Method and apparatus for determining channel quality indicator in wireless communication system

Abstract

A method of a wireless communication system may comprise the steps of: determining, by a terminal, whether to apply a CQI correction value on the basis of higher layer signaling provided from a base station, wherein each of the terminal and the base station has an AI / ML model, and the AI / ML model provided in the terminal corresponds to a CSI generation part and the AI / ML model provided in the base station corresponds to a CSI reconfiguration part; receiving a channel state information-reference signal (CSI-RS) from the base station; and transmitting, to the base station, an AI / ML-based CSI report including CSI output information on the basis of the CSI generation part and a CQI determined on the basis of the CSI-RS.

Description

Method and device for determining channel quality indicator in a wireless communication system

[0001] The present disclosure relates to a method and device for determining a channel quality indicator (CQI) in a wireless communication system. Specifically, the present disclosure relates to a method and device for reporting CQI for an artificial intelligence / machine learning (AI / ML)-based system in a wireless communication system.

[0002]

[0003] The International Telecommunication Union (ITU) is developing the International Mobile Telecommunication (IMT) framework and standards, and is currently discussing fifth-generation (5G) communications through a program called "IMT for 2020 and beyond."

[0004] To meet the requirements presented in "IMT for 2020 and beyond," the 3rd Generation Partnership Project (3GPP) NR (New Radio) system is being discussed to support various numerologies based on time-frequency resource units, taking into account various scenarios, service requirements, and potential system compatibility.

[0005] Additionally, 5G communications can support the transmission of physical signals or physical channels through multiple beams to overcome adverse channel conditions such as high path loss, phase noise, and frequency offset that occur at high carrier frequencies. This enables 5G communications to support applications such as enhanced Mobile Broadband (eMBB), massive Machine Type Communications (mMTC), and Ultra Reliable and Low Latency Communication (URLLC).

[0006]

[0007] The technical problem of the present disclosure is a method and device for determining CQI in a wireless communication system.

[0008] The technical problem of the present disclosure is a method and device for determining CQI in a wireless communication system to which an AI / ML model is applied.

[0009] The technical problem of the present disclosure is a method and device for determining CQI by considering a mismatch between actual channel information of a terminal and decoded channel information in a CSI reconstruction part on the base station side.

[0010] The technical problem of the present disclosure is a method and device for determining CQI using a proxy model provided in a terminal.

[0011] 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 disclosure belongs from the description below.

[0012] According to one aspect of the present disclosure, a method may include a step of determining, by a terminal, whether to apply a CQI (channel quality indicator) correction value based on upper layer signaling provided from a base station, wherein each of the terminal and the base station is provided with an AI (artificial intelligence) / ML (machine learning) model, wherein the AI / ML model provided in the terminal corresponds to a CSI generation part, and the AI / ML model provided in the base station corresponds to a CSI reconstruction part, and a step of receiving a CSI-RS (channel state information-reference signal) from the base station, and a step of transmitting, to the base station, an AI / ML-based CSI report including CSI output information based on the CSI generation part and a CQI determined based on the CSI-RS.

[0013] In addition, according to one aspect of the present disclosure, a wireless user device includes at least one processor and a memory storing instructions for causing the wireless user device to perform a specific operation by the at least one processor, wherein the specific operation is: determining whether to apply a channel quality indicator (CQI) correction value based on upper layer signaling provided from a base station, wherein each of a terminal and the base station has an artificial intelligence (AI) / machine learning (ML) model, wherein the AI / ML model provided in the terminal corresponds to a CSI generation part, and the AI / ML model provided in the base station corresponds to a CSI reconstruction part, and receiving a CSI-RS (channel state information-reference signal) from the base station, and transmitting an AI / ML-based CSI report including CSI output information based on the CQI determined based on the CSI-RS and the CSI generation part to the base station.

[0014] Additionally, the following may be commonly applied:

[0015] According to one aspect of the present disclosure, a terminal may receive CQI assistance information from a base station, apply the CQI assistance information to a CQI determined based on a CSI-RS, and transmit the corrected CQI to the base station by including it in an AI / ML-based CSI report.

[0016] In addition, according to one aspect of the present disclosure, the terminal may receive reconstructed CSI from a CSI reconstruction part of a base station as auxiliary information, generate CQI auxiliary information from the reconstructed CSI, apply the information to a CQI determined based on a CSI-RS, and transmit the corrected CQI to the base station by including it in an AI / ML-based CSI report.

[0017] Additionally, according to one aspect of the present disclosure, the base station can directly determine CQI assistance information based on the CSI generation part in the AI / ML-based CSI report received from the terminal, thereby correcting the CQI received from the terminal.

[0018] Additionally, according to one aspect of the present disclosure, the terminal can directly generate CQI assistance information without receiving assistance information from the base station, apply the CQI assistance information to the CQI determined based on the CSI-RS, and transmit the corrected CQI to the base station by including it in the AI / ML-based CSI report.

[0019] In addition, according to one aspect of the present disclosure, the terminal may further include a proxy model corresponding to a CSI reconstruction part of the base station, generate CQI assistance information based on the proxy model, apply the CQI assistance information to a CQI determined based on a CSI-RS, and transmit the corrected CQI to the base station by including it in an AI / ML-based CSI report.

[0020]

[0021] According to the present disclosure, a method for determining CQI in a wireless communication system can be provided.

[0022] According to the present disclosure, a method for determining CQI in a wireless communication system to which an AI / ML model is applied can be provided.

[0023] According to the present disclosure, a method for determining CQI by considering a mismatch between actual channel information of a terminal and decoded channel information in a base station-side CSI reconstruction part can be provided.

[0024] According to the present disclosure, a method for determining CQI using a proxy model provided in a terminal can be provided.

[0025] The effects that can be obtained from the present disclosure are not limited to the effects mentioned above, and other effects that are not mentioned will be clearly understood by a person having ordinary skill in the art to which the present disclosure pertains from the description below.

[0026]

[0027] FIG. 1 is a drawing for explaining the frame structure of a wireless communication system to which the present disclosure can be applied.

[0028] FIG. 2 is a diagram showing the resource structure of a wireless communication system to which the present disclosure can be applied.

[0029] FIG. 3 is a diagram illustrating CSI-RS resources to which the present disclosure can be applied.

[0030] FIG. 4 is a diagram illustrating a type 1 CSI applicable to the present disclosure.

[0031] FIG. 5 is a diagram illustrating a method of using multiple SRSs that can be applied to the present disclosure.

[0032] FIG. 6 is a diagram illustrating a method for performing non-codebook based precoding that can be applied to the present disclosure.

[0033] Figure 7 is a diagram illustrating AI / ML model life cycle management applicable to the present disclosure.

[0034] FIG. 8 is a diagram illustrating a method for performing AI / ML model training based on Type 1 applicable to the present disclosure.

[0035] FIG. 9 is a diagram illustrating a method for performing AI / ML model training based on Type 2 applicable to the present disclosure.

[0036] FIG. 10 is a diagram illustrating a method for performing AI / ML model training based on Type 3 applicable to the present disclosure.

[0037] FIG. 11 is a diagram showing an AI encoder structure and input and output values ​​to which the present disclosure can be applied.

[0038] FIG. 12 is a diagram illustrating an inference process and procedure for CSI compression purposes to which the present disclosure can be applied.

[0039] FIG. 13 is a diagram illustrating a signaling procedure for using an AI / ML model applied to the present disclosure.

[0040] FIG. 14 is a diagram illustrating a method in which a terminal applied to the present disclosure receives auxiliary information from a base station and determines and reports a CQI for an AI / ML model.

[0041] FIG. 15 is a diagram illustrating a method in which a terminal applied to the present disclosure receives reconstructed CSI information from a base station and determines and reports CQI for an AI / ML model.

[0042] FIG. 16 is a diagram illustrating a method in which a base station applied to the present disclosure receives a CQI decision for an AI / ML model from a terminal and applies a CQI correction value.

[0043] FIG. 17 is a diagram illustrating a method for a terminal applied to the present disclosure to determine whether to perform additional CQI correction for an AI / ML model and derive and report a CQI value.

[0044] FIG. 18 is a diagram illustrating a method for deriving a CQI value based on a proxy model provided in a terminal applied to the present disclosure.

[0045] FIG. 19 is a diagram illustrating a method for deriving a CQI value based on a proxy model and auxiliary information provided in a terminal applied to the present disclosure.

[0046] FIG. 20 is a diagram illustrating a method for transmitting a precoded CSI-RS based on target CSI information of a terminal acquired in advance by a base station applied to the present disclosure.

[0047] Figure 21 is a flowchart of a method for a terminal to derive a CQI value applied to the present disclosure.

[0048] Figure 22 is a drawing showing a base station device and a terminal device to which the present disclosure can be applied.

[0049]

[0050] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the attached drawings so that those skilled in the art can easily implement the present disclosure. However, the present disclosure may be implemented in various different forms and is not limited to the embodiments described herein.

[0051] In describing embodiments of the present disclosure, detailed descriptions of known components or functions will be omitted if they are deemed to obscure the gist of the present disclosure. Furthermore, portions of the drawings that are irrelevant to the description of the present disclosure have been omitted, and similar portions are designated with similar reference numerals.

[0052] In the present disclosure, when a component is said to be "connected," "coupled," or "connected" to another component, this may include not only a direct connection, but also an indirect connection in which another component exists in between. Furthermore, when a component is said to "include" or "have" another component, unless otherwise specifically stated, this does not exclude the other component, but rather implies that the other component may be included.

[0053] In this disclosure, terms such as first, second, etc. are used solely to distinguish one component from another, and do not limit the order or importance of components unless specifically stated otherwise. Accordingly, within the scope of this disclosure, a first component in one embodiment may be referred to as a second component in another embodiment, and similarly, a second component in one embodiment may be referred to as a first component in another embodiment.

[0054] In this disclosure, distinct components are used to clearly illustrate their respective characteristics, and do not necessarily imply that the components are separated. That is, multiple components may be integrated into a single hardware or software unit, or a single component may be distributed into multiple hardware or software units. Therefore, even if not specifically mentioned, such integrated or distributed embodiments are also included within the scope of this disclosure.

[0055] In the present disclosure, the components described in various embodiments are not necessarily essential components, and some may be optional components. Therefore, embodiments comprising a subset of the components described in one embodiment are also within the scope of the present disclosure. Furthermore, embodiments including other components in addition to the components described in various embodiments are also within the scope of the present disclosure.

[0056] The present disclosure describes a wireless communication network, and operations performed in the wireless communication network may be performed in a process of controlling the network and transmitting or receiving a signal in a system (e.g., a base station) that manages the wireless communication network, or in a process of transmitting or receiving a signal in a terminal connected to the wireless network.

[0057] It is self-evident that various operations performed for communication with terminals in a network consisting of multiple network nodes including a base station can be performed by the base station or other network nodes other than the base station. 'Base station (BS)' can be replaced by terms such as fixed station, Node B, eNodeB (eNB), ng-eNB, gNodeB (gNB), and access point (AP). In addition, 'terminal' can be replaced by terms such as UE (User Equipment), MS (Mobile Station), MSS (Mobile Subscriber Station), SS (Subscriber Station), and non-AP station (non-AP STA).

[0058] In the present disclosure, transmitting or receiving a channel means transmitting or receiving information or a signal through the channel. For example, transmitting a control channel means transmitting control information or a signal through the control channel. Similarly, transmitting a data channel means transmitting data information or a signal through the data channel.

[0059] In the following description, the term NR (New Radio) system is used for the purpose of distinguishing the system to which various examples of the present disclosure are applied from existing systems; however, the scope of the present disclosure is not limited by this term.

[0060] NR systems support a variety of subcarrier spacings (SCS) to accommodate diverse scenarios, service requirements, and potential system compatibility. Furthermore, NR systems can support the transmission of physical signals / channels across multiple beams to overcome challenging channel conditions, such as high path loss, phase noise, and frequency offsets that occur at high carrier frequencies. This enables NR systems to support applications such as enhanced Mobile Broadband (eMBB), massive Machine Type Communications (mMTC) / ultra Machine Type Communications (uMTC), and Ultra Reliable and Low Latency Communications (URLLC).

[0061] Hereinafter, 5G mobile communication technology can be defined to include not only the NR system, but also the existing LTE-A (Long Term Evolution-Advanced) system and LTE (Long Term Evolution) system. 5G mobile communication may include technology that operates in consideration of backward compatibility with previous systems as well as the newly defined NR system. Therefore, the 5G mobile communication below may include technology that operates based on the NR system and technology that operates based on previous systems (e.g., LTE-A, LTE), and is not limited to a specific system.

[0062] First, we would like to briefly explain the physical resource structure of the wireless communication system to which the present invention is applied.

[0063] FIG. 1 is a drawing for explaining an NR frame structure to which the present disclosure can be applied.

[0064] In NR, the basic unit of time domain is It can be, , and N can be 4096. Meanwhile, in LTE, the basic unit of the time domain is It can be, And, =2048. The constant for the multiplication relationship between the NR time base unit and the LTE time base unit is k= can be defined as

[0065] Referring to Figure 1, the time structure of a frame for downlink / uplink (DL / UL) transmission is can have. Here, one frame is It consists of 10 subframes corresponding to time. The number of consecutive OFDM symbols in each subframe is = It can be. In addition, each frame is divided into two half frames of the same size, half frame 1 can be composed of sub frames 0-4, and half frame 2 can be composed of sub frames 5-9.

