Method and apparatus for performing communication in wireless communication system

Abstract

A method according to various embodiments and an apparatus therefor are disclosed. The apparatus may transmit a reference signal using at least one far-field beam selected from a far-field beam set that is based on a first beamforming matrix, receive channel state information of a partial channel corresponding to the at least one far-field beam, the channel state information being measured on the basis of the reference signal, input the channel state information of the partial channel and information related to the first beamforming matrix into an artificial neural network-based prediction model, obtain channel state information of an entire channel corresponding to the far-field beam set, the channel state information being predicted by the prediction model, and transmit, on the basis of the channel state information of the entire channel, a first signal beamformed with a near-field beam or a far-field beam.

Description

Method for performing communication in a wireless communication system and device for the same

[0001] This relates to a method for a terminal to perform communication in a wireless communication system and a device for doing so.

[0002] A wireless communication system is a multiple access system that supports communication with multiple users by sharing available system resources (e.g., bandwidth, transmission power, etc.). Examples of multiple access systems include CDMA (code division multiple access), FDMA (frequency division multiple access), TDMA (time division multiple access), OFDMA (orthogonal frequency division multiple access), SC-FDMA (single carrier frequency division multiple access), and MC-FDMA (multi carrier frequency division multiple access) systems.

[0003] As more communication devices require larger communication capacities, the need for enhanced mobile broadband communication compared to existing radio access technology (RAT) is emerging. Furthermore, Massive Machine Type Communications (MTC), which connects multiple devices and objects to provide various services anytime and anywhere, is also one of the major issues to be considered in next-generation communication. In addition, communication system designs that consider services / terminals sensitive to reliability and latency are being discussed. Thus, the introduction of next-generation radio access technology considering enhanced mobile broadband communication, massive MTC, and Ultra-Reliable and Low Latency Communication (URLC) is being discussed, and for convenience, this technology is referred to as new RAT or NR in this invention.

[0004] The technical objective to be achieved is to provide a method for transmitting and receiving signals more accurately and efficiently.

[0005] The technical problems are not limited to those mentioned above, and other technical problems not mentioned will be clearly understood by those skilled in the art to which the present invention belongs from the description below.

[0006] A method by a base station according to one aspect may include: transmitting a reference signal using at least one far-field beam selected within a far-field beam set based on a first beamforming matrix; receiving channel state information of a partial channel corresponding to the at least one far-field beam measured based on the reference signal from a UE (user equipment); inputting the channel state information of the partial channel and information related to the first beamforming matrix into a prediction model based on an artificial neural network, and obtaining channel state information of the entire channel corresponding to the far-field beam set predicted by the prediction model; and transmitting a first signal beamformed into a near-field beam or a far-field beam based on the channel state information of the entire channel to the UE.

[0007] Alternatively, the channel state information of the entire channel is characterized by including a precoding matrix and a second beamforming matrix for the entire channel.

[0008] Alternatively, the second beamforming matrix is ​​characterized as being a hybrid analog beamforming matrix that simultaneously supports near-field beamforming and far-field beamforming.

[0009] Alternatively, the channel state information of the entire channel is characterized by being predicted based on the PMI (precoding matrix indicator) included in the channel state information of the partial channel.

[0010] Alternatively, the channel state information of the partial channel is characterized by including only the PMI (precoding matrix indicator) calculated for at least one far beam based on the reference signal.

[0011] Alternatively, the base station selects M far-field beams from among N far-field beams included in the far-field beam set as the at least one far-field beam, and is characterized in that M is smaller than N.

[0012] Alternatively, the prediction model is characterized as being a model trained to predict channel information of the entire channel based on the spatial and frequency characteristic patterns of the channel measurement information of the partial channel.

[0013] Alternatively, the above reference signal is characterized as being a CSI-RS (Channel State Information - Reference Signal) or an SSB (Synchronization Signal Block).

[0014] Depending on other aspects, a non-transient computer-readable storage medium may be provided that records instructions for performing the method by the base station described above.

[0015] Depending on another aspect, a base station that performs the method described above may be provided.

[0016] Depending on another aspect, a processing device for controlling a base station that performs the method described above may be provided.

[0017] A method by a UE (user equipment) according to another aspect comprises: receiving a reference signal transmitted using at least one far-beam selected within a far-beam set; transmitting channel state information of a partial channel corresponding to the at least one far-beam measured based on the reference signal; and receiving a first signal transmitted through a far-beam or a near-beam determined based on channel state information of the entire channel; wherein the channel state information of the entire channel may be channel state information for all far-beams included in the far-beam set predicted based on the channel state information of the partial channel and a first beamforming matrix applied to the at least one far-beam using a prediction model based on an artificial neural network.

