal can predict the channel state for all antenna ports based on the estimated channel state for the partial antenna port. For example, the terminal can predict the channel state for all M antenna ports based on the channel state obtained based on the CSI-RS measurements for L partial antenna ports. The process of predicting the channel state for all antenna ports can be defined as a spatial-domain channel prediction process. An AI can be used in the channel prediction process for all antenna ports. The AI ​​used at this time can be, for example, a channel predictor (220, 340), but is not limited thereto.

[0130] In step 1060, the terminal may report channel state information (CSI) to the base station. The channel state information may be channel state information for all antenna ports. The channel state information for all antenna ports may be predicted channel state information from the estimated channel states for partial antenna ports. The channel state information for all antenna ports may include at least one of a spatial dimension number (rank indicator, RI), a precoding matrix index (precoding matrix indicator, PMI), and channel state information (channel quality indicator, CQI). According to an embodiment, the terminal may transmit the channel state information for all antenna ports to the base station when the base station requests it or when a preset reporting condition is satisfied.

[0131] In step 1070, the terminal can transmit a message requesting a port pattern switch to the base station. The channel measured based on the received full antenna port CSI-RS is called the ground-truth channel, and the full antenna port channel predicted from the partial antenna port channel is called the restored full antenna port channel. Channel measurements based on the full antenna port CSI-RS can be performed periodically, or aperiodic or periodic at the terminal's request. Step 1070 may be performed before step 1060 or after step 1060.

[0132] In step 1070, the terminal can transmit a request signal to the base station requesting a port pattern transition when a specific event occurs. The request signal transmission step by the terminal is an optional step and can be performed as needed. Since the terminal is the entity that can actually measure the raw channel, it has an advantage in measuring prediction performance at the terminal. The terminal typically measures the partial antenna port channel state information reference signal (partial antenna port CSI-RS) transmitted from the base station and predicts the full antenna port channel based on this. In addition, the terminal can measure all antenna port CSI-RS transmitted by the base station at a specific interval. When measuring all antenna port CSI-RS, the terminal can measure prediction performance using AI performance metrics. AI performance metrics can include the square of the generalized cosine similarity (SGCS) or the mean squared error (MSE). Additionally, the terminal can obtain the channel relationship in the form of a channel variance matrix whenever measuring all antenna port CSI-RS, which also enables performance comparison analysis.

[0133] For example, for port patterns that a terminal is using at a regular interval, the MSE between the restored entire antenna port channel and the ground-truth channel can be compared and a request can be made to the base station to use only port patterns that satisfy an MSE below a certain threshold value. As another example, for port patterns that a terminal is using at a regular interval, the SGCS between the restored entire antenna port channel and the ground-truth channel can be compared and a request can be made to the base station to use only port patterns that satisfy an SGCS below a certain threshold value.

[0134] FIG. 11 is a diagram illustrating a CSI-RS-based spatial domain channel prediction method of a base station according to one embodiment of the present disclosure.

[0135] The content of Fig. 11 may be related to the description of the relationship between the terminal and the base station in the spatial-domain channel estimation process of Fig. 9. In other words, the content of Fig. 11 may be related to the content of Fig. 9. Operations or content omitted in Fig. 11 may refer to the corresponding configuration of Fig. 9.

[0136] Referring to FIG. 11, at step 1110, a base station may receive capability information from a terminal. Capability information may be defined as performance information. The base station may receive capability information from the terminal through a capability report. The capability report may include a UE capability information report. The capability report may include capability information related to spatial domain channel prediction. The base station may request the terminal to report the UE capability information. Based on the capability report received from the terminal, the base station may select a partial antenna port from all antenna ports. This selection process may be defined as a port selection process. The base station configures partial antenna ports through the port selection process, and the port pattern may vary depending on the type of the configured partial antenna port.

[0137] The information in the capability report received by the base station from the terminal may vary depending on whether the terminal uses its own AI model for spatial domain channel prediction or whether the base station transmits the AI ​​model for spatial domain channel prediction to the terminal.

[0138] a) When a terminal uses its own AI model for spatial domain channel prediction, the base station can receive a port pattern trained by itself from the terminal. In this case, the information received by the base station from the terminal can be referred to as UE-assisted information. In general, the greater the number of measured ports, the easier it is to predict the full antenna port channel state, and thus channel prediction is possible with a low AI model complexity (e.g., number of model parameters, number of FLOPs). In the present invention, a partial antenna port may be a measured port. The AI ​​model complexity may be related to the number of model parameters and / or FLOPs (floating point operations). Typically, since the AI-related complexity capability information is different for each terminal, the minimum number of measured ports required for full port channel prediction may vary. Since the minimum required measured port, i.e., the partial antenna port, is different, the port pattern trained internally by the terminal is different. Therefore, when using a terminal-trained AI model, a procedure may be required for the base station to receive port patterns internally trained by the terminal from the terminal. The format for reporting port patterns varies depending on the presence or absence of a standard definition.