[0066] represents the timing advance (TA) between the downlink (DL) and uplink (UL). Here, the transmission timing of the uplink transmission frame i is determined based on the downlink reception timing at the terminal, based on the following mathematical expression 1.

[0067] [Mathematical Formula 1]

[0068]

[0069]

[0070] Here, It may be a TA offset value that occurs due to differences in duplex mode, etc. In FDD (Frequency Division Duplex), has a value of 0, but in TDD (Time Division Duplex), it takes into account the margin for DL-UL switching time. It can be defined as a fixed value. For example, in TDD (Time Division Duplex) of FR1 (Frequency Range 1), which is a frequency below 6 GHz, is 39936 or 25600 It could be 39936 is 20.327μs, and 25600 is 13.030μs. Also, at FR2 (Frequency Range 2), which is a millimeter wave (mmWave) frequency, is 13792 It could be. At this time, 13792 is 7.020 μs.

[0071] FIG. 2 is a diagram showing an NR resource structure to which the present disclosure can be applied.

[0072] Resource elements (REs) within a resource grid can be indexed according to each subcarrier spacing. Here, one resource grid can be created for each antenna port and each subcarrier spacing. Uplink and downlink transmission and reception can be performed based on the corresponding resource grid.

[0073] In the frequency domain, one Resource Block (RB) consists of 12 REs, and each of the 12 REs can be configured with an index (nPRB) for one RB. The index for an RB can be utilized within a specific frequency band or system bandwidth. The index for an RB can be defined as in the following mathematical expression 2. Here, represents the number of subcarriers per RB, and k represents the subcarrier index.

[0074] [Equation 2]

[0075]

[0076]

[0077] Different numerologies can be configured to meet the diverse services and requirements of NR systems. For example, while LTE / LTE-A systems can support a single subcarrier spacing (SCS), NR systems can support multiple SCSs.

[0078] A new numerology for NR systems supporting multiple SCSs can operate in frequency ranges or carriers such as below 3GHz, 3GHz-6GHz, 6GHZ-52.6GHz or above 52.6GHz to address the issue of not being able to use wide bandwidth in frequency ranges or carriers such as 700MHz or 2GHz.

[0079] Table 1 below shows examples of numerologies supported by the NR system.

[0080] [Table 1]

[0081]

[0082]

[0083] Referring to Table 1 above, the numeral can be defined based on the subcarrier spacing (SCS), cyclic prefix (CP) length, and number of OFDM symbols per slot used in the Orthogonal Frequency Division Multiplexing (OFDM) system. The above values ​​can be provided to the terminal through the upper layer parameters DL-BWP-mu and DL-BWP-cp for the downlink, and through the upper layer parameters UL-BWP-mu and UL-BWP-cp for the uplink.

[0084] In Table 1 above, when the subcarrier spacing setting index (u) is 2, the subcarrier spacing (Δf) is 60 kHz, and normal CP and extended CP can be applied. For other numerology indices, only normal CP can be applied.

[0085] A normal slot can be defined as the basic time unit used to transmit a single piece of data and control information in an NR system. The length of a normal slot can be set to the number of OFDM symbols, which is 14 by default. Furthermore, unlike slots, a subframe has an absolute time length equivalent to 1 ms in an NR system and can be used as a reference time for the length of other time intervals. Here, for coexistence or backward compatibility between LTE and NR systems, a time interval similar to an LTE subframe may be required in the NR standard.

[0086] For example, in LTE, data can be transmitted based on a unit of time called a Transmission Time Interval (TTI), which can be set to one or more subframes. Here, one subframe can be set to 1ms and can contain 14 OFDM symbols (or 12 OFDM symbols).

[0087] In addition, a non-slot can be defined in NR. A non-slot can mean a slot having a number that is at least one symbol smaller than a normal slot. For example, in the case of providing low latency such as URLLC service, latency can be reduced through a non-slot having a number of symbols smaller than a normal slot. Here, the number of OFDM symbols included in a non-slot can be determined by considering the frequency range. For example, a non-slot with a length of 1 OFDM symbol can be considered in a frequency range of 6 GHz or higher. As an additional example, the number of OFDM symbols defining a non-slot can include at least 2 OFDM symbols. Here, the range of the number of OFDM symbols included in a non-slot can be set as the length of a mini-slot up to a predetermined length (e.g., the normal slot length - 1). However, as a specification of a non-slot, the number of OFDM symbols may be limited to 2, 4, or 7 symbols, but is not limited thereto.

[0088] Additionally, for example, in unlicensed bands below 6 GHz, subcarrier spacing where u equals 1 and 2 may be used, and in unlicensed bands above 6 GHz, subcarrier spacing where u equals 3 and 4 may be used. For example, when u equals 4, it may be used for SSB (Synchronization Signal Block).

[0089] [Table 2]

[0090]

[0091]

[0092] Table 2 shows the number of OFDM symbols per slot for normal CP, by subcarrier spacing setting (u). ), number of slots per frame ( ), number of slots per subframe ( ) is shown. Table 2 shows the above-described values ​​based on a normal slot having 14 OFDM symbols.

[0093] [Table 3]

[0094]

[0095]

[0096] Table 3 shows the number of slots per frame and the number of slots per subframe based on normal slots with 12 OFDM symbols per slot when extended CP is applied (i.e., when u is 2 and the subcarrier spacing is 60 kHz).

[0097] As mentioned above, one subframe may correspond to 1ms on the time axis. Additionally, one slot may correspond to 14 symbols on the time axis. For example, one slot may correspond to 7 symbols on the time axis. Accordingly, the number of slots and symbols that can be considered within 10ms corresponding to one radio frame may be set differently. Table 4 may show the number of slots and symbols according to each SCS. In Table 4, the 480kHz SCS may not be considered, but is not limited to these examples.

[0098] [Table 4]

[0099]

[0100]

[0101] A terminal can calculate downlink CSI parameters (e.g., CQI, PMI, RI, L1-RSRP, etc.) through a DL (downlink) CSI (channel state information) reporting procedure and report the calculated CSI information to a base station. The base station can utilize the received CSI information for downlink data transmission. The CSI parameters are measurement values ​​associated with a channel state, and the base station can perform downlink data transmission based on the measurement values. The base station can instruct the terminal about data scheduling information based on the CSI parameters. The terminal generally reports the CSI calculation and derived CSI parameters to the base station as feedback information by receiving a CSI-RS transmitted by the base station. Alternatively, the terminal can measure the channel state through another downlink signal (e.g., SSB (synchronization signal block)) in addition to the CSI-RS and report the measured value to the base station.

[0102]

[0103] Select RI (rank indicator):

[0104] RI can be a CSI parameter that indicates the number of layers available for downlink transmission in a specific channel environment. RI can also indicate the maximum number of uncorrelated paths available for downlink transmission. Here, other CSI parameters (e.g., PMI, CQI) can be calculated based on the rank provided by RI.

[0105]

[0106] Select precoding matrix index (PMI):

[0107] A PMI may include a set of indices corresponding to a precoding matrix. The base station may apply the PMI reported by the terminal to downlink transmission. However, the base station may have the freedom to apply a precoding matrix other than the PMI. For example, the standard may support multiple types of codebooks to report the precoding matrix for the PMI. Each codebook type is described below.

[0108] Codebook type

[0109] - Type 1 single-panel codebooks

[0110] - Type 1 multi-panel codebooks

[0111] - Type 2 codebooks

[0112] - Enhanced Type 2 codebooks

[0113]

[0114] For example, PMI selection may be performed based on the codebook type, the number of transmission layers represented by RI values, and other CSI reporting configuration parameters (e.g., antenna panel dimensions). Each codebook may include a set of precoding matrices. For example, in NR, a dual-stage precoding matrix codebook design may be applied, but is not limited thereto.

[0115] The dual-stage precoding matrix may be as shown in Equation 3 below. In Equation 3, W1 may be a matrix representing a DFT (discrete Fourier transform) beam or a beam group for both polarizations according to the codebook type, and W2 may be selected from one of the following according to the codebook type.

[0116] W2 codebook type

[0117] - Beam selection from W1

[0118] - Weighting coefficients for the beams in W1

[0119] - Cophasing values ​​between two polarizations

[0120]

[0121] [Equation 3]

[0122]

[0123]

[0124] The terminal calculates a PMI value that represents the highest SINR (signal interference noise ratio) value at the receiver using the codebook selected for "Type I single-panel codebook" and "Type I multi-panel codebook" as the codebook type and all possible precoding matrices in the given channel environment. For example, a Type 1 codebook may be considered for PMI selection in a single user multi-input multi-output (SU-MIMO) scenario, but is not limited thereto.

[0125] A Type II codebook can consider a set of orthogonal DFT beams. In Equation 3, the W1 matrix contains information associated with the DFT beams, and the PMI selection function calculates beam amplitude scaling and cophasing values ​​(W2) for all DFT beams within each orthogonal beam group for a given channel environment and number of DFT beams. The terminal can report the i1 and i2 values ​​to the base station based on the PMI selection function, and each index corresponds to a precoding matrix (W) that provides the maximum SINR value. A Type II codebook-based PMI can provide more accurate channel information than a Type I codebook-based PMI. A Type II codebook can provide more accurate channel information because multiple beams are applied to estimate the channel eigenvector. For example, a Type II codebook-based PMI can be used in a MU (multi)-MIMO scenario because it considers multiple beams, but is not limited thereto. For example, a Type 2 codebook can only support up to Rank 2, while an enhanced Type 2 codebook can support up to Rank 4 with overhead reduction methods, which will be described later.

[0126] Furthermore, based on the above, CSI reporting performed by a terminal may consider explicit CSI feedback and implicit CSI feedback. Explicit CSI feedback is a method in which the terminal directly feeds back channel information (e.g., channel eigenvector / eigenvalue, SVD) indicating the channel state measured by the terminal, and may not consider how the base station processes the reported CSI. On the other hand, implicit CSI feedback may be a method in which a selected precoder matrix (desired precoder matrix) W within a set of candidate precoder matrices from a possible precoder codebook is indicated.

[0127] For example, CSI parameters (or CSI contents) related to CSI feedback may be as shown in Table 5 below, but may not be limited thereto.

[0128] [Table 5]

[0129]

[0130]

[0131] The CSI content within a CSI report can be set by different combinations of the CSI parameters in Table 5. Here, the CSI content within a CSI report can be set by reportQuantity.

[0132] For example, if set to "none", the base station may trigger aperiodic CSI-RS transmission. Aperiodic CSI-RS transmission may be triggered by the terminal for channel measurement and receiver parameter adjustment, and related CSI reporting may not be performed by the terminal. Aperiodic CSI-RS transmission may be, but is not limited to, aperiodic transmission of CSI-RS or TRS (tracking reference signal) for tracking purposes for synchronization purposes, or spatial aperiodic receive beam sweeping for the terminal to adjust the receive filter. The following operations do not require the base station to report aperiodic CSI to the terminal, but may operate within the same framework as aperiodic CSI requests.

[0133] - Aperiodic transmission of the Tracking Reference Signal (TRS or CSI-RS for tracking)

[0134] - Aperiodic Rx beam sweeping (so called P3 beam sweep) - adjusts the Rx spatial received filter to the UE

[0135]

[0136] Additionally, CSI contents within a single CSI report may be set as shown in Table 6 below, but may not be limited thereto. For example, the combinations of CSI contents in Table 6 below are representative combinations of CSI contents that a base station will primarily need.

[0137] [Table 6]

[0138]

[0139]

[0140] For example, “cri-RI-i1” in Table 6 can operate based on “Hybrid beamformed / non-precoded CSI acquisition operation” as shown in FIG. 3.

[0141] The base station can mainly use UE-specific beamformed CSI-RS (320) with several ports per CSI-RS resource for the purpose of CSI acquisition. However, in order for the base station to recognize how to beamform the CSI-RS for each UE, non-precoded CSI-RS resources (310) can be transmitted in the middle together with multiple antenna ports (e.g., 32 ports). Here, the UE can report PMI corresponding to all antenna ports (e.g., 32 ports) based on the received CSI-RS, thereby indicating a preferred beam direction. For example, beamforming of the CSI-RS can be based on the wideband / long-term part of the reported precoder. That is, the reported W1 metrics can be used to perform UE-specific CSI-RS beamforming. Additionally, the short-term subband CS (short-term / subband CSI) (W2) can be determined from the beamforming CSI-RS.

[0142] For example, non-precoded CSI-RS (310) can be used to recognize how CSI-RS beamforming is performed, and based on this, the terminal may only need to report CSI corresponding to W1 and not report for W2. This can reduce overhead.

[0143] Additionally, the terminal can select a PMI based on a specific number of transmit antennas. Specifically, the terminal can select a PMI based on a specific number of transmit antennas given by the number of antenna ports of a configured CSI-RS associated with the reporting configuration.

[0144] Additionally, in the case of MU-MIMO scheduling, the base station may select a PMI other than the PMI reported by the target terminal, taking into account interference or channel conditions from other terminals simultaneously scheduled. Considering the above, two types of CSI can be defined in a wireless communication system, as shown in Table 7 below.