[0018] Depending on another aspect, a UE that performs the method described above may be provided.

[0019] Various embodiments enable the UE to perform communication with the base station more accurately and efficiently.

[0020] The effects obtainable from various embodiments are not limited to those mentioned above, and other unmentioned effects will be clearly understood by those skilled in the art from the description below.

[0021] The drawings attached to this specification are intended to provide an understanding of the present invention, to illustrate various embodiments of the invention, and to explain the principles of the invention together with the description in the specification.

[0022] Figure 1 shows the structure of an LTE system.

[0023] Figure 2 shows the structure of the NR system.

[0024] Figure 3 shows the structure of a wireless frame of NR.

[0025] Figure 4 shows the slot structure of an NR frame.

[0026] FIG. 5 is a diagram illustrating physical channels that can be used in various embodiments and a signal transmission method using them.

[0027] FIG. 6 is a diagram showing the structure of an SSB (Synchronization Signal Block) to which various examples of the present disclosure are applicable.

[0028] FIG. 7 is a diagram illustrating an example of a transmission method of an SSB to which various examples of the present disclosure are applicable.

[0029] FIG. 8 illustrates multi-beam transmission of an SSB applicable to various examples of the present disclosure.

[0030] Figure 10 is a diagram illustrating the concept of AI / ML / Deep learning.

[0031] FIGS. 11 to 14 illustrate various AI / ML models of deep learning.

[0032] Figure 15 is a diagram illustrating segmented AI inference.

[0033] Figure 16 is a diagram illustrating the framework for 3GPP RAN Intelligence.

[0034] Figure 17 is a diagram illustrating a method for performing beam prediction based on AI / ML.

[0035] FIGS. 18 to 21 are diagrams illustrating a method for a base station to transmit a data stream to a UE based on a hybrid beamforming system.

[0036] FIG. 22 is a diagram illustrating the performance of a beam setting or determination method based on a hybrid beamforming system.

[0037] Figure 23 is a diagram illustrating how a base station transmits data to a UE.

[0038] FIG. 24 illustrates a communication system to which the present invention is applied.

[0039] FIG. 25 illustrates a wireless device that can be applied to the present invention.

[0040] FIG. 26 shows another example of a wireless device to which the present invention is applied.

[0041] A wireless communication system is a multiple access system that supports communication with multiple users by sharing available system resources (e.g., bandwidth, transmission power, etc.). Examples of multiple access systems include CDMA (code division multiple access), FDMA (frequency division multiple access), TDMA (time division multiple access), OFDMA (orthogonal frequency division multiple access), SC-FDMA (single carrier frequency division multiple access), and MC-FDMA (multi carrier frequency division multiple access) systems.

[0042] Sidelink refers to a communication method in which User Equipment (UE) establishes a direct link to directly exchange voice or data between terminals without passing through a Base Station (BS). Sidelink is being considered as a solution to address the burden on base stations caused by rapidly increasing data traffic.

[0043] V2X (vehicle-to-everything) refers to a communication technology that exchanges information with other vehicles, pedestrians, and infrastructure-equipped objects through wired or wireless communication. V2X can be classified into four types: V2V (vehicle-to-vehicle), V2I (vehicle-to-infrastructure), V2N (vehicle-to-network), and V2P (vehicle-to-pedestrian). V2X communication can be provided through PC5 interfaces and / or Uu interfaces.

[0044] Meanwhile, as more communication devices require larger communication capacities, the need for improved mobile broadband communication compared to existing Radio Access Technology (RAT) is emerging. Accordingly, communication systems considering services or terminals sensitive to reliability and latency are being discussed; next-generation radio access technology that incorporates improved mobile broadband communication, Massive MTC, and URLLC (Ultra-Reliable and Low Latency Communication) can be referred to as new radio access technology (new RAT) or new radio (NR). Vehicle-to-everything (V2X) communication can also be supported in NR.