[0139] For example, if a port pattern table such as FIG. 8a, FIG. 8b, FIG. 8c, FIG. 8d, FIG. 8e, FIG. 8f, FIG. 8g, or FIG. 8h is defined in the standard, the base station can receive from the terminal the index of the port patterns trained by the terminal for each CSI-RS port. The base station can receive from the terminal the index of the port patterns trained by the terminal in the number of ports configured.

[0140] b) When the base station transmits an AI model for spatial domain channel prediction to the terminal, the terminal can perform spatial domain channel prediction using the AI ​​model received from the base station. This method of performing the prediction can be defined as a prediction performance in the form of AI model inference based on model transfer. When the prediction performance in the form of AI model inference based on model transfer is performed, the AI ​​model itself may be owned by the base station. Therefore, the base station can receive a capability report for operating the AI ​​from the terminal. The capability report that the base station receives from the terminal may be an AI complexity capability report. The AI ​​complexity capability information received by the base station from the terminal may include at least one or more of maximum allowable memory, storage space, and computational capability information. For example, the base station can receive the allowable complexity for each terminal from the categorized AI-related terminal capability information from the terminal. The base station can determine an AI model suitable for the capability information received from the terminal and transmit the determined AI model to the terminal.

[0141] In step 1120, the base station may transmit port pattern information to the terminal. The port pattern information may include information about at least one port pattern. The port pattern information may be defined as a candidate port pattern configuration, but is not limited thereto. The port pattern candidate configuration information may be transmitted to the terminal via an RRC (radio resource control) message. When configuring the port pattern candidate, the base station may consider the terminal capability information of step 910. Furthermore, when configuring the port pattern candidate, the base station may use a standardized table or a UE-specific (vendor-specific or site-specific) port pattern.

[0142] When configuring a standard defined port pattern, the base station may have received the port pattern index from the terminal in step 1110. Therefore, the base station can transmit configuration information to the terminal by setting the indices of patterns to be selectively used among the entire port pattern. If all port patterns corresponding to a specific number of ports are to be used, the index corresponding to the use of the entire pattern can be transmitted. For example, if only the number of ports is indicated, rather than the index of a specific port pattern, it can be understood that the terminal and base station are using the entire port pattern that supports that number of ports.

[0143] Spatial domain channel prediction performance is significantly affected by port patterns, which can vary depending on the site's channel environment (e.g., scatter, reflector, etc.) and the vendor's antenna configuration implementation. Therefore, if there is a high degree of freedom in port pattern design, vendors and / or sites can operate port patterns in a site-specific or site-common manner. If each vendor and / or site uses its own specific pattern, the base station can set the port pattern it wants to operate in the form of a bitmap and transmit the port pattern candidate configuration information to the terminal. The base station can transmit to the terminal not only the candidate port pattern but also an AI model trained with the candidate port pattern. The base station can provide the terminal with information about the AI ​​model in advance, or provide the terminal with information about the AI ​​model along with the port pattern candidate configuration.

[0144] In step 1130, the base station can transmit port pattern configuration-related information to the terminal. The port pattern configuration-related information can be transmitted to the terminal in the form of DCI, MAC CE, etc. For example, among multiple port patterns configured based on candidate port pattern configuration, a port pattern to be measured by the terminal can be indicated through the port pattern configuration-related information. The terminal can check the port pattern according to the configured port pattern and perform AI-related inference. In step 930, one or multiple port patterns can be indicated. When using one port pattern, the corresponding index can be set, and when using multiple port patterns, port pattern configuration can be performed by setting all of the corresponding indices. Multiple port patterns can be indicated by a method of indicating a specific group, a specific number of ports, or a specific ratio.

[0145] When configuring a port pattern, additional parameters can be configured. That is, when configuring multiple port patterns, the port patterns can be used in a mixed manner based on the CSI-RS transmission cycle and / or the performance monitoring results of the base station.

[0146] The base station can configure the port pattern to change when a specific event occurs. The base station can also set the index of the port pattern that changes when a specific event occurs. For example, the base station can configure the port pattern to change and set the related port pattern index when the performance metric deteriorates during AI / ML-related life-cycle-management (LCM) execution. In another example, when the base station monitors cell throughput as a performance metric, it can configure the port pattern to change according to the throughput decrease and set the related port pattern index. In another example, the base station can configure the port pattern to change and set the related port pattern index when the mean squared error (MSE) performance between the predicted (or reconstructed) full antenna port channel and the ground-truth full antenna port channel exceeds a specific threshold value. As another example, when a base station changes the port pattern for each specific location range within a cell (e.g., zone 1, zone 2, ...), it can set a port pattern change and set a related port pattern index whenever the terminal's location changes outside the specific location range (e.g., zone 1-> zone 2).