[0145] [Table 7]

[0146]

[0147]

[0148] Specifically, type 1 CSI may have different sub-codebook formats based on different antenna configurations at the transmitter. As an example, FIG. 4 is a diagram illustrating type 1 CSI applicable to the present disclosure. Specifically, FIG. 4(a) may be a case where 16 antenna ports are configured with (N1, N2) = (4, 2). The precoding matrix W for type 1 single panel CSI may be expressed as the product of two matrices (W1, W2), each of which may be independently reported as a different part of the overall PMI. For example, W1 may reflect long-term frequency independent channel characteristics, and thus may report channel information corresponding to the entire bandwidth (Wideband reporting). For example, W1 may be as shown in Equation 4 below. W1 may indicate a set of beams pointing in different directions, and matrix B may define one beam. In Equation 4, based on the 2 X 2 block structure, there can be two polarizations, and the selection of W1 can be a limited set that has the same beam direction.

[0149] [Equation 4]

[0150]

[0151]

[0152] As a concrete example, for rank 1 or 2, a single beam or four neighboring beams can be indicated by the W1 matrix. The correct beam among the four neighboring beams can be selected by W2 reported for each subband, thereby selecting an optimized beam for each subband. Furthermore, W2 can provide cophasing between two polarizations. If W1 defines a single beam, then W2 in B, which is a single-column matrix, can only provide cophasing between two polarizations. On the other hand, if multiple neighboring beams are defined by W1, W2 can select one correct beam and provide cophasing between two polarizations within that beam.

[0153] On the other hand, if the rank is greater than 2, W1 can define N neighboring orthogonal beams, which can be as shown in Equation 5. In Equation 5, N beams used for transmission of R transmission layers can be indicated, and W2 can only provide cophasing between two polarizations. Here, transmission to the same terminal can be possible for up to 8 transmission layers.

[0154] [Equation 5]

[0155]

[0156]

[0157] On the other hand, W2 is short-term and can potentially indicate different channel characteristics depending on the frequency. W2 may not be reported in certain cases. Here, the base station can randomly select W2 for each physical resource block group (PRG), and the terminal can select the channel quality indicator (CQI) based on this assumption.

[0158] Also, Fig. 4(b) may be a Type 1 multi-panel CSI. Multi-panel CSI may be applied considering a case where multiple antenna panels are used at a base station. Here, coherence between different panels may be difficult to guarantee. For example, Fig. 4(b) may be a case where (N1, N2) = (4, 1) and there are 32 antenna ports by 4 panels, but the present invention is not limited thereto. Here, the difference between the single panel of Fig. 4(a) and the multi-panel of Fig. 4(b) is that coherence may be difficult to guarantee between the antenna ports of different antenna panels. In the multi-panel, W1 may be the same set of beams for different polarizations and panels, just like in the single panel. On the other hand, in the case of the multi-panel, W2 may provide cophasing not only between polarizations but also between multi-panels for each subband.

[0159] Type 2 CSI can provide channel information with higher spatial granularity than Type 1 CSI and can be primarily applied to MU-MIMO scenarios. For example, Type 2 CSI may be limited to a maximum of two ranks, while Enhanced Type 2 CSI may be extended to four ranks, but may not be limited to a specific form. For convenience of explanation, the terms Type 2 CSI and Enhanced Type 2 CSI are referred to below, but may not be limited thereto.

[0160] Here, in the above-described mathematical expressions 3 and 4, W1 may be a wideband report. Type 2 CSI may report up to four beams, corresponding to four columns within B. W2 may provide amplitude values ​​(partially wideband and partially subband reporting) and phase values ​​(subband reporting) for each of the four reported beams and two polarizations within B.

[0161] Type 2 CSI can provide more detailed channel model information than Type 1, including main ray and amplitude and phase information. Based on the CSI reported from multiple terminals, the base station can determine which combinations of terminals are appropriate for simultaneous transmission on the same time / frequency resource.

[0162] Additionally, enhanced Type 2 CSI (e.g., Rel-16 enhanced Type 2 CSI) may be considered. Enhanced Type 2 CSI can report a set of beams on the wideband, similar to Type 2 CSI, along with a set of combining coefficients on an additional narrowband. Here, the reported beams can be linearly combined using the combining coefficients to provide a set of precoder vectors for each transmission layer. For example, in the existing Type 2 CSI, the combining coefficient value can be reported independently for each subband. Therefore, even considering the fact that the channels between adjacent subbands are highly correlated, the existing Type 2 CSI must report a relatively large combining coefficient value on a per-subband basis, which can increase the overhead. Enhanced Type 2 CSI can utilize frequency correlation to reduce this overhead. Furthermore, enhanced Type 2 CSI can improve the frequency-domain granularity of the PMI report. For example, compression operations can be applied to subbands or half-subbands based on frequency domain units. While conventional Type 2 CSI uses one precoder per subband, enhanced Type 2 CSI can use a recommended precoder for each frequency domain unit. However, actual reporting can be performed together for all frequency units. Specifically, given layer k, the reported precoders for all frequency domain units can be expressed as in Equation 6 below, where N can be the number of frequency domain units.

[0163] [Equation 6]

[0164]

[0165]

[0166] Here, Equation 6 may not be a mapping of layers and antennas, but may mean a set of precoder vectors for the entire set of frequency domain units for layer k. For the entire k layers, there may be a set of k precoder vectors, each set may be composed of N precoder vectors, and the layer and antenna port mapping for a specific frequency domain unit n may be as shown in Equation 6 below.

[0167] Additionally, in the expression for vectors in Equation 7, W1 can be the same as Equation 4 described above, which is the same as the existing Type 2 CSI.

[0168] [Equation 7]

[0169]

[0170] Here, the B columns can correspond to the L selected beams, and W1 can be the same for all frequency domain units. (wideband reporting) Also, W1 can be the same for all layers.

[0171] Additionally, as an example, the enhanced Type 2 CSI has a compression matrix (size M x N) can be considered. can be constructed as a set of row vectors from the DFT basis, and can provide a transformation from N dimensions in the frequency domain (N frequency domain units) to M dimensions in the smaller delay domain. can be frequency independent, but can be reported independently for each layer. That is, can be a single common metric for all frequency domain units, but can be reported independently for each layer. The number of rows can be equal to M in Equation 8. In Equation 8, R can be the number of frequency domain units per subframe (R=1 or 2), and p can be a configurable parameter that controls the amount of compression.

[0172] [Equation 8]

[0173]

[0174] Finally (2L x M) can perform mapping from the delay domain to the beam domain. For example, W2 of the existing Type 2 CSI can generate a similar matrix for all subbands, and the matrix can be mapped from the frequency (subband) domain to the beam domain. On the other hand, can perform allocation from a smaller delay domain to a beam domain. can be adjusted by the reported rank and compression factor parameters p, which allows reporting based on fewer parameters. In addition, You can adjust the amount of CSI to be reported through the value, which allows you to get a smaller Fewer parameters can be reported based on size.

[0175] Additionally, as an example, the enhanced Type 2 CSI may be expanded to a maximum of 4 reported ranks and a maximum number of selectable beams to 6, with reduced overhead and improved frequency domain granularity. Table 8 below may illustrate possible configurations of the enhanced Type 2 CSI, but is not limited thereto.

[0176] [Table 8]

[0177]

[0178]

[0179] Additionally, multi-antenna precoding can be considered in uplink transmission. PUSCH (physical uplink shared channel) MIMO precoding can support up to four layers. For uplink transmission, DFT-based precoding can only support single-user (SU) MIMO. For example, a terminal can perform codebook-based transmission and non-codebook-based transmission with two different PUSCH precoding modes. In case of codebook-based precoding, the uplink grant transmitted by the base station can include precoder-related information (e.g., PMI). It is assumed that the terminal uses the precoder provided by the base station in the uplink differently from the downlink transmission. Furthermore, coherence can be assumed between the terminal antennas in uplink MIMO transmission, and the phases of the signals transmitted on the two antennas can be adjusted. Here, if there is no coherence between the antenna ports, each antenna port may result in a random relative phase, which may mean that the use of antenna port specific weight factors is limited.

[0180] For example, the coherence between antenna ports can correspond to any one of full coherence, partial coherence, and no coherence, and the corresponding information can be provided as terminal capability information. In addition, the use of multiple sounding reference signals (SRS) for codebook-based PUSCH can be considered. FIG. 5 is a diagram illustrating a method of using multiple SRSs applicable to the present disclosure. Referring to FIG. 5, a terminal can transmit relatively large beams on a multi-port SRS. For example, the beams can correspond to different terminal antenna panels (in different directions), and each panel can include a set of antenna elements. That is, each panel can correspond to antenna ports of a multi-port SRS. The base station can transmit an SRI (SRS Resource Indicator) to the terminal to determine whether to perform transmission through which beam. The terminal can determine which transmission to perform within the selected beam based on the number of layers and precoder using SRI and precoder information.

[0181] FIG. 6 is a diagram illustrating a method for performing non-codebook-based precoding applicable to the present disclosure. Referring to FIG. 6, when performing non-codebook-based precoding, a terminal can perform transmission based on a precoder indication and channel measurement received from a base station. In codebook-based precoding, the terminal performs transmission based on an uplink precoder selected based on channel measurement by the base station, but in non-codebook, uplink transmission can be performed through a precoder selected based on measurement based on channel reciprocity. Here, each column of the precoder matrix W can be a digital beam for a corresponding layer. When the terminal selects a precoder for N layers, it can be regarded as selecting N different beam directions, and each beam can correspond to one possible layer.

[0182] For example, referring to FIG. 6, a terminal may be capable of PUSCH transmission based on the selected precoding. However, the precoder selected by the terminal based on downlink channel measurements may not be the optimal precoder from the base station's perspective. Considering the above, the base station may modify the precoder selected by the terminal. For example, the base station may remove some beams or some columns selected by the terminal from the precoder and transmit information about this to the terminal. Specifically, the base station may indicate the result of beam selection based on measured SRS information back to the terminal through SRI. Based on the indicated information, the terminal may perform only some beam transmissions, which may implicitly indicate reduced layer transmission.

[0183]

[0184] A Channel Quality Indicator (CQI) may be a channel measurement value reported by a terminal to a base station for the purpose of reporting a desired Modulation and Coding Scheme (MCS) of the terminal to the base station. A precoding matrix determined based on a precoding matrix indicator (PMI) and a rank indicator (RI) may be applied by the base station to CSI-RS transmission. CSI-RS may be transmitted from the base station to the terminal for channel measurement of the terminal. For example, multiple CSI-RS transmissions for a virtual PDSCH transmission may assume different PMIs and RIs, and thus multiple CSI-RSs may be transmitted to the terminal. The terminal may first determine (or derive) an RI value through the received CSI-RS, and then determine (or derive) a PMI value based on the determined RI value. The terminal may finally calculate a CQI based on the determined (or derived) RI and PMI values. A precoding matrix indicated based on specific RI and PMI values ​​can be used for subsequent PDSCH transmissions by the base station. Here, when the base station applies the precoding matrix reported by the terminal to PDSCH precoding, the base station can set the MCS based on the reported CQI. That is, the terminal can determine the CQI based on the assumptions about the precoding matrix and RI values ​​based on the PDSCH and report it to the base station. The base station can determine the MCS value for subsequent PDSCH scheduling based on the CQI value reported by the terminal. Here, the base station may not only use the CSI information reported by the terminal, but may utilize the information for PDSCH scheduling for the corresponding terminal by considering the CSI information reported by the terminal.That is, the CSI parameters (PMI, RI, CQI, LI (layer indicator), R1-RSRP (reference signal received power), CRI (CSI-RS resource indicator)) reported by the terminal may be channel state report values ​​recommended by the terminal to the base station. Among the CSI parameters, CQI may be a value corresponding to one SINR (signal interference noise ratio) value and may be used as an input value for link adaptation. When determining the CQI value, the terminal may consider a virtual PDSCH transmission that schedules and transmits one TB (transport block). Here, it may be assumed that the virtual PDSCH is allocated to a radio resource set by the CSI reference resource parameter. The CSI reference resource may mean information corresponding to assumptions about information on the used precoding, RS (reference signal) overhead, bandwidth, and other information. In addition, the CSI reference resource may include relationship information on time resources associated with the CSI. That is, the CSI reference resource may be used as a timing reference. A CSI reference resource may refer to a specific downlink slot to which it is allocated. Accordingly, slots later than the specific downlink slot may not be used as slot resources for CSI reporting.

[0185] The terminal can determine the highest CQI index that can provide virtual PDSCH scheduling with a TB error probability that satisfies the target BLER (Block Error Rate) as the CQI value. For example, the target BLER may be 10%, but may not be limited thereto. As another example, in case of supporting URLLC (ultra reliable low latency communications), the target BLER may be It could be a level.

[0186] In addition, the CQI table may be indicated in the form of a subset of the MCS table. The CQI table may consist of 4 or 5 bits, and four tables may be used to support up to 64QAM, 256QAM, URLLC, and 1024QAM, respectively, but may not be limited to the embodiment. In addition, a single wideband MCS value may be applied and used to the entire TB, but if the base station wishes to use frequency selective scheduling considering multiple sub-bands, the terminal may measure an independent CQI value for each subband and report it to the base station. For example, if a subband CQI is configured, the terminal may use a differential subband CQI in addition to the wideband CQI for each configured subband, as shown in Table 9 below. Additionally, in Table 9, the sub-band offset level can be as shown in Equation 9 below, and the terminal can report an independent CQI value for each subband through a 2-bit subband CQI difference value.