[0045] The following technologies can be used in various wireless communication systems such as CDMA (code division multiple access), FDMA (frequency division multiple access), TDMA (time division multiple access), OFDMA (orthogonal frequency division multiple access), and SC-FDMA (single carrier frequency division multiple access). CDMA can be implemented using wireless technologies such as UTRA (universal terrestrial radio access) or CDMA2000. TDMA can be implemented using wireless technologies such as GSM (global system for mobile communications), GPRS (general packet radio service), and EDGE (enhanced data rates for GSM evolution). OFDMA can be implemented using wireless technologies such as IEEE (institute of electrical and electronics engineers) 802.11 (Wi-Fi), IEEE 802.16 (WiMAX), IEEE 802-20, and E-UTRA (evolved UTRA). IEEE 802.16m is an evolution of IEEE 802.16e and provides backward compatibility with systems based on IEEE 802.16e. UTRA is part of UMTS (universal mobile telecommunications system). 3GPP (3rd generation partnership project) LTE (long term evolution) is part of E-UMTS (evolved UMTS) which uses E-UTRA (evolved-UMTS terrestrial radio access), employing OFDMA in the downlink and SC-FDMA in the uplink.LTE-A (advanced) is an evolution of 3GPP LTE.

[0046] 5G NR is a successor technology to LTE-A and is a new clean-slate type mobile communication system with characteristics such as high performance, low latency, and high availability. 5G NR can utilize all available spectrum resources, ranging from low frequency bands below 1 GHz to mid-frequency bands from 1 GHz to 10 GHz, and high frequency (millimeter wave) bands above 24 GHz.

[0047] For clarity of explanation, the description focuses on LTE-A or 5G NR, but the technical concept of the embodiment(s) is not limited thereto.

[0048] Figure 1 shows the structure of an applicable LTE system. This can be called an E-UTRAN (Evolved-UMTS Terrestrial Radio Access Network), or an LTE (Long Term Evolution) / LTE-A system.

[0049] Referring to FIG. 1, the E-UTRAN includes a base station (20; Base Station, BS) that provides a control plane and a user plane to a terminal (10). The terminal (10) may be fixed or mobile and may be referred to by other terms such as MS (Mobile Station), UT (User Terminal), SS (Subscriber Station), MT (Mobile Terminal), or Wireless Device. The base station (20) refers to a fixed station that communicates with the terminal (10) and may be referred to by other terms such as eNB (evolved-NodeB), BTS (Base Transceiver System), or Access Point.

[0050] Base stations (20) can be connected to each other through an X2 interface. The base station (20) is connected to the EPC (Evolved Packet Core, 30) through the S1 interface, more specifically to the MME (Mobility Management Entity) through the S1-MME and to the S-GW (Serving Gateway) through the S1-U.

[0051] The EPC (30) is configured with an MME, an S-GW, and a P-GW (Packet Data Network-Gateway). The MME holds information regarding the terminal's connection information or capabilities, and this information is primarily used for managing the terminal's mobility. The S-GW is a gateway with an E-UTRAN as its endpoint, and the P-GW is a gateway with a PDN as its endpoint.

[0052] The layers of the Radio Interface Protocol between a terminal and a network can be classified into L1 (Layer 1), L2 (Layer 2), and L3 (Layer 3) based on the lower three layers of the Open System Interconnection (OSI) model, which is widely known in communication systems. Among these, the Physical Layer, belonging to Layer 1, provides Information Transfer Services using a physical channel, while the Radio Resource Control (RRC) layer, located at Layer 3, performs the role of controlling radio resources between the terminal and the network. To this end, the RRC layer exchanges RRC messages between the terminal and the base station.

[0053] Figure 2 shows the structure of the NR system.

[0054] Referring to FIG. 2, the NG-RAN may include gNBs and / or eNBs that provide user plane and control plane protocol termination to terminals. The gNBs and eNBs are connected to each other via Xn interfaces. The gNBs and eNBs are connected to the 5G Core Network (5GC) via NG interfaces. More specifically, they are connected to the access and mobility management function (AMF) via NG-C interfaces and to the user plane function (UPF) via NG-U interfaces.

[0055] Figure 3 shows the structure of a wireless frame of NR.

[0056] Referring to FIG. 3, radio frames can be used for uplink and downlink transmission in NR. The radio frame has a length of 10 ms and can be defined as two 5 ms half-frames (HF). A half-frame may contain five 1 ms subframes (SF). A subframe may be divided into one or more slots, and the number of slots within a subframe may be determined by the subcarrier spacing (SCS). Each slot may contain 12 or 14 OFDM(A) symbols according to the cyclic prefix (CP).

[0057] When normal CP is used, each slot may contain 14 symbols. When extended CP is used, each slot may contain 12 symbols. Here, the symbols may include OFDM symbols (or CP-OFDM symbols) and SC-FDMA (Single Carrier - FDMA) symbols (or DFT-s-OFDM (Discrete Fourier Transform-spread-OFDM) symbols).

[0058] Table 1 below shows the number of symbols per slot (N) according to the SCS setting (u) when normal CP is used. slot symb ), number of slots per frame (N frame,u slot ) and the number of slots per subframe((N subframe,u slot ) exemplifies.