[0147] At step 1140, the base station can transmit a CSI-RS to the terminal. The base station can transmit the CSI-RS to the terminal based on the port pattern configuration. The base station can transmit the CSI-RS to the terminal based on CSI-RS configuration-related information and the port pattern configuration. The CSI-RS may be a CSI-RS based on a partial antenna port. The relationship between the full antenna port and the partial antenna port is described in the embodiment of FIG. 4.

[0148] At step 1150, the base station may receive channel state information (CSI) from the terminal. The CSI may be channel state information for all antenna ports. The CSI for all antenna ports may be predicted channel state information based on the estimated channel states for partial antenna ports. The CSI for all antenna ports may include at least one of a spatial dimension number (rank indicator, RI), a precoding matrix index (precoding matrix indicator, PMI), and channel state information (channel quality indicator, CQI).

[0149] In step 1160, the base station may receive a message requesting a port pattern switch from the terminal. The channel measured based on the received full antenna port CSI-RS is called the ground-truth channel, and the full antenna port channel predicted from the partial antenna port channel is called the restored full antenna port channel. Channel measurements based on the full antenna port CSI-RS may be performed periodically, or aperiodic or periodic at the terminal's request. Step 1160 may be performed before step 1150 or after step 1150.

[0150] At step 1160, the base station may receive a request signal from the terminal requesting a port pattern transition. For example, the base station may be requested by the terminal to use only port patterns that satisfy an MSE below a certain threshold value. In another example, the base station may be requested by the terminal to use only port patterns that satisfy an SGCS below a certain threshold value.

[0151] FIG. 12 is a diagram illustrating a transmit antenna switching (TAS) technology related to SRS according to one embodiment of the present disclosure.

[0152] Although the present disclosure describes an embodiment based on a case where the number of RX ports is 4, the concept can be applied in the same or similar manner even when the number of RX ports is 4, 8, 16, 32, 64, 128, 256 or more.

[0153] In the case of a sounding reference signal (SRS), the SRS is transmitted by switching the tx port (1205) within the terminal. According to one embodiment of the present invention, the base station can measure a sounding reference signal (SRS) corresponding to a partial rx port among rx port #1 (1210), rx port #2 (1220), rx port #3 (1230), and rx port #4 (1240). Thereafter, based on the sounding reference signal corresponding to the measured partial rx port, the channel state of the entire rx port can be predicted.

[0154] FIG. 13 is a diagram illustrating an SRS-based spatial domain channel prediction method between a terminal and a base station according to one embodiment of the present disclosure.

[0155] Referring to FIG. 13, in step 1310, the base station (1302) may transmit information about a port pattern to the terminal (1301). The information about the port pattern may include information about at least one port pattern. The information about the port pattern may be defined as candidate port pattern configuration information, but is not limited thereto. The candidate port pattern configuration information may be transmitted to the terminal (1301) via an RRC message. When configuring the candidate port pattern, the base station (1302) may use a standard defined table or a UE-specific (vendor-specific or site-specific) port pattern. [Table 4] below is an example of an SRS-related port pattern according to a port reduction ratio by the number of ports according to an embodiment of the present disclosure. If a table such as [Table 4] is defined in the standard, the base station (1302) may use [Table 4] when configuring the candidate port pattern.

[0156] Num. of ports1 (port reduction ratio = 1 / 4)2 (port reduction ratio = 1 / 2)3 (port reduction ratio = 3 / 4)4Port pattern{1}{2}{3}{4}{1, 2}{1, 3}{1, 4}{2, 3}{2, 4}{3, 4}{1, 2, 3}{1, 2, 4}{2, 3, 4}{1, 2, 3, 4}

[0157] When setting among the standard defined port patterns, the base station (1302) can set the indexes of patterns to be selectively used among the entire port patterns and transmit the setting information to the terminal (1301). If all port patterns corresponding to a specific number of ports are to be used, the base station (1302) can transmit the index corresponding to the use of the entire pattern to the terminal (1301). For example, when only the number of ports is indicated and not the index of a specific port pattern, it can be understood that the terminal (1301) and the base station (1302) use the entire port pattern that supports the corresponding number of ports.