[0187] [Table 9]

[0188]

[0189]

[0190] [Equation 9]

[0191]

[0192]

[0193] For example, AI / ML (artificial intelligence / machine learning)-based operations may be performed below. The terms in Table 10 below may be used within the AI / ML-based framework, but may not be limited thereto.

[0194] [Table 10]

[0195]

[0196]

[0197]

[0198] For example, deep learning, a subcategory of machine learning (ML), focuses on parameterizing multi-layer neural networks. Deep learning can be a method for learning data representation and has demonstrated high performance in image classification, speech recognition, and natural language processing. However, deep learning can require a large amount of data during the training process, and the computational complexity required to train deep learning-based models on large data sets can be high, necessitating high hardware performance.

[0199] Regarding AI / ML, AI / ML-based mobile communications can be implemented. For example, AI has been able to reduce CAPEX and OPEX costs by managing data from devices operating on existing wireless communication systems. Furthermore, the introduction of next-generation networks may require low operating costs and high return-to-cost compensation, and AI can help address the complexity and optimization challenges of network operations. Furthermore, network intelligence and automation may be critical issues in next-generation mobile communication systems, and these challenges can be addressed using AI / ML. For example, budget management, network management, equipment lifecycle management, service level agreement (SLA) management, network performance management, and network planning can be implemented using AI / ML. AI / ML can enhance the Quality of Experience (QoE) of network operation and usage, but is not limited to this. Furthermore, AI / ML can improve network quality and enhance the provision of more personalized services, but is not limited to a specific form.

[0200] For example, AI / ML algorithms could be implemented within network equipment for some purposes of equipment and network operation. Here, the issue of optimizing network operation between manufacturers and operators is a major technical issue in operating next-generation communication systems, as described above. Therefore, an AI / ML model that considers this issue may be necessary. In particular, a method of applying AI / ML algorithms within the wireless interface required to transmit and receive wireless signals between multiple network entities, such as between base stations and terminals or between terminals, may be necessary. For example, AI / ML technologies may be utilized in 5G NR (New Radio), a mobile communication technology, or in future 6G mobile communication systems to significantly improve wireless performance and reduce complexity / latency, but may not be limited thereto. Table 11 below may be a basic framework considered in a wireless interface based on AI / ML technologies, but may not be limited thereto.

[0201] [Table 11]

[0202]

[0203]

[0204]

[0205] For example, AI / ML-based operations may consider two types of operations: model generation and inference operations. Model generation may be performed based on model training, model validation, model testing, and application stages. For example, model training may be performed based on at least one of input / output, pre-processing / post-processing, and online / offline application, but may not be limited thereto.

[0206] In addition, the inference operation can be performed based on the input / output, pre / post-processing, and application stages. The life cycle management (LCM) procedure of a general AI / ML framework and an AI / ML model can be performed based on the following steps. As an example, FIG. 7 is a diagram illustrating the AI / ML model life cycle management applicable to the present disclosure. Referring to FIG. 7, data collection for an AI / ML model can be performed. Here, the collected data can include data for model training, monitoring data for model management and performance monitoring, and inference data for model inference. Model training can perform model learning based on training data, and the learned and updated model can be stored. Here, the model can be stored based on a function or model identifier or an identifier related to the model and additional information, but is not limited to a specific form. In addition, the model training can be performed based on model training control information based on model management and performance monitoring.

[0207] Model management and performance monitoring can control model inference based on monitoring data. Model management and performance monitoring can control the activation / deactivation / selection / switching / fallback of model inference, as well as other operations. Model management and performance monitoring can obtain information about the results of model inference. Model inference can obtain inference data, perform inference operations, and obtain result information. Model inference can receive models stored in storage and perform inference operations based on them.

[0208] In relation to the LCM (life cycle management) process, one AI / ML model may have one model ID and related information and / or model functions for AI / ML operations, as described above.

[0209] Actions within the LCM procedure

[0210] - Data Collection

[0211] - Model Training

[0212] - Functionality / model identification

[0213] - Model transfer

[0214] - Model Inference

[0215] - Functionality / model selection, activation, deactivation, switching and fallback operation

[0216] - Functionality / model monitoring

[0217] - Model update

[0218] - Model Storage

[0219] - UE capability

[0220]

[0221] For example, AI / ML-based operations in wireless communication systems can consider the following cases. However, this is only one example, and AIML-based operations can also be applied to other cases.

[0222] AI / ML-based operation of wireless communication systems

[0223] CSI (Channel State Information): compression (**), temporal prediction (*)

[0224] BM(Beam Management): spatial prediction (*), temporal prediction (*)

[0225] Positioning: direct (*), assisted (*)

[0226] (**) is a two-side model (*) is a single-side model

[0227]

[0228] As a concrete example, we can consider AI / ML model training collaborations for CSI compression using a two-sided model.

[0229] For example, with respect to CSI compression, AI / ML model inference can be performed on both the terminal and the base station. That is, AI / ML model inference can be performed on both sides (both the terminal and the base station). For this purpose (model inference for CSI compression), a CSI generation part can be deployed on the terminal, and a CSI reconstruction part corresponding to the CSI generation part can be deployed on the base station. The CSI generation part is a part for CSI compression, and the CSI reconstruction part can be used for an operation for CSI reconstruction to obtain compressed CSI information from the generation part, and to obtain (recover) more accurate CSI for better MU (multi-user) scheduling operation in a massive MIMO scenario, but is not limited thereto, and cooperation on the two-side AI / ML model can be performed in various forms.

[0230] As another example, considering limitations in the capabilities or computing power of the terminal, it may be considered that a third entity (e.g., a server for the terminal) performs AI / ML model training, monitoring, or inference operations within the LCM procedure. That is, the terminal-side operations (e.g., model training / inference / monitoring / switching / activation / deactivation / validation / testing) as the AI / ML operations described above may include operations of other entities and are not limited to a specific form.

[0231] For example, training collaboration can consider Types 1 through 3 below. In the following, all AI / ML model-related operations performed at the base station or network are assumed to be performed at the network level. However, this may not be limited. Since the "network" is an inclusive term that includes the base station, the two terms may be used interchangeably. What is performed at the network can encompass both the base station side and the broader network side. For convenience of explanation, the following description will be described as being performed at the network.

[0232]

[0233] For example, joint training can be performed based on Type 1. Type 1 can be a method in which training for an AI / ML model is performed on one side / entity. Other sides / entities can then obtain specific model parts through downloading / uploading, and CSI feedback operations can be performed based on these.

[0234] As a specific example, FIG. 8 is a diagram illustrating a method for performing AI / ML model training based on Type 1. Referring to FIG. 8, model training for the CSI generation part of the terminal and the CSI reconstruction part of the base station can be performed on the network (820) side in a joint training manner. Thereafter, the network side can provide the trained model information to the terminal (821) through downloading and can also apply it to the network (822). As another example, model training for the CSI generation part of the terminal and the CSI reconstruction part of the base station can be performed on the terminal (810) side in a joint training manner. Thereafter, the terminal side can apply the trained model information to the terminal (811) and provide it to the network (812) through an uplink. In other words, an AI / ML model can be trained on one side / entity, and the trained AI / ML model can be transferred to another side / entity.

[0235] As a specific example, network-side model 1 training can train the CSI generation part using a data set related to CSI generation for training purposes in the network, and train the CSI reconstruction part by transferring the result value into the training loop of the CSI reconstruction part. Here, the CSI reconstruction part located in the base station can be utilized for subsequent AI / ML model-based CSI feedback. On the other hand, the CSI generation part (i.e., CSI encoder) that performed model training by the base station can transfer the trained model information to the terminal via the wireless interface (UE download AI / ML model). The terminal can perform the AI / ML model-based CSI feedback operation using the downloaded CSI generation part model. That is, CSI compression can be performed through the CSI generation part to perform CSI feedback.

[0236] On the other hand, when the AI / ML model is trained on the terminal side, terminal-side Type 1 training can train the CSI generation part using a data set of CSI generation for training purposes at the terminal. Then, the output value can be provided to the training loop of the CSI reconstruction part, and the CSI reconstruction part can be trained based on the output value. For example, the CSI generation part located at the terminal can be used for subsequent generation of AI / ML model-based CSI feedback information (e.g., CSI compression). The CSI reconstruction part (i.e., CSI decoder) that performed model training by the base station can transmit the trained model information to the base station via the wireless interface (UE upload AI / ML model). Thereafter, the base station can perform an AI / ML model-based CSI feedback operation using the CSI reconstruction part model received through the terminal upload. That is, CSI feedback information can be recovered (reconstructed) through the CSI reconstruction part to obtain CSI feedback information.

[0237] Additionally, AI / ML model joint training type 1 may require AI / ML model exchange. For example, AI / ML model inference may require common reference models / parameters for model inference and common AI / ML model inference algorithms, which may require AL / ML model exchange. For example, terminal-specific joint training type 1 may perform AI / ML model transfer based on the open format of the AI / ML model structure recognized by the terminal. On the other hand, terminal-non-specific joint training type 1 may mean model transfer within an open format without the terminal being aware of the AI / ML model structure. For example, when training a terminal-side model on the network side, the terminal-specific model may be trained on the network side based on the terminal capabilities and terminal model structure, but may not be limited to a specific form. Furthermore, the model transfer method in the above training type 1 is not limited to the wireless interface, and can be transferred to another wireless entity in advance via a wired connection.

[0238]

[0239]

[0240] As another example, FIG. 9 is a diagram illustrating a method for performing AI / ML model training based on Type 2 applicable to the present disclosure. Referring to FIG. 9, Type 2 model training may be identical to Type 1 in that both the CSI generation part and the CSI reconstruction part are performed within a single, identical loop. However, Type 2 model training differs from Type 1 in that training is not performed on a specific side / entity, but on both sides / entities. That is, both the network and the terminal may be involved in the training.

[0241] For example, referring to FIG. 9, the terminal side can perform a forward propagation (FP) operation based on a data sample. That is, the terminal (910) can generate a CSI feedback result and transmit the FP result as information thereon to the network side. Thereafter, the network (920) can perform a CSI reconstruction operation based on the FP result, thereby training the CSI reconstruction part. The network side can generate BP (backward propagation) information (e.g., gradients) based on the above-described method and transmit the generated BP information back to the terminal side. The terminal side can perform training for the CSI generation part based on the BP information received from the network side. Here, the data set can be aligned on both the terminal side and the network side, and training can be performed through the above-described method.

[0242] Also, as an example, FIG. 10 is a diagram illustrating a method for performing AI / ML model training based on Type 3 applicable to the present disclosure. Referring to FIG. 10, Type 3 model training is a sequential training method, and unlike the joint training of Types 1 and 2, model training can be performed sequentially / independently for each node. That is, Type 3 model training can be performed based on the exchange of aligned data set information between a branch station and a terminal within a separate loop, without training the CSI generation part and the CSI reconstruction part within the same loop. As an example, Type 3 model training can be distinguished into network-first training or terminal-first training, similar to Type 1 model training.

[0243] Referring to Figure 10, in the case of network-first training, the network may use the CSI generation part for training the CSI reconstruction model set on the network side. However, the CSI generation part is for training purposes and may not be used in subsequent model inference.

[0244] The network (1020) side inputs the CSI generation part's input values ​​( ) based on the corresponding CSI generation part and its output value ( ) and the model training of the CSI reconstruction part based on this to obtain the final output value ( ) can be generated. The CSI reconstruction part for which training is completed can be used for the AI / ML-based CSI feedback operation of the network. Thereafter, the network (1020) side can share the data set information (e.g., dataset including labels and intermediate results) utilized in the model training of the network side with the terminal after the model training is completed. The terminal (1010) side can apply the data set information to the CSI generation part training. Thereafter, a joint inference operation can be performed through the CSI generation part performed on the terminal (1010) side and the CSI reconstruction part performed on the network (1020). In addition, information on the CSI generation part trained on the terminal (1020) side (e.g., model / functionality identification process - information on available CSI generation parts) can be reported or transmitted to the network (1020). The network (1020) can select or optimize a CSI reconstruction part corresponding to the CSI generation part model of the terminal based on information about the CSI generation part obtained from the terminal (1010).

[0245] As an example, options for training a terminal-side CSI generation model for network-first type 3 joint training can be considered as shown in Table 12 below.

[0246] [Table 12]

[0247]

[0248]

[0249] The terminal-side CSI output format or the base station-side CSI input format may be as shown in Table 13 below. That is, the CSI generated by the terminal by considering AI / ML and transmitted to the base station may be in the form of Table 13 below. In Table 13, Option 1 may be a case where the terminal-side CSI output format or the base station-side CSI input format is in the form of a precoding matrix similar to the existing CSI feedback. The terminal may report CSI feedback to the base station in the form of a precoding matrix together with the confirmed layer number information. For example, the terminal may report CSI feedback in the angular-delay domain as the domain of Option 1-b of Table 13, but may not be limited to the embodiment. (e.g., via spatial-frequency DFT domain transformation to angular / delay domain)

[0250] On the other hand, option 2 in Table 13 may be a case where the terminal-side CSI output format or the base station-side CSI input format is a channel matrix with explicit CSI feedback. That is, the explicit CSI feedback of option 2 may not include precoding vectors assigned to each layer as a channel matrix. The explicit channel matrix may be based on channels in the space-frequency domain (option 2-a) or based on channels in the angle-delay domain (option 2-b), but is not limited to a specific form. For example, in the angle-delay domain of option 2-b, loss may be caused by sub-selection of angle-delay domain-based indices, but is not limited thereto.