[0158] Spatial domain channel prediction performance is significantly influenced by port patterns, which can vary depending on the site's channel environment (e.g., scattering, reflection, etc.) and the vendor's antenna configuration implementation. Therefore, if there is a high degree of freedom in port pattern design, vendors and / or sites can operate port patterns site-specifically or site-commonly. If each vendor and / or site independently uses a specific pattern, the base station (1302) can set the port pattern it wants to operate in bitmap form and transmit port pattern candidate configuration information to the terminal (1301).

[0159] At step 1320, the base station (1302) can transmit port pattern configuration-related information to the terminal (1301). The port pattern configuration-related information can be transmitted to the terminal (1301) in the form of DCI (downlink control information), MAC CE (medium access control - control element), etc. When using one port pattern, the corresponding index can be set, and when using multiple port patterns, the port pattern configuration can be performed by setting all of the corresponding indices. In addition, multiple port patterns can be indicated by a method of indicating a specific group or a method of indicating a specific number of ports or a specific ratio.

[0160] When the base station (1302) configures a port pattern for the terminal (1301), additional parameters can be configured. That is, when configuring multiple port patterns, the port patterns can be used in a mixed manner based on the SRS transmission cycle and / or the performance monitoring results of the base station (1302). When using a mixed manner of port patterns, the base station (1302) can utilize the port channels measured before the port pattern change to predict the new port pattern. Therefore, using a mixed manner of port patterns by additionally setting parameters can achieve performance gains in terms of implementation.

[0161] For example, if different port patterns are used for odd-numbered and even-numbered SRS transmissions, the results may be as shown in [Table 5]. The numbers in the second row of [Table 5] indicate the port pattern index. (1: Port pattern 1, 2: Port pattern 2)

[0162] First->Second->Third->Fourth->Fifth->Sixth...121212...

[0163] Another example is when each port pattern has a different impact on AI inference performance. A dominant pattern, which has a significant impact on performance, may be used more frequently than other patterns, as shown in Table 6. The numbers in the second row of Table 6 represent the port pattern numbers (1: port pattern 1, 2: port pattern 2).

[0164] [Table 6] is an example of a case where port pattern 1 is a dominant pattern that has a greater impact on performance than port pattern 2.

[0165] First->Second->Third->Fourth->Fifth->Sixth...111121...

[0166] The base station (1302) can be configured to change the port pattern when a specific event occurs. In addition, the base station (1302) can set the index of the port pattern that changes when a specific event occurs. For example, the base station (1302) can be configured to change the port pattern when a performance metric deteriorates during AI / ML-related life-cycle-management (LCM) execution, and can set the related port pattern index. In another example, when the base station (1302) monitors cell throughput using a performance metric, the base station (1302) can be configured to change the port pattern according to a decrease in throughput, and can set the related port pattern index. In yet another example, the base station (1302) can be configured to change the port pattern when the MSE performance between the predicted (or restored) full port channel and the ground-truth channel exceeds a specific threshold value, and can set the related port pattern index. As another example, when the base station (1302) changes the port pattern for each specific location range within the cell (e.g., zone 1, zone 2, ...), it can set a port pattern change every time the location of the terminal changes outside the specific location range (e.g., zone 1-> zone 2), and can set a related port pattern index.

[0167] At step 1330, the terminal (1301) can transmit an SRS to the base station (1302). The terminal (1301) can transmit the SRS to the base station (1302) based on port pattern settings. The terminal (1301) can transmit the SRS to the base station (1302) based on SRS setting related information and port pattern settings. The SRS may be an SRS based on a partial rx port.

[0168] At step 1340, the base station (1302) can predict (or estimate) a channel state. The channel state may be a space-domain channel state. The base station (1302) can predict the channel state for the full rx port based on the channel state measurement for the partial rx port. The base station (1302) can measure the SRS for the partial rx port transmitted from the terminal (1301). The base station (1302) can estimate the channel state for the partial rx port based on the SRS measurement for the partial rx port. Thereafter, the base station (1302) can predict the channel state for the full rx port based on the estimated channel state for the partial rx port. For example, the base station (1302) can predict the channel state for the M full rx ports based on the channel state obtained based on the SRS measurements for the L partial rx ports. The process of predicting the overall RX port channel state can be defined as a spatial domain channel prediction process. AI can be used in this process. The AI ​​used here can be, for example, channel predictor (220, 340), but is not limited thereto.

[0169] FIG. 14 is a diagram illustrating a method for predicting a terminal's SRS-based spatial domain channel according to an embodiment of the present disclosure.

[0170] The content of Fig. 14 may be related to the description of the relationship between the terminal and the base station in the spatial domain channel estimation process of Fig. 13. That is, the content of Fig. 14 may be related to the content of Fig. 13. Operations or content omitted in Fig. 14 may refer to the corresponding configuration of Fig. 13.

[0171] Referring to FIG. 14, at step 1410, a terminal may receive information about a port pattern from a base station. The information about the port pattern may include information about at least one port pattern. The information about the port pattern may be defined as candidate port pattern configuration information, but is not limited thereto. The terminal may receive the candidate port pattern configuration information via an RRC message. The candidate port pattern configuration information may consider a standard-defined table or a UE-specific port pattern.

[0172] When setting a standard defined port pattern, the terminal can receive configuration information from the base station that sets the indexes of patterns to be selectively used among the entire port pattern. If all port patterns corresponding to a specific number of ports are being used, the terminal can receive an index corresponding to the use of the entire pattern.

[0173] Spatial domain channel prediction performance is significantly influenced by port patterns, which can vary depending on the site's channel environment (e.g., scattering, reflection, etc.) and the vendor's antenna configuration implementation. Therefore, with greater freedom in port pattern design, vendors and / or sites can implement site-specific or site-common port patterns. If each vendor and / or site uses its own unique pattern, the terminal can receive the specific port pattern from the base station in bitmap form.

[0174] At step 1420, the terminal may receive port pattern configuration-related information from the base station. Port pattern configuration-related information may be received in the form of DCI, MAC CE, etc. At step 1420, one or more port patterns may be used. When one port pattern is used, the corresponding index may be set, and when multiple port patterns are used, the port pattern configuration may be performed by setting all of the corresponding indices.

[0175] When a port pattern is configured, additional parameters can be configured. Specifically, when multiple port patterns are configured, the port patterns can be used in a mixed manner based on the SRS transmission cycle and / or base station performance monitoring results. When using mixed port patterns, the terminal can use the port channels measured before the port pattern change to predict the new port pattern. Therefore, using mixed port patterns by additionally setting parameters can yield performance gains from an implementation perspective.

[0176] At step 1430, the terminal can transmit an SRS to the base station. The terminal can transmit the SRS to the base station based on port pattern settings. The terminal can transmit the SRS to the base station based on SRS configuration-related information and port pattern settings. The SRS may be an SRS based on a partial RX port.

[0177] FIG. 15 is a diagram illustrating a method for predicting a spatial domain channel based on SRS of a base station according to an embodiment of the present disclosure.

[0178] That is, the contents of Fig. 15 may be related to the contents of Fig. 13. Operations or contents omitted in Fig. 15 may refer to the corresponding configuration of Fig. 13.

[0179] Referring to FIG. 15, at step 1510, the base station may transmit port pattern information to the terminal. The port pattern information may include information about at least one port pattern. The port pattern information may be defined as candidate port pattern configuration information, but is not limited thereto. The candidate port pattern configuration information may be transmitted to the terminal via an RRC message. When configuring the port pattern candidate group, the base station may use a standardized table or a UE-specific (vendor-specific or site-specific) port pattern.

[0180] When configuring standard defined port patterns, the base station can transmit configuration information to the terminal by setting the indexes of patterns to be selectively used among the entire set of port patterns. If all port patterns corresponding to a specific number of ports are to be used, the index corresponding to the use of the entire set of patterns can be transmitted. For example, if only the number of ports, rather than the index of a specific port pattern, is indicated, the terminal and base station can be understood to be using the entire set of port patterns supported by that number of ports.

[0181] Spatial domain channel prediction performance is significantly influenced by port patterns, which can vary depending on the site's channel environment (e.g., scattering, reflection, etc.) and the vendor's antenna configuration implementation. Therefore, with greater freedom in port pattern design, vendors and / or sites can operate port patterns site-specifically or site-commonly. If each vendor and / or site uses its own unique pattern, the base station can set the port patterns it wishes to operate in bitmap form and transmit port pattern candidate configuration information to the terminal.

[0182] In step 1520, the base station can transmit information related to port pattern configuration to the terminal. The information related to port pattern configuration can be transmitted to the terminal in the form of DCI, MAC CE, etc. In step 1520, one or more port patterns can be indicated. When using one port pattern, the corresponding index can be set, and when using multiple port patterns, port pattern configuration can be performed by setting all of the corresponding indices. Multiple port patterns can be indicated by a method of indicating a specific group or a method of indicating a specific number of ports or a specific ratio.

[0183] When configuring a port pattern, additional parameters can be configured for the terminal. Specifically, when configuring multiple port patterns, the port patterns can be used in a mixed manner based on the SRS transmission cycle and / or base station performance monitoring results. When using mixed port patterns, the base station can use the port channels measured before the port pattern change to predict the new port pattern. Therefore, setting additional parameters to use mixed port patterns can yield performance gains from an implementation perspective.