[0251] [Table 13]

[0252]

[0253]

[0254] Additionally, as an example, applicable new content and transmission methods between AI CSI reports for AI / ML-based CSI compression may be as shown in Tables 14 and 15 below. The priority applied by reflecting AI CSI reports may be determined based on Table 14 below.

[0255] [Table 14]

[0256]

[0257]

[0258] [Table 15]

[0259]

[0260]

[0261] For example, the input values ​​for AI / ML-based encoders can be the eigenvectors of the raw channel matrix (H) or the ray channel matrix (V). Here, if the raw channel matrix is ​​the input, the eigenvector calculation may be performed at the base station, which may increase the complexity of the base station. For example, the raw channel matrix may contain unnecessary information for precoder selection, which may increase unnecessary feedback overhead.

[0262] Therefore, considering the aforementioned characteristics, the AI ​​reporting procedure can be performed by considering which CSI content information will be used as input values ​​for the AI ​​encoder. Furthermore, by determining which CSI content information may impact the AI ​​decoder and how its performance may change accordingly, the AI ​​CSI reporting procedure can be performed by combining appropriate input values ​​from the input values ​​in Tables 14 and 15 described above, but is not limited to a specific format.

[0263] For example, when comparing AI / ML-based quantized CSI compression information for an AI / ML-based CSI feedback mechanism with existing CSI report / contents, the AI / ML-based CSI report fields may have differences as shown in Table 16 below.

[0264] [Table 16]

[0265]

[0266]

[0267] Below, we describe a method for performing CSI compression based on AI / ML algorithms. AI / ML algorithm-based CSI compression methods can consider compression of CSI information that can be generated and measured in the spatial-frequency domain. Here, AI / ML algorithm-based CSI compression (or AI / ML-based CSI compression) methods can also be applied to compression of CSI information that can be generated and measured in the time domain, and may not be limited to a specific form. Below, we describe issues that may arise in relation to CQI determination when CSI compression methods are applied to mobile communication systems (e.g., LTE / NR and 6G wireless access technologies) and solutions therefor.

[0268] FIG. 11 is a diagram illustrating an AI encoder structure and input and output values ​​to which the present disclosure can be applied. Referring to FIG. 11, an AI encoder (1110) can perform AI encoding using AI CSI inputs (AI CSI inputs) 1121. Thereafter, the AI ​​encoder (1110) can perform a quantization procedure, thereby deriving an AI encoder output (AI encoder output) 1122. Here, the AI ​​CSI inputs (1121) can include an original channel eigenvector. In addition, as an example, the AI ​​CSI inputs (1121) can additionally include at least one of a terminal performance-related index, channel environment information, cell-related information, location information, previous channel environment information, and various other values ​​based on the AI ​​encoder design, and are not limited to a specific form.

[0269] The AI ​​encoder output (1122) may be information in the form of a bit stream or bit sequence. The AI ​​encoder output (1122) may be transmitted from the terminal to the base station, and an AI decoder located at the base station may reconstruct AI CSI information through a decoding process. This process may be performed in at least one of training, inference, and monitoring within the above-described LCM procedure, and may also be utilized for monitoring or switching other AI models.

[0270] FIG. 12 is a diagram illustrating an inference process and procedure for CSI compression purposes to which the present disclosure can be applied.

[0271] Referring to FIG. 12, CSI compression may be applied based on a two-sided AI / ML model. For example, a CSI generation part (CSI generation part or CSI compression part, 1210) may be located at a terminal, and a corresponding CSI reconstruction part (CSI reconstruction part or CSI decompression part, 1220) may be located at a base station. In other words, AI / ML models may be included at both the terminal and the base station. Here, each of the CSI parts (1210, 1220) is configured as a pair, and the CSI decoder needs to decode and match the signal transmitted from the CSI encoder. That is, the models for the CSI generation part and the CSI reconstruction part need to be matched and operated, and in order to create such a matching structure, a paired model specific to various conditions such as the base station / terminal structure setting, network conditions, wireless transmission / reception structure, and network / terminal wireless environment assumed during training can be located at the base station and the terminal and utilized for the purpose of CSI compression.

[0272] The terminal side can determine the AI ​​CSI input values ​​described above for the CSI generation part (1210) and input them to the CSI generation part (1210), and the CSI generation part (1210) can derive an output value of the CSI generation part through an inference process. The derived output value of the CSI generation part can be included in AI / ML-based CSI information for CSI feedback through a quantization process. The terminal can transmit the AI / ML-based CSI information including the output value of the CSI generation part to which quantization has been applied to the base station. The base station can perform a dequantization procedure on the AI / ML-based CSI information received (or reported) from the terminal, and apply a value for the result of the dequantization procedure to the CSI reconstruction part (1220) to generate an output value of the CSI reconstruction part through inference.

[0273] For example, both the output value of the CSI generation part and the input value of the CSI reconstruction part may be in the form of a precoding matrix. As a specific example, the output value of the CSI generation part and the input value of the CSI reconstruction part may be configured as a precoding matrix in the space-frequency domain. As another example, the output value of the CSI generation part and the input value of the CSI reconstruction part may be configured as a precoding matrix generated using channel information in the angular-delay domain. As another example, the terminal may report AI / ML-based CSI information to the base station based on explicit channel information in the space-frequency domain or the angular-delay domain, and may not be limited to a specific embodiment.

[0274] A terminal may report AI / ML-based CSI information to a base station. For example, the CSI information may be composed of CSI Part 1 and CSI Part 2, but may not be limited thereto. When a terminal performs CSI reporting, CSI Part 1 may include size information for CQI, RI, and CSI Part 2. In addition, CSI Part 2 may include information about an output value of a CSI generation part, but may not be limited thereto. Thereafter, the base station may utilize the output value of the decoded CSI reconstruction part as channel information for PDSCH transmission scheduling for the terminal. For example, the AI / ML model inference operation of the CSI reconstruction part may further consider channel information measured based on the AI / ML model and method. As another example, the AI / ML model inference operation of the CSI reconstruction part may additionally consider input / output values ​​to which pre / post precoding operations are applied, and may not be limited to a specific form.

[0275] Here, the terminal cannot accurately recognize the decoded channel output information through the CSI reconstruction part of the base station. Specifically, since the terminal cannot recognize the information about the algorithm and structure of the CSI reconstruction part of the base station, it cannot recognize the decoded channel output information through the CSI reconstruction part of the base station. In the above-described situation, when the terminal performs a CSI report to the base station, the terminal can transmit CQI information to the base station by including CSI part 1 as described above. Here, the terminal and the base station may perceive different channel-related information (e.g., PMI (precoding matrix information), channel matrix / eigenvector (explicit channel information)) considered to derive the CQI information, and thus, a problem may arise in which the derivation of the CQI information transmitted to the base station is performed differently. Specifically, referring to Table 17 below, the base station can determine the CQI by determining the channel output value of the CSI reconstruction part using the CSI feedback information received from the terminal. On the other hand, the terminal can determine the CQI based on the original channel information measured on the terminal side. That is, the channel assumptions for CQI determination may differ between the base station and the terminal. Here, the CQI value may be the same CQI value reported from the terminal to the base station.

[0276] That is, as shown in Table 17, the base station and the terminal may both have the same understanding (or assumption) about the corresponding CQI value derived from the terminal and reported to the base station, but the channel assumptions applied to determine (or derive) the CQI value may be different. The terminal basically derives the CQI value based on the original channel information (target CSI) assumed by the terminal side, and the base station can derive the CQI value based on the decoded channel information through the base station-side CSI reconstruction part, and thus a misalignment problem may occur for the CQI value.

[0277] That is, the CQI value (target CSI) based on the original (realistic) channel information by the terminal and the CQI value determined based on the decoded channel information through the CSI reconstruction part on the base station side may be different. For example, the CQI value provided by the terminal is likely to be an overestimated value. Consequently, data channel transmission scheduled based on such an overestimated CQI value is likely to be determined by an error on the terminal side, which may cause problems such as a negative impact on the data transmission rate and reliability. Alternatively, if the original channel information by the terminal is based on the existing codebook and the CQI value based on it is assumed, the CQI value calculated by the terminal may be an underestimated value. Considering the above, the following describes a method for solving the CQI misalignment problem between the base station and the terminal.

[0278] For example, in a wireless communication system (e.g., NR), a channel understanding (or assumption) for a terminal to calculate CQI may be determined. The channel understanding (or assumption) may be determined so that there is no mismatch between the MCS applied by the base station for PDSCH transmission and the CQI value reported by the terminal, but is not limited to this embodiment. Here, if the channel understanding (or assumption) for CQI calculation is unclear, a link adaptation error may occur between the base station and the terminal, which may result in performance degradation. When CQI calculation is performed in the two-side model described above, the terminal may not recognize the trained decoder structure of the base station, and thus, the channel understanding (or assumption) may be different, which may result in a link adaptation error. As a specific example, the terminal may not recognize the reconstructed output channel matrix / eigenvector because the trained decoder is located at the base station. Here, the terminal can determine the CQI value based on the actual measured value. The terminal can assume a virtual PDSCH transmission by the base station and select the highest CQI index among the virtual PDSCH scheduling schemes that result in a TB error probability lower than the target BLER. However, errors may occur during this process.

[0279] Hereinafter, a method for determining a CQI value is described considering a case where an AI / ML-based CSI compression technique is applied in the above-described manner. For example, in the case where the terminal includes a CSI generation part and the base station includes a CSI reconstruction part in the above-described two-side model, AI / ML-based CSI compression is performed in the CSI generation part of the terminal and transmitted to the base station, and the base station can decode the AI / ML-based CSI information received from the terminal through the CSI reconstruction part. Here, the channel output value (e.g., output channel matrix / eigenvector) decoded from the CSI reconstruction part of the base station may be information recognized only by the base station and may not be recognized by the terminal. Therefore, the terminal may have limitations in calculating an accurate CQI value. The terminal may calculate the CQI based on at least one of the measured original channel matrix and eigenvector, but the information may mismatch with the channel information decoded from the CSI reconstruction part of the base station. Below, a CQI determination method for improving the efficiency of AI / ML-based CSI compression effects based on the above is described.

[0280] For example, the base station can calculate the CQI value based on the PMI value in the CSI report reported by the terminal. Accordingly, the base station and the terminal can maintain a mutual understanding (or assumption) about the CQI value. Since the base station uses the PMI value reported by the terminal to generate a precoder, the results for the same CQI value can be maintained and applied between the base station and the terminal. On the other hand, in the case of PDSCH scheduling with an AI / ML-based CSI compression method applied, the channel information decoded by the base station and the original channel information measured by the terminal may be mismatched, as described above.

[0281] In the following, based on the above, the AI / ML-based CSI compression configuration between the base station and the terminal and the CQI reporting configuration within the CSI report are described based on at least one of the upper layer signaling and the L1 signaling. In addition, the AI / ML-based CSI compression configuration between the base station and the terminal and the CQI reporting configuration within the CSI report can be enabled or activated based on at least one of the upper layer signaling and the L1 signaling.

[0282] The base station and the terminal can apply AI / ML models for two-side model AI / ML-based CSI compression, and the base station and the terminal can exchange information about the corresponding capabilities (e.g., UE capabilities, hardware / software implementation) to perform operations based on the capabilities. For example, the base station and the terminal can perform model identification / functionality identification through the LCM procedure, thereby recognizing applicable AI / ML models and functions between them. In other words, interaction can be performed between AI / ML model parts located on the base station and the terminal sides, respectively. In the above-described environment, the base station and the terminal can perform AI / ML-based CSI reporting based on the same understanding (or assumption) to derive a highly reliable CQI value.

[0283] FIG. 13 is a diagram illustrating a signaling procedure for using an AI / ML model applied to the present disclosure. FIG. 13 may be a method for a terminal (1310) and a base station (or network 1320, hereinafter referred to as a base station) to set a CSI compression function based on terminal capability (UE capability) before applying an AI / ML-based CSI compression method. For example, the base station (1320) may transmit a UE capability inquiry (UECapabilityEnquiry) message to the terminal (1310) to determine whether AI / ML-based CSI compression is applicable. The terminal (1310) may transmit a UE capability information (UECapabilityInformation) message including information on whether the AI / ML-based CSI compression function is supported to the base station (1320). Additionally, the terminal may further transmit information about the training collaboration type for the two-sided model (e.g., Type 1 (Joint training of the two-sided model at a single side / entity), Type 3 (Separate training at network side and UE side)) to the base station (1320). For example, the above-described information may be transmitted as part of the configuration information transmitted by the base station (1320) to the terminal (1310), but may not be limited to a specific form.

[0284] As a specific example, the base station (1320) may transmit upper layer configuration information required for the AI / ML-based CSI compression function to the terminal (1310). Here, the configuration information may include a configuration that allows the terminal to report UE assistance information (UAI) to the base station for reporting a function required or desired by the terminal. In addition, the configuration information may include at least one of configuration information for additional conditions on the network / terminal side, configuration information required for inference within the CSI generation part, configuration information for the CSI reconstruction part on the network side, configuration information for a model corresponding to the CSI reconstruction part, configuration information for a proxy model related to the CSI reconstruction part, and information on the applied learning collaboration type. The configuration information is upper layer configuration information, and the base station (1320) may transmit the configuration information to the terminal (1310) through upper layer signaling.