[0184] At step 1530, the base station can receive an SRS from the terminal. The base station can receive an SRS from the terminal based on port pattern settings. The base station can receive an SRS from the terminal based on SRS configuration-related information and port pattern settings. The SRS may be an SRS based on a partial RX port.

[0185] At step 1540, the base station can predict (or estimate) a channel state. The channel state may be a spatial-domain channel state. The base station can predict the channel state for the full RX port based on the channel state measurement for the partial RX port. The base station can measure an SRS for the partial RX port transmitted from the terminal. The base station can estimate the channel state for the partial RX port based on the SRS measurement for the partial RX port. Thereafter, the base station can predict the channel state for the full RX port based on the estimated channel state for the partial RX port. For example, the base station can predict the channel state for the M full RX ports based on the channel state obtained based on the SRS measurements for the L partial RX ports. The process of predicting the full RX port channel state can be defined as a spatial-domain channel prediction process. AI can be used in the process of predicting the full RX port channel state. The AI ​​used at this time may be, for example, a channel predictor (220, 340), but is not limited thereto.

[0186] FIG. 16 is a diagram illustrating the structure of a terminal in a wireless communication system according to an embodiment of the present disclosure.

[0187] Referring to FIG. 16, a terminal according to an embodiment of the present disclosure may be configured to include a transceiver unit (1610), a control unit (1620), and a memory (1630). In the present disclosure, the control unit (1620) of the terminal may be defined as a circuit or an application-specific integrated circuit or at least one processor.

[0188] The transceiver (1610) can transmit and receive signals. The transceiver (1610) can transmit signals to a satellite or a base station according to an embodiment of the present disclosure, and receive signals from the satellite or the base station, for example.

[0189] The control unit (1620) can control the overall operation of the terminal according to various embodiments proposed in the present disclosure. For example, the control unit (1620) can control the signal flow between each block to perform operations according to the drawings (or, flowcharts, flow charts) described above.

[0190] For example, the control unit (1620) may control the terminal to receive information related to a channel state information reference signal (CSI-RS) from a base station, measure a partial antenna port channel state information reference signal (partial antenna port CSI-RS) transmitted from the base station based on the information related to the channel state information reference signal, estimate a channel state for a partial antenna port based on the measurement of the partial antenna port channel state information reference signal, obtain channel state information for a full antenna port based on the channel state for the partial antenna port, and report the channel state information for the full antenna port to the base station. The full antenna port is an antenna port through which the base station wishes to receive channel state information, and the partial antenna port may be a partial antenna port of the full antenna port.

[0191] In addition, the control unit (1620) can control the terminal to transmit channel prediction related capability information to the base station, receive candidate port pattern related information including a plurality of port patterns determined based on the channel prediction related capability information from the base station, and receive port pattern configuration related information indicating a port pattern to be measured by the terminal among the plurality of port patterns from the base station. The port pattern candidate related information can be received through an RRC (radio resource control) message, and the port pattern configuration related information can be received as downlink control information (DCI) or medium access control-control element (MAC-CE).

[0192] Additionally, the control unit (1620) can obtain channel state information for all antenna ports based on the results of artificial intelligence (AI) training performed on the port pattern candidate group and the channel state for the partial antenna port.

[0193] In addition, the control unit (1620) can control the terminal to measure the entire antenna port channel state information reference signal transmitted from the base station, compare the channel state information for the entire antenna port measured through the entire antenna port channel state information reference signal with the channel state information for the entire antenna port acquired based on the channel state for the partial antenna port, and transmit a message requesting the use of a port pattern that satisfies a preset criterion based on the comparison result to the base station.

[0194] The functions of the control unit are not limited to the examples above.

[0195] The memory (1630) can store at least one of information transmitted and received through the transceiver (1610) and information generated through the control unit (1620).

[0196] FIG. 17 is a diagram illustrating the structure of a base station in a wireless communication system according to an embodiment of the present disclosure.

[0197] Referring to FIG. 17, a base station according to an embodiment of the present disclosure may be configured to include a transceiver unit (1710), a control unit (1720), and a memory (1730). In the present disclosure, the control unit (1720) of the base station may be defined as a circuit or an application-specific integrated circuit or at least one processor.

[0198] The transceiver (1710) can transmit and receive signals. For example, the transceiver (1710) can transmit signals to a terminal according to an embodiment of the present disclosure, or receive signals from the terminal. Alternatively, the transceiver (1710) can transmit and receive signals with other satellites via an inter-satellite link, and can also perform multi-hop communication between satellites.

[0199] The control unit (1720) can control the overall operation of the base station according to various embodiments proposed in the present disclosure. For example, the control unit (1720) can control the signal flow between each block to perform operations according to the drawings (or, flowcharts, flow charts) described above.

[0200] For example, the control unit (1720) may control the base station to transmit information related to a channel state information reference signal (CSI-RS) to a terminal, transmit a partial antenna port channel state information reference signal (partial antenna port CSI-RS) to the terminal based on the information related to the channel state information reference signal, and receive channel state information for a full antenna port. The full antenna port may be an antenna port through which the base station wishes to receive channel state information, and the partial antenna port may be a part of the full antenna port.

[0201] In addition, the control unit (1720) can control the base station to receive channel prediction-related capability information from the terminal, transmit candidate port pattern-related information including a plurality of port patterns determined based on the channel prediction-related capability information to the terminal, and transmit port pattern configuration-related information indicating a port pattern to be measured by the terminal among the plurality of port patterns to the terminal. The port pattern candidate-related information can be received through an RRC (radio resource control) message, and the port pattern configuration-related information can be received as downlink control information (DCI) or a MAC-CE (medium access control-control element).

[0202] Additionally, the control unit (1720) can obtain channel state information for all antenna ports based on the results of artificial intelligence (AI) training performed on the port pattern candidate group and the channel state for the partial antenna port.

[0203] In addition, the control unit (1720) can control the base station to transmit a reference signal for full antenna port channel state information to the terminal and to receive a message requesting the use of a port pattern that satisfies a preset reference from the terminal.

[0204] The functions of the control unit are not limited to the examples above.

[0205] The memory (1730) can store at least one of information transmitted and received through the transceiver (1710) and information generated through the control unit (1720).

[0206] The methods proposed in this disclosure may be implemented by combining some or all of the contents included in each embodiment within a scope that does not harm the essence of the invention.

[0207] Meanwhile, the embodiments of the present invention disclosed in this specification and drawings are merely specific examples presented to easily explain the technical content of the present invention and aid in understanding the present disclosure, and are not intended to limit the scope of the present disclosure. In other words, it will be apparent to those skilled in the art to which the present disclosure pertains that other modified examples based on the technical concepts of the present disclosure are possible.

[0208] Furthermore, the present specification and drawings have disclosed preferred embodiments of the present disclosure, and although specific terms have been used, they are used in a general sense only to easily explain the technical contents of the present disclosure and to assist in the understanding of the invention, and are not intended to limit the scope of the present disclosure. It will be apparent to those skilled in the art to which the present disclosure pertains that other modified examples based on the technical idea of ​​the present disclosure are possible in addition to the embodiments disclosed herein.

Claims

1. A method performed by a terminal in a wireless communication system, A step of receiving information related to a channel state information-reference signal (CSI-RS) from a base station; A step of measuring a partial antenna port channel state information reference signal (partial antenna port CSI-RS) transmitted from the base station based on information related to the channel state information reference signal; A step of estimating a channel state for a partial antenna port based on measurement of a reference signal of the partial antenna port channel state information; A step of obtaining channel state information for a full antenna port based on a channel state for the partial antenna port; and A method comprising the step of reporting channel state information for all of the antenna ports to the base station.

2. In paragraph 1, The above total antenna ports are antenna ports through which the base station wishes to receive channel state information. A method characterized in that the partial antenna port is a part of the entire antenna port.

3. In paragraph 1, A step of transmitting channel prediction related capability information to the above base station; A step of receiving information related to a candidate port pattern group including a plurality of port patterns determined based on the channel prediction-related capability information from the base station; and Further comprising a step of receiving information related to port pattern configuration, which indicates a port pattern to be measured by the terminal among the plurality of port patterns, from the base station; The above port pattern candidate information is received via an RRC (radio resource control) message, A method characterized in that the above port pattern setting related information is received as DCI (downlink control information) or MAC-CE (medium access control-control element).

4. In paragraph 1, A method characterized in that the channel state information for the entire antenna port is obtained based on the results of artificial intelligence (AI) training performed on a group of port pattern candidates and the channel state for the partial antenna port.

5. In paragraph 1, A step of measuring a reference signal of the entire antenna port channel state information transmitted from the above base station; A step of comparing channel state information for all antenna ports measured through the above-mentioned entire antenna port channel state information reference signal with channel state information for all antenna ports obtained based on the channel state for the partial antenna port; and A method characterized by further comprising the step of transmitting a message requesting the use of a port pattern satisfying a preset criterion to a base station based on the comparison result.

6. A method performed by a base station in a wireless communication system, A step of transmitting information related to a channel state information-reference signal (CSI-RS) to a terminal; A step of transmitting a partial antenna port channel state information reference signal (partial antenna port CSI-RS) to the terminal based on information related to the channel state information reference signal; and A step of receiving channel state information for a full antenna port from the terminal, A method characterized in that the channel state information for the entire antenna port is obtained based on the channel state for the partial antenna port measured based on the measurement of the partial antenna port channel state information reference signal.