[0285] The terminal (1310) can determine the AI / ML-based CSI compression function (or AI / ML-based CSI compression model) using at least one of the configuration information acquired through the above-described method and the terminal-side internal configuration information. For example, the terminal (1310) can determine the AI / ML-based CSI compression function (or AI / ML-based CSI compression model) using at least one of model implementation information, model utilization information, and other information in addition to the acquired configuration information and the terminal-side internal configuration information. The terminal (1310) can report the AI / ML-based CSI compression function (or AI / ML-based CSI compression model) determined based on the above-described method to the base station (1320) through terminal assistance information (UAI). Here, the base station (1320) can additionally provide the terminal (1310) with configuration information for inference based on the applicable AI / ML function as described above. As a specific example, setting information that is not included in the above-described upper layer setting information but is necessary for inference to the terminal may be additionally provided from the base station (1320) to the terminal (1310).

[0286] As another example, although the configuration information is already provided in the upper layer configuration information, the base station (1320) may provide the terminal (1310) with information to indicate whether to update the configuration information or maintain the existing configuration, and is not limited to a specific form.

[0287] For example, the above-described configuration information may be performed based on the RRC reconfiguration procedure of the base station (1320) and the terminal (1310), but may not be limited thereto. The terminal (1310) may perform the CSI compression function based on the latest information set through the RRC reconfiguration procedure in which the upper layer configuration information is transmitted. Here, an activation or deactivation procedure including instruction information for a CSI generation part-CSI reconfiguration part pair may be performed, and is not limited to a specific form. In addition, when an AI / ML model corresponding to the CSI generation part-CSI reconfiguration part pair is operated on each of the terminal (1310) and the base station (1320), an LCM operation may be performed based on inference and model monitoring. Through the above-described signaling procedure, applicable AI / ML models (or AI / ML functions) can be matched between the terminal (1310) and the base station (1320), and an inference operation for CSI compression can be performed based on this. A method for resolving the CQI mismatch problem when CQI reporting is performed based on the above-described method is described.

[0288]

[0289] For example, the terminal may measure original channel information (e.g., target CSI) for calculating a CQI value, and calculate and report the CQI value to the base station based on the measured original channel information. Here, the target CSI, which is the original channel information, may be actual measurement data information (or ground-truth data information) measured on the terminal side and used as at least one of the AI / ML model input value and output value for AI / ML model training or monitoring purposes. Here, the base station may provide assistance information to the terminal, thereby resolving the above-described mismatch problem. The assistance information may be information that can be reflected in the terminal determining (or calculating or deriving, hereinafter referred to as determination) the CQI value. That is, the terminal may perform additional correction by utilizing the assistance information in the process of determining the CQI value. As a specific example, the assistance information may include channel information derived by the CSI reconstruction part of the base station. The terminal can recognize the difference (or offset) between the channel information actually estimated by the terminal and the channel information derived by the base station based on the received auxiliary information, and the terminal can correct the CQI value based on the difference (or offset). For example, the auxiliary information may include at least one of the information in Table 17 below. That is, the methods in Table 17 may be considered to resolve the CQI mismatch problem caused by the compression and reconstruction process between the base station and the terminal.

[0290] Specifically, the base station can calculate a CQI value using the output value of the CSI reconstruction part, and the calculated CQI value can be provided to the terminal. The terminal can compare the provided CQI value with the CQI value calculated by the terminal based on actual channel information to determine a corrected final CQI value, and report the corresponding CQI value to the base station. In other words, the base station can provide the terminal with the CQI value obtained using the output value of the CSI reconstruction part. (Option 1)

[0291] Additionally, the base station can transmit a CQI difference offset value (differential CQI offset) to the terminal. The terminal can derive a final CQI value by applying the CQI difference offset value provided by the base station to the CQI value calculated using actual channel information, and report the resulting CQI value to the base station. In other words, the base station can directly determine (or derive) the offset value for the CQI difference and transmit the offset value for the CQI difference to the terminal so that the terminal can apply the value (option 2).

[0292] As another example, the base station may provide at least one of the reconstructed CSI and channel state information to the terminal as the output value of the CSI reconstruction part. The base station may transmit at least one of the reconstructed CSI (e.g., PMI / RI or channel eigenvector) and channel state information to the terminal in the form of PMI / RI or channel eigenvector. The terminal may compare at least one of the reconstructed CSI and channel state information provided by the base station with target CSI information measured by the terminal to independently correct the CQI value and report the corrected CQI value to the base station. (Option 3)

[0293] As another example, the base station can transmit a CSI-RS to the terminal based on the CSI information reconstructed using the output value of the CSI reconstructed part. The base station can transmit the CSI-RS to the terminal in the form of auxiliary information, with the CSI / channel state information corresponding to the output value of the CSI reconstructed part applied. The terminal can receive the CSI-RS, check the output value of the CSI reconstructed part determined by the base station based on this, and derive a CQI value based on this information. (Option 4)

[0294] [Table 17]

[0295]

[0296]

[0297] Based on the above-described Table 17, the terminal can compare the CQI value calculated based on the measured actual channel information (target CSI) with the auxiliary information provided to the terminal by the base station to correct the CQI value. Thereafter, the terminal can report the corrected CQI value to the base station. That is, the terminal can correct the CQI value based on the auxiliary information signaling provided from the base station, and report the corrected CQI index value to the base station by including it in the CSI report (e.g., CSI Part 1). For example, the auxiliary information signaling by the base station can be transmitted to the terminal in the form of CSI channel information (e.g., PMI / RI) or channel matrix / eigenvector, which are past CSI reconstruction part output values. The corresponding CSI reconstruction part output values ​​can be provided to the terminal in the form of all values, part of values, an average value, the most recent value, or other forms of past output values, depending on the base station implementation, and is not limited to a specific form.

[0298] Here, the terminal can determine the original CQI value based on the target CSI corresponding to the actual channel information measured by the terminal and the SINR value measured based on the channel matrix / eigenvector of the actual channel. The terminal can finally determine a corrected CQI value based on the determined CQI value and the auxiliary information provided by the base station, and the corrected CQI value can be reported to the base station. In order to determine the corrected CQI value through the above-described process, the terminal can select the highest CQI index by considering the following factors.

[0299] For example, the terminal may select the highest CQI index that does not exceed the next TB reception error probability within the downlink physical resource block group of the CSI reference resource based on at least one of the Transport Block Size (TBS), the Target Code Rate, and the modulation method corresponding to the corrected CQI index in one PDSCH TB.

[0300] For example, in FIG. 13, a terminal may be configured with a higher layer parameter (e.g., AI-CQI-Assistance) through an RRC reconfiguration procedure by a base station. Here, the higher layer parameter (e.g., AI-CQI-Assistance) may be a setting indicating whether to apply an additional CQI correction value.

[0301] In addition, as an example, the terminal may report to the base station whether CQI correction can be performed based on the procedure described above in FIG. 13. If CQI correction can be applied to the terminal, the base station may provide the terminal with upper layer configuration information for performing the corresponding function, as illustrated in FIG. 13. The terminal may perform additional correction on the CQI value calculated through the target CSI through the configuration information provided from the base station. As an example, the upper layer configuration information may include information related to enabling / disabling CQI correction application, CQI correction application values ​​(e.g., differential CQI offset values ​​and / or CQI compensation value range), CQI correction type (base station correction type, terminal compensation type and / or compensation method), and other information, and is not limited to a specific form.

[0302] As another example, the setting for whether to perform CQI correction may be a quality measurement value (e.g., RSRP, RSRQ, SINR, or CSI) for a wireless channel including PDSCH BLER performance or an AI / ML model performance monitoring result value and a threshold value set between the base station and the terminal, based on which the terminal may determine whether to perform CQI correction. Here, the threshold value may be provided to the terminal by a configuration of a base station upper layer (e.g., RRC, MAC CE). The terminal may determine whether the current CQI report and the PDSCH scheduling based thereon are reliably transmitted in the channel environment of the terminal based on at least one of a quality measurement value (e.g., RSRP, RSRQ, SINR, or CSI) for a wireless channel including PDSCH BLER performance or an AI / ML model performance monitoring result value. Here, the terminal may perform the above-described judgment based on the determined threshold value. As an example, the threshold value may be one of a PDSCH BLER error rate, RSRP / RSRQ / SINR / CSI value. As another example, a threshold value may be set as the criterion for AI / ML model evaluation performance. For example, if the current AI / ML model evaluation performance / results are worse than the model evaluation-related threshold value, the terminal may trigger an operation to perform CQI correction. The terminal operation for the aforementioned CQI correction may be controlled based on the threshold value settings provided by the base station, and the base station (network) may control the terminal operation for the corresponding CQI correction.

[0303]

[0304] As another example, one or more CQI tables may be considered for CQI correction and calculation. For example, the CQI value may be determined based on at least one of the CQI, SINR, and MCS information for PDSCH based on one of the multiple CQI tables. Accordingly, additional configuration information may be provided to the terminal regarding which CQI table is applied for the proposed CQI correction and what the target error probability is. For example, if a parameter indicating a CQI table value is set, the terminal may apply a target error probability of 0.1 or 0.0001 based on the parameter setting for the corresponding CQI table value. For example, Tables 18 to 21 below may be CQI tables for 64 / 256 / 1024 QAM modulation, respectively, and may apply a target error probability of 0.1. On the other hand, Table 21 may be a CQI table for a higher target error probability of 0.00001 for URLLC service support. The terminal may select the highest CQI index that satisfies the error probability based on the CQI values ​​determined based on the CQI tables in Tables 18 to 21 and the aforementioned criteria, and report this to the base station. The following items are not limited to the aforementioned criteria and may also be applied to other cases.

[0305] Here, the above-described methods may be equally applicable to the methods of FIGS. 14 to 23 below, and may not be limited to a specific form. However, for convenience of explanation, the following description focuses on a method for correcting CQI values.

[0306] [Table 18]

[0307]

[0308]

[0309] [Table 19]

[0310]

[0311]

[0312] [Table 20]

[0313]

[0314]

[0315]

[0316] [Table 21]

[0317]

[0318]

[0319] FIG. 14 is a diagram illustrating a method for a terminal applied to the present disclosure to determine and report a CQI value for an AI / ML model. Referring to FIG. 14, a terminal (1410) and a base station (1420) can share configuration information related to whether to apply CQI compensation based on the signaling and procedure for the AI / ML model function of FIG. 13. For example, the base station (1420) can provide CQI compensation information to the terminal (1410). The base station (1420) can provide information related to CQI compensation to the terminal (1410) through an RRC reconfiguration procedure, as described above. In addition, as shown in FIG. 14, the compensation information for CQI compensation can be transmitted after Steps 1 to 6, or can be transmitted naturally within the process of Steps 1 to 6 (the following methods are also the same).

[0320] For example, the above-described procedure may be performed based on FIG. 13 or may be set by the base station while the AI / ML model is being applied and used, and is not limited to a specific form.

[0321] The terminal (1410) can measure target CSI information corresponding to original channel information based on a reference signal (e.g., CSI-RS for CSI measurement, SSB (synchronization signal block)) transmitted by the base station (or network, 1420). In addition, the base station (1420) can provide auxiliary information for CQI correction to the terminal (1410). The terminal (1410) can determine a final CQI value through a CQI value calculated through target CSI information, which is channel information actually measured by the terminal based on the CSI-RS transmitted from the base station (1420), and auxiliary information provided to the terminal (1410) from the base station (1420). Thereafter, the terminal (1410) can report the final CQI value to the base station (1420) by including it in the AI / ML-based CSI report. The base station (1420) can report channel information (e.g., RI / PMI or channel eigenvector and the CQI) corresponding to the output value of the CSI reconstruction part. It can be decoded and restored. Thereafter, the base station (1420) can perform data scheduling for the PDSCH to be transmitted to the corresponding terminal (1320). Through this, the base station (1420) can finally determine at least one of the most suitable optimal MCS, TBS, beamforming, and power and perform PDSCH transmission to the terminal. The terminal can receive the PDSCH transmitted through the above-described method, decode the PDSCH channel, and then perform HARQ (hybrid automatic repeat and request)-ACK reporting. For example, the CQI compensation information in FIG. 14 may be the auxiliary information described above and may not be limited to the name.

[0322] FIG. 15 is a diagram illustrating a method for a terminal applied to the present disclosure to determine and report a CQI for an AI / ML model.

[0323] Referring to FIG. 15, the terminal (1510) can correct the CQI value by reflecting the auxiliary information derived by the terminal to the CQI value determined based on the actual channel information (Target CSI) measured by the terminal. FIG. 15 shows that, unlike FIG. 14, the terminal (1510) can dynamically receive recovered (restored) CSI (Recovered CSI) information, which is an output value of the CSI reconstruction part of the base station, from the base station (1520). In FIG. 15, based on FIG. 13 described above, the terminal (1510) and the base station (1520) can share configuration information regarding the terminal capability for CQI correction. Thereafter, an AI / ML-based CSI compression method can be performed between the base station (1520) and the terminal (1510) for CSI reporting, and a CQI correction method can be triggered (or enabled) to solve the CQI mismatch problem. For example, in FIG. 15, the base station (1520) can transmit the reconstructed CSI corresponding to the output value of the CSI reconstruction part to the terminal (1510) as compensation information. The above-described signaling can be immediately utilized in a situation where a CQI mismatch occurs. That is, the base station (1520) can provide the reconstructed CSI information to the terminal (1510) as compensation information. The terminal (1510) can determine (or derive) a value for CQI compensation based on the reconstructed CSI information transmitted by the base station (1520). The terminal (1510) can generate a compensated CQI value using the determined CQI compensation value and the original CQI value. Here, the compensated CQI value can be used as an input of the CSI generation part on the terminal side, and the compensated CQI value can be included in an AI / ML-based CSI report together with an output generated through inference and reported to the base station (1520). The base station (1520) can restore the received CQI value through the CSI reconstruction part.The restored CQI value may be a value in which a CQI correction value is reflected in the CQI value for the original channel measured by the terminal (1510), and the base station (1520) may perform subsequent PDSCH scheduling based on this. Thereafter, the base station (1520) may perform a scheduled PDSCH transmission to the corresponding terminal, and the terminal (1510) may decode the received PDSCH and report a HARQ-ACK value thereafter.

[0324] That is, in FIG. 14, the base station (1520) derived a value for CQI correction and transmitted it to the terminal (1510), but in FIG. 15, the terminal (1510) can determine a value for CQI correction based on the reconstructed CSI information received from the base station (1520). The terminal (1520) can directly determine a corrected CQI value by combining the CQI correction value determined based on the above and the target CSI information measured through reception of a reference signal (e.g., CSI-RS). The terminal (1510) can include the determined CQI value in the AI / ML CSI report and transmit it to the base station (1530), and the base station (1520) can utilize the CQI value obtained from the terminal (1510) for subsequent PDSCH scheduling and transmission.

[0325] FIG. 16 is a diagram illustrating a CQI determination and reporting method for an AI / ML model applied to the present disclosure. Referring to FIG. 16, a terminal (1610) can report a CQI value determined based on actual channel information (Target CSI) measured by the terminal to a base station (1620). The base station (1620) can perform final correction on the CQI value using CQI assistance information derived by the base station. For example, in FIG. 16, the base station can directly calculate a CQI compensation value and determine a final CQI value by considering the CQI mismatching problem. In FIGS. 14 and 15, the terminal is the entity performing the CQI compensation, while in FIG. 16, the base station can calculate the CQI compensation value and directly determine the final CQI value.

[0326] For example, the terminal (1610) and the base station (1620) can be configured (or activated) to perform two-side AI / ML CSI compression based on models located in the base station and the terminal, respectively, for AI / ML model-based CSI compression operation based on the above-described FIG. 13. In addition, the terminal (1610) and the base station (1620) can apply the same configuration method as described above for determining whether to perform CQI correction. Here, the base station (1620) can transmit a CSI-RS, and the terminal (1610) can measure the target CSI with actual original channel information using the CSI transmission resource and the CSI reference resource. For example, the terminal (1610) can transmit the target CSI information to the base station (1620) through an AI / ML-based CSI report or a non-AI / ML-based CSI report, and is not limited to a specific form. The base station (1620) can determine (or derive) compensation information by comparing the received target CSI information with previously received and restored CSI information. The base station (1620) can apply the determined (or derived) compensation information to the CQI value restored by the CSI reconstruction part of the base station. Thereafter, the base station (1620) can perform PDSCH scheduling based on the CQI value to which the compensation information has been applied, and the base station (1620) can perform PDSCH transmission to the corresponding terminal (1610) based on the information. Here, the terminal (1610) can perform HARQ-ACK reporting for the received PDSCH to transmit and receive data.

[0327] As described above, the base station (1620) and the terminal (1610) can compensate for the CQI value caused by mismatch due to potential channel information differences, and the compensation can be performed at the base station (1620). That is, the terminal (1610) reports the measured actual channel information (Target CSI) to the base station (1620), and the base station (1620) can calculate the final CQI value based on the reported target CSI information and previously derived auxiliary information. For example, the base station (1620) can obtain the target CSI information from the terminal (1610) in advance, and can obtain compensation information by analyzing the difference between the target CSI information and the CSI information reconstructed through the CSI reconstruction part of the base station. The base station (1620) can directly derive the final corrected CQI value by applying the correction information obtained as described above to the CQI value reported by the terminal (1610).

[0328]

[0329] FIG. 17 is a diagram illustrating a CQI determination and reporting method for an AI / ML model applied to the present disclosure. Referring to FIG. 17, a terminal (1710) can independently determine (or derive) compensation information for CQI compensation without signaling from a base station (1720) regarding compensation information or related information. Therefore, in FIG. 17, signaling for transmitting compensation information may not be required, unlike FIGS. 14 and 15 . The terminal (1710) can perform channel measurement and performance monitoring for at least one of a plurality of reference signals, CSI measurement resources, CSI reference resources, and data channels / control channels transmitted from the base station (1720). Through this, the terminal (1710) can evaluate (or estimate) whether there is a mismatch in the CQI value generated by the CSI generation part and reported to the base station.

[0330] As another example, when an AI / ML model performance monitoring evaluation procedure is performed, the base station (1720) and the terminal (1710) can determine whether the CSI compression / recover operation by the AI / ML models within the procedure well reflects the channel environment based on the monitoring type. As a specific example, the first type of monitoring type may be a method in which at least one of model change, activation, and deactivation is determined by the base station. On the other hand, the second type of monitoring type may be a method in which at least one of model change, activation, and deactivation is determined by the terminal. Here, when considering a method of performing CQI correction based on AI / ML model performance evaluation, the entity performing the CQI correction may differ depending on which performance evaluation type is applied. For example, when the monitoring type is the first type, correction information for the CQI value may be determined by the base station, and when the monitoring type is the second type, correction information for the CQI value may be determined by the terminal. However, the present invention may not be limited thereto.

[0331] As another example, if a situation regarding model change and CQI mismatch is detected by the base station or terminal regardless of the monitoring type described above, a CQI correction operation can be performed based on the detected information. For example, a CQI mismatch can be detected through an AI / ML performance evaluation by the base station or terminal. Here, the entity that detected the CQI mismatch can derive and apply a CQI correction value on its own. As a specific example, if a terminal detects a CQI mismatch based on an AI / ML performance evaluation, the terminal can derive and apply CQI correction information based on the detected CQI mismatch.

[0332] Consequently, the entity deriving the CQI correction information may be determined based on at least one of signaling overhead, accuracy for CQI correction, improvement in data transmission / reception performance, and CSI compression performance, and is not limited to a specific embodiment.

[0333]

[0334] As another example, the model for the CSI reconstruction part of the base station can be applied to the training collaboration type on the terminal side. For example, the terminal can download the AI / ML model corresponding to the CSI reconstruction part located on the base station side in the same manner. In addition, the terminal can implement a proxy model with a structure similar to the CSI reconstruction part located on the base station side, and the proxy model can be used for the purpose of CQI calculation and correction. For example, the terminal can have a proxy model that is compatible with or has a similar form to the CSI reconstruction part on the base station side. In addition, the terminal can download the proxy model based on model transfer / delivery by the base station. This method can clearly determine the CQI value, but may have the disadvantage of requiring tighter and closer engineering work between the base station manufacturer and the terminal manufacturer. In another example, the terminal can utilize the proxy model corresponding to the AI / ML model for at least one of model training, model inference, and model monitoring purposes other than CQI calculation, and may not be limited to a specific form.

[0335] Figure 18 is a diagram illustrating a CQI determination and reporting method for an AI / ML model applicable to the present disclosure. Referring to Figure 18, a terminal (1810) may have a proxy model corresponding to the CSI reconstruction part of the base station. The terminal can directly derive CSI output values ​​through the proxy model and determine CQI values ​​based on the CSI output values.

[0336] That is, in FIG. 18, the terminal (1810) can avoid the CQI mismatch problem through a proxy model corresponding to the CSI reconstruction part of the base station. Here, if the proxy model is identical or similar to the CSI reconstruction part of the base station, the output value of the CSI reconstruction part can be obtained through the proxy model of the terminal, and the terminal can resolve the CQI mismatch issue related to CQI determination through the CSI output value of the proxy model. For example, referring to FIG. 18, the terminal can provide information about the channel measured by the terminal as an input value of the CSI generation part for CSI compression, and perform an inference operation based on the information to derive the output value of the CSI generation part. Here, the CSI output value can be utilized as an input of the proxy model located on the terminal side, and based on the information, the terminal can directly acquire (restore) the CQI value, which is an output value obtained through the inference procedure of the CSI reconstruction part. In addition, the terminal can utilize additional auxiliary information to determine the final corrected CQI value, and it can be included in the AI / ML-based CSI feedback information and transmitted to the base station.

[0337] However, the proxy model of the terminal (1810) may have different aspects from the actual CSI reconstruction part model of the base station based on various situations or conditions. Considering the above, the base station may signal necessary auxiliary information to the terminal. The terminal may use the auxiliary information signaled from the base station for CQI determination, as shown in FIG. 18. For example, the auxiliary information may be information based on the difference between the proxy model of the terminal and the model corresponding to the actual CSI reconstruction part of the base station, and may be information that offsets the difference as much as possible. For example, the auxiliary information may include at least one of network additional condition setting information, reference model structure information, related parameter value information, related setting information, and algorithm difference information, but may not be limited thereto. For example, if the proxy model of the terminal is different from the CSI reconstruction model of the base station, the terminal (1810) and the base station may perform an adjustment operation for the difference. The base station can signal model-related information (e.g., model ID, pairing ID, model structure and parameter information, etc.) to the terminal as auxiliary information to enable verification of the CSI reconstruction part. When the terminal receives the auxiliary information from the base station, the terminal can configure the proxy model to be maximally compatible with the model of the base station's CSI reconstruction part using the received auxiliary information. As another example, the terminal can utilize the auxiliary information transmitted from the base station to perform correction on the final CQI value and report the corrected CQI value to the base station by including it in an AI / ML-based CSI report.

[0338] FIG. 19 is a diagram illustrating a method for deriving a CQI value based on a proxy model and auxiliary information provided in a terminal applied to the present disclosure.

[0339] Fig. 19 shows the signaling and its procedures between the base station (1920) and the terminal. As described above, the terminal (1910) can utilize the internal proxy model corresponding to the CSI reconstruction part model of the base station (1920) to use the output value of the CSI reconstruction part for the LCM procedure for CQI correction and determination or other additional purposes. Here, through the signaling procedures such as the UE capabilities and the RRC reconfiguration procedure by the base station (1920) through steps 1 to 6 described above, the base station (1920) and the terminal (1910) can know in advance whether the application of Fig. 19 is possible. In particular, it is assumed that the presence of a proxy model on the terminal (1910) side and related configuration information are mostly shared with the terminal (1910) through a training procedure or the like in advance. However, due to the combination of channel environment and multiple different AI / ML models, continuous adjustment work for the proxy model will be required, and this adjustment work is performed by the base station (1920) to ensure the reliability of the proxy model of the terminal (1910) through configuration information such as Assistance information as shown in the above figure. Thereafter, the terminal (1910) corrects the CQI value by using the output value of its proxy model and the Assistance information together to resolve the CQI mismatch issue. The corrected CQI value is included in the AI / ML CSI report and input to the model corresponding to the CSI reconstruction part located in the base station (1920). As a result, the CQI value is restored through the base station (1920) AI / ML model, and based on the corrected CQI value, the proposed method 2-1 can be utilized for the data transmission and reception procedure by performing the subsequent PDSCH data scheduling and HARQ-ACK feedback procedure.

[0340] The present invention proposes using the corresponding CSI reconstruction part output value as an input value for CQI determination. For example, the CQI value can be calculated based on channel information, which is the CSI reconstruction part output value. If the proxy model output (the CSI reconstruction part output value on the terminal (1910) side) and the CSI reconstruction part output value on the base station (1920) side are almost identical, the CQI value calculated by the terminal (1910) can be directly applied on the base station (1920) for PDSCH scheduling (link adaptation) of the terminal (1910). Otherwise, as described above, additional auxiliary information from the base station (1920) is utilized to generate and correct the final CQI value.

[0341] As another example, the base station (1920) performs precoded CSI-RS transmission based on the reconstructed CSI information, which is the CSI reconstruction part output value, to assist the CQI calculation of the terminal (1910). This is a method of transmitting the CSI-RS transmission based on the restored CSI information corresponding to the CSI reconstruction part output value of the base station (1920) to the terminal (1910) in the form of auxiliary information. The terminal (1910) receives the precoded CSI-RS and performs CSI measurement. In the process, the terminal (1910) calculates the CQI value through the CSI-RS, thereby calculating the CQI based on the channel information corresponding to the output of the CSI reconstruction part of the base station (1920). This means that it is assumed that the channel information corresponding to the CSI reconstruction part output derived from the base station (1920) and the channel information acquired through the CSI-RS transmitted based on the channel information are almost similar. Under such assumptions, the CQI values ​​corrected and derived by the terminal (1910) are reported to the base station (1920) through a CSI reporting method related to the inference procedure of the CSI generation model for AI / ML CSI compression. Thereafter, the CQI values ​​are restored through the CSI reconstruction part, similar to other methods. The restored CQI values ​​are then utilized for PDSCH scheduling and data transmission and reception operations are performed through the HARQ-ACK feedback procedure.