7. In paragraph 6, The above total antenna ports are antenna ports through which the base station wishes to receive channel state information. A method characterized in that the partial antenna port is a part of the entire antenna port.

8. In paragraph 6, A step of receiving channel prediction related capability information from the terminal; A step of transmitting information related to a candidate port pattern group including a plurality of port patterns determined based on the channel prediction-related capability information to the terminal; and Further comprising a step of transmitting port pattern configuration related information that instructs the terminal to measure a port pattern among the plurality of port patterns, to the terminal; The above port pattern candidate information is transmitted via an RRC (radio resource control) message. A method characterized in that the above port pattern setting related information is transmitted as DCI (downlink control information) or MAC-CE (medium access control-control element).

9. In paragraph 6, A method characterized in that the channel state information for the entire antenna port is obtained based on the results of artificial intelligence (AI) training performed on a group of port pattern candidates and the channel state for the partial antenna port.

10. In paragraph 6, A step of transmitting a reference signal for the entire antenna port channel state information to the terminal; and A method characterized by further comprising the step of receiving a message requesting the use of a port pattern satisfying preset criteria from the terminal.

11. In a terminal in a wireless communication system, Transmitter and receiver; and Including a control unit, The above control unit Receive information related to a channel state information-reference signal (CSI-RS) from a base station, Measure a partial antenna port channel state information reference signal (partial antenna port CSI-RS) transmitted from the base station based on information related to the above channel state information reference signal, Estimate the channel state for the partial antenna port based on the measurement of the reference signal for the partial antenna port channel state information, Obtain channel state information for the full antenna port based on the channel state for the above partial antenna port, A terminal that controls to report channel status information for all antenna ports to the base station.

12. In paragraph 11, The above total antenna ports are antenna ports through which the base station wishes to receive channel state information. A terminal characterized in that the above partial antenna port is a part of the above full antenna port.

13. In the 11th paragraph, the control unit, Transmit channel prediction related capability information to the above base station, Receive information related to a candidate port pattern including a plurality of port patterns determined based on the channel prediction-related capability information from the base station, Control to receive port pattern configuration-related information indicating a port pattern to be measured by the terminal among the plurality of port patterns from the base station, The above port pattern candidate information is received via an RRC (radio resource control) message, A terminal characterized in that the above port pattern setting related information is received as DCI (downlink control information) or MAC-CE (medium access control-control element).

14. In paragraph 11, A terminal characterized in that the channel state information for the entire antenna port is obtained based on the results of artificial intelligence (AI) training performed on a group of port pattern candidates and the channel state for the partial antenna port.

15. In the 11th paragraph, the control unit, Measure the entire antenna port channel state information reference signal transmitted from the above base station, The channel state information for the entire antenna port measured through the above-mentioned entire antenna port channel state information reference signal is compared with the channel state information for the entire antenna port obtained based on the channel state for the partial antenna port, A terminal characterized by controlling to transmit to a base station a message requesting the use of a port pattern that satisfies preset criteria based on the above comparison results.

16. In a base station in a wireless communication system, Transmitter and receiver; and Including a control unit, The above control unit Transmits information related to the channel state information-reference signal (CSI-RS) to the terminal, Transmitting a partial antenna port channel state information reference signal (partial antenna port CSI-RS) to the terminal based on information related to the above channel state information reference signal, Control to receive channel state information for the full antenna port, A base station, characterized in that the channel state information for the entire antenna port is obtained based on the channel state for the partial antenna port measured based on the measurement of the partial antenna port channel state information reference signal.

17. In paragraph 16, The above total antenna ports are antenna ports through which the base station wishes to receive channel state information. A base station, characterized in that the above partial antenna port is a part of the above full antenna port.

18. In paragraph 16, the control unit, Receive channel prediction related capability information from the above terminal, Transmitting candidate port pattern-related information including a plurality of port patterns determined based on the channel prediction-related capability information to the terminal, Control to transmit to the terminal information related to port pattern configuration that instructs the terminal to measure a port pattern among the plurality of port patterns, The above port pattern candidate information is transmitted via an RRC (radio resource control) message. A base station, characterized in that the above port pattern setting related information is transmitted as DCI (downlink control information) or MAC-CE (medium access control-control element).

19. A base station characterized in that the channel state information for the entire antenna port is obtained based on the results of artificial intelligence (AI) training performed on a group of port pattern candidates and the channel state for the partial antenna port.

20. In paragraph 16, the control unit, Transmits a reference signal for the entire antenna port channel status information to the above terminal, A base station characterized by controlling to receive a message requesting the use of a port pattern satisfying preset criteria from the terminal.

Images