[0342] FIG. 20 is a diagram illustrating a method for transmitting a precoded CSI-RS based on target CSI information of a terminal acquired in advance by a base station applied to the present disclosure.

[0343] Referring to FIG. 20, the base station (2020) can obtain the output value of the CSI reconstruction part through the CSI previously received from the terminal (2010). Thereafter, the base station (2020) performs Precoded CSI-RS transmission to the terminal (2010) based on the CSI information corresponding to the output value of the CSI reconstruction part. The terminal (2010) receives the Precoded CSI-RS transmission and performs CSI measurement to derive channel information corresponding to PMI / RI / CRI or channel eigenvector. Based on the channel information, the terminal (2010) calculates a CQI value and then reports the CSI information (including CQI) set in the AI / ML-based CSI report to the base station (2020). The base station (2020) restores the CQI value through the reported AI / ML-based CSI information and utilizes it to perform data scheduling such as PDSCH. This method is characterized by performing precoded CSI-RS transmission to the terminal (2010) based on output value information from a CSI reconstruction part that has been performed in advance. Accordingly, the terminal (2010) can also derive channel information on the spatial domain applied by the base station (2020) through the received precoded CSI-RS. Assuming that the understanding of the corresponding channel information between the base station (2020) and the terminal (2010) is the same or similar, it is expected that the CQI value measured and reported by the terminal (2010) will not be overestimated. Through this advantage, the terminal (2010) may not request prior auxiliary information or other operations regarding CQI correction from the base station (2020).

[0344] Fig. 21 is a flowchart illustrating a method for a terminal to derive a CQI value applied to the present disclosure. Referring to Fig. 21, the terminal may determine whether to apply a CQI (channel quality indicator) correction value based on upper layer signaling provided from a base station (S2110). For example, the terminal may determine whether to apply a CQI correction value based on the RRC reconfiguration procedure of Fig. 13 described above, but may not be limited thereto. The terminal and the base station may each be equipped with an AI / ML model, and the AI / ML model equipped in the terminal may correspond to a CSI generation part, and the AI / ML model equipped in the base station may correspond to a CSI reconfiguration part, as described above. After that, the terminal can receive the CSI-RS from the base station. (S2120) The terminal can transmit to the base station an AI / ML-based CSI report including the CQI determined based on the CSI-RS received from the base station and the CSI output information generated based on the CSI generation part. (S2130) For example, the terminal can receive CQI assistance information from the base station and apply the CQI assistance information to the CQI determined based on the CSI-RS to generate a corrected CQI. Thereafter, the terminal can transmit to the base station an AI / ML-based CSI report including the corrected CQI.

[0345] As another example, a terminal can receive reconstructed CSI as auxiliary information from the CSI reconstruction part of a base station. The terminal can directly generate CQI auxiliary information from the reconstructed CSI and apply it to the CQI determined based on the CSI-RS to generate a corrected CQI. The terminal can then transmit an AI / ML-based CSI report containing the corrected CQI to the base station.

[0346] As another example, a base station can directly determine CQI assistance information based on the CSI generation part of the AI / ML-based CSI report received from the terminal, and directly apply this information to correct the CQI received from the terminal. In other words, the base station can directly correct the CQI value.

[0347] As another example, a terminal can directly generate CQI assistance information without receiving assistance information from a base station. The terminal can apply the CQI assistance information to the CQI determined based on the CSI-RS to generate a corrected CQI. The terminal can then include the corrected CQI in an AI / ML-based CSI report and transmit it to the base station.

[0348] As another example, the terminal may further include a proxy model corresponding to the CSI reconstruction part of the base station, generate CQI assistance information based on the proxy model, apply the CQI assistance information to the CQI determined based on the CSI-RS, and transmit the corrected CQI to the base station by including it in the AI / ML-based CSI report, as described above.

[0349]

[0350] Figure 22 is a drawing showing a base station device and a terminal device to which the present disclosure can be applied.

[0351] The base station device (2200) may include a processor (2220), an antenna unit (2212), a transceiver (2226), and a memory (2216).

[0352] The processor (2220) performs baseband-related signal processing and may include a higher layer processing unit (2230) and a physical layer processing unit (2240). The higher layer processing unit (2230) may process operations of a MAC (Medium Access Control) layer, an RRC (Radio Resource Control) layer, or higher layers. The physical layer processing unit (2240) may process operations of a physical (PHY) layer (e.g., uplink reception signal processing, downlink transmission signal processing). In addition to performing baseband-related signal processing, the processor (2220) may also control the overall operation of the base station device (2200).

[0353] The antenna unit (2212) may include one or more physical antennas, and when it includes multiple antennas, it may support MIMO (Multiple Input Multiple Output) transmission and reception. In addition, it may support beamforming.

[0354] The memory (2216) can store information processed by the processor (2220), software related to the operation of the base station device (2200), an operating system, applications, etc., and may also include components such as a buffer.

[0355] The processor (2220) of the base station (2200) may be configured to implement the operations of the base station in the embodiments described in the present invention.

[0356] The terminal device (2250) may include a processor (2270), an antenna unit (2262), a transceiver (2264), and a memory (2266). For example, the terminal device (2250) in the present invention may communicate with a base station device (2200). As another example, the terminal device (2250) in the present invention may perform sidelink communication with another terminal device. That is, the terminal device (2250) in the present invention refers to a device that can communicate with at least one of the base station device (2200) and another terminal device, and is not limited to communication with a specific device.

[0357] The processor (2270) performs baseband-related signal processing and may include a higher layer processing unit (2280) and a physical layer processing unit (2290). The higher layer processing unit (2280) may process operations of the MAC layer, the RRC layer, or higher layers. The physical layer processing unit (2290) may process operations of the PHY layer (e.g., downlink reception signal processing, uplink transmission signal processing). In addition to performing baseband-related signal processing, the processor (2270) may also control the overall operation of the terminal device (2250).

[0358] The antenna unit (2262) may include one or more physical antennas, and when it includes multiple antennas, it may support MIMO transmission and reception. In addition, it may support beamforming.

[0359] The memory (2266) can store information processed by the processor (2270), software related to the operation of the terminal device (2250), an operating system, applications, etc., and may also include components such as a buffer.

[0360] A terminal device (2250) according to an example of the present invention may be associated with a vehicle. For example, the terminal device (2250) may be integrated into, positioned in, or located on the vehicle. Furthermore, the terminal device (2250) according to the present invention may be the vehicle itself. Furthermore, the terminal device (2250) according to the present invention may be at least one of a wearable terminal, an AV / VR, an IoT terminal, a robot terminal, and a public safety terminal. The terminal device (2250) to which the present invention is applicable may include any type of communication device that supports interactive services utilizing sidelink for services such as Internet access, service execution, navigation, real-time information, autonomous driving, and safety and risk diagnosis. Furthermore, any type of communication device capable of sidelink operation or an AR / VR device or a sensor that performs relay operation may be included.

[0361] Here, the vehicles to which the present invention is applied may include autonomous vehicles, semi-autonomous vehicles, non-autonomous vehicles, etc. Meanwhile, while the terminal device (2250) according to one example of the present invention is described as being associated with a vehicle, one or more of the UEs may not be associated with a vehicle. This is merely an example, and should not be construed as limiting the application of the present invention to the described example.

[0362] In addition, the terminal device (2250) according to an example of the present invention can determine whether to apply a CQI (channel quality indicator) correction value based on upper layer signaling provided from the base station device (2200). For example, the terminal device (2250) can determine whether to apply a CQI correction value based on the RRC reconfiguration procedure of FIG. 13 described above, but may not be limited thereto. Each of the terminal device (2250) and the base station device (2220) may be equipped with an AI / ML model, and the AI / ML model equipped in the terminal device (2250) may correspond to a CSI generation part, and the AI / ML model equipped in the base station device (2220) may correspond to a CSI reconfiguration part, as described above. Thereafter, the terminal device (2250) may receive a CSI-RS from the base station device (2220). The terminal device (2250) can transmit an AI / ML-based CSI report including CQI determined based on CSI-RS received from the base station device (2220) and CSI output information generated based on a CSI generation part to the base station device (2220). For example, the terminal device (2250) can receive CQI assistance information from the base station device (2220) and apply the CQI assistance information to the CQI determined based on the CSI-RS to generate a corrected CQI. Thereafter, the terminal device (2250) can transmit an AI / ML-based CSI report including the corrected CQI to the base station device (2200). As another example, the terminal device (2250) can receive the reconstructed CSI from the CSI reconstruction part of the base station device (2200) as assistance information. The terminal device (2250) can directly generate CQI assistance information from the reconstructed CSI and apply it to the CQI determined based on the CSI-RS to generate a corrected CQI. Thereafter, the terminal device (2250) can transmit an AI / ML-based CSI report including the corrected CQI to the base station device (2200).As another example, the base station device (2200) can directly determine CQI assistance information through CSI output information based on the CSI generation part in the AI / ML-based CSI report received from the terminal device (2250), and can directly apply it to correct the CQI received from the terminal device (2250). That is, the base station device (2200) can directly correct the CQI value. As another example, the terminal device (2250) can directly generate CQI assistance information without receiving assistance information from the base station device (2200). The terminal device (2250) can apply the CQI assistance information to the CQI determined based on the CSI-RS to generate a corrected CQI. Thereafter, the terminal device (2250) can include the corrected CQI in the AI / ML-based CSI report and transmit it to the base station device (2200). As another example, the terminal device (2250) may further include a proxy model corresponding to the CSI reconstruction part of the base station device (2200), generate CQI assistance information based on the proxy model, apply the CQI assistance information to the CQI determined based on the CSI-RS, and transmit the corrected CQI to the base station device (2200) by including it in the AI / ML-based CSI report, as described above.

[0363] Additionally, various embodiments of the present disclosure may be implemented by hardware, firmware, software, or a combination thereof. In the case of hardware implementation, the embodiments may be implemented by one or more Application Specific Integrated Circuits (ASICs), Digital Signal Processors (DSPs), Digital Signal Processing Devices (DSPDs), Programmable Logic Devices (PLDs), Field Programmable Gate Arrays (FPGAs), general processors, controllers, microcontrollers, microprocessors, etc.

[0364] The scope of the present disclosure includes software or machine-executable instructions (e.g., operating systems, applications, firmware, programs, etc.) that cause operations according to the methods of various embodiments to be executed on a device or a computer, and a non-transitory computer-readable medium having such software or instructions stored thereon and executable on the device or computer.

[0365] The various embodiments of the present disclosure are not intended to list all possible combinations but rather to illustrate representative aspects of the present disclosure, and the matters described in the various embodiments may be applied independently or in combination of two or more.

[0366]

[0367] The above may also apply to other systems.

Claims

1. In terms of method, A step for determining whether to apply a CQI (channel quality indicator) correction value based on upper layer signaling provided by a base station, wherein each of the terminal and the base station is equipped with an AI (artificial intelligence) / ML (machine learning) model, wherein the AI / ML model equipped in the terminal corresponds to a CSI generation part, and the AI / ML model equipped in the base station corresponds to a CSI reconstruction part; A step of receiving a CSI-RS (channel state information-reference signal) from the base station; and A method comprising the step of transmitting, to the base station, an AI / ML-based CSI report including CQI determined based on the CSI-RS and CSI output information based on the CSI generation part.

2. In paragraph 1, A method in which the terminal receives CQI assistance information from the base station, applies the CQI assistance information to the CQI determined based on the CSI-RS, and transmits the corrected CQI to the base station by including it in the AI / ML-based CSI report.

3. In paragraph 1, A method in which the terminal receives reconstructed CSI from the CSI reconstruction part of the base station as auxiliary information, generates CQI auxiliary information from the reconstructed CSI, applies the corrected CQI to the CQI determined based on the CSI-RS, and transmits the corrected CQI to the base station by including it in the AI / ML-based CSI report.

4. In paragraph 1, A method in which the base station directly determines CQI assistance information through the CSI output information based on the CSI generation part in the AI / ML based CSI report received from the terminal, thereby correcting the CQI received from the terminal.

5. In paragraph 1, A method in which the terminal directly generates CQI assistance information without receiving assistance information from a base station, applies the CQI assistance information to a CQI determined based on the CSI-RS, and transmits the corrected CQI to the base station by including it in the AI / ML-based CSI report.

6. In paragraph 1, A method wherein the terminal further comprises a proxy model corresponding to the CSI reconstruction part of the base station, generates CQI assistance information based on the proxy model, applies the CQI assistance information to the CQI determined based on the CSI-RS, and transmits the corrected CQI to the base station by including it in the AI / ML-based CSI report.

7. For wireless devices, at least one processor; and A memory storing instructions that cause the wireless user device to perform a specific operation by the at least one processor, The above specific actions are: Whether to apply a CQI (channel quality indicator) correction value is determined based on upper layer signaling provided from a base station, wherein each of the terminal and the base station is equipped with an AI (artificial intelligence) / ML (machine learning) model, wherein the AI / ML model equipped in the terminal corresponds to a CSI generation part, and the AI / ML model equipped in the base station corresponds to a CSI reconstruction part, Receive a CSI-RS (channel state information-reference signal) from the above base station, and A device that transmits to the base station an AI / ML-based CSI report including CQI determined based on the CSI-RS and CSI output information based on the CSI generation part.

Images