[0059] SCS (15*2 u )N slot symb N frame,u slot N subframe,u slot 15KHz (u=0)1410130KHz (u=1)1420260KHz (u=2)14404120KHz (u=3)14808240KHz (u=4)1416016

[0060] Table 2 shows the number of symbols per slot, the number of slots per frame, and the number of slots per subframe according to the SCS when an extended CP is used.

[0061] SCS (15*2 u )N slot symb N frame,u slot N subframe,u slot 60KHz (u=2)12404

[0062] In an NR system, the OFDM(A) numerology (e.g., SCS, CP length, etc.) can be configured differently among multiple cells that are merged into a single terminal. Accordingly, the (absolute time) interval of a time resource (e.g., subframe, slot, or TTI) (collectively referred to as TU (Time Unit) for convenience) composed of the same number of symbols can be configured differently among the merged cells.

[0063] In NR, multiple numerologies or SCSs may be supported to support various 5G services. For example, if the SCS is 15 kHz, a wide area in traditional cellular bands may be supported, and if the SCS is 30 kHz / 60 kHz, dense-urban, lower latency, and wider carrier bandwidth may be supported. If the SCS is 60 kHz or higher, a bandwidth greater than 24.25 GHz may be supported to overcome phase noise.

[0064] The NR frequency band can be defined by two types of frequency ranges. The two types of frequency ranges may be FR1 and FR2. The numerical values ​​of the frequency ranges may change, for example, as shown in Table 3 below. Among the frequency ranges used in an NR system, FR1 may mean "sub 6GHz range" and FR2 may mean "above 6GHz range" and may be referred to as millimeter wave (mmW).

[0065] Frequency Range designationCorresponding frequency rangeSubcarrier Spacing (SCS)FR1450MHz - 6000MHz15, 30, 60kHzFR224250MHz - 52600MHz60, 120, 240kHz

[0066] As described above, the numerical value of the frequency range of the NR system may change. For example, FR1 may include a band of 410 MHz to 7125 MHz as shown in Table 4 below. That is, FR1 may include a frequency band of 6 GHz (or 5850, 5900, 5925 MHz, etc.) or higher. For example, the frequency band of 6 GHz (or 5850, 5900, 5925 MHz, etc.) or higher included within FR1 may include an unlicensed band. The unlicensed band may be used for various purposes, for example, for communication for vehicles (e.g., autonomous driving).

[0067] Frequency Range designationCorresponding frequency rangeSubcarrier Spacing (SCS)FR1410MHz - 7125MHz15, 30, 60kHzFR224250MHz - 52600MHz60, 120, 240kHz

[0068] Figure 4 shows the slot structure of an NR frame.

[0069] Referring to FIG. 4, a slot contains multiple symbols in the time domain. For example, in the case of a normal CP, one slot may contain 14 symbols, but in the case of an extended CP, one slot may contain 12 symbols. Alternatively, in the case of a normal CP, one slot may contain 7 symbols, but in the case of an extended CP, one slot may contain 6 symbols.

[0070] A carrier includes multiple subcarriers in the frequency domain. A Resource Block (RB) can be defined as multiple (e.g., 12) consecutive subcarriers in the frequency domain. A Bandwidth Part (BWP) can be defined as multiple consecutive (P)RBs ((Physical) Resource Blocks) in the frequency domain and can correspond to a single numerology (e.g., SCS, CP length, etc.). A carrier can include up to N (e.g., 5) BWPs. Data communication can be performed through the active BWPs. Each element can be referred to as a Resource Element (RE) in a resource grid and can be mapped to a single complex symbol.

[0071] Meanwhile, a wireless interface between terminals or a wireless interface between a terminal and a network may be composed of L1, L2, and L3 layers. In various embodiments of the present disclosure, L1 layer may refer to the physical layer. Additionally, for example, L2 layer may refer to at least one of the MAC layer, RLC layer, PDCP layer, and SDAP layer. Additionally, for example, L3 layer may refer to the RRC layer.

[0072] Bandwidth part (BWP)

[0073] NR systems can support up to 400 MHz per component carrier (CC). If a terminal operating within such a wideband CC keeps the RF for the entire CC constantly turned on, it may lead to high battery consumption. Alternatively, considering various use cases (e.g., eMBB, URLLC, Mmtc, V2X, etc.) operating within a single wideband CC, different numerologies (e.g., sub-carrier spacing) may be supported for different frequency bands within that CC. Furthermore, the capability regarding maximum bandwidth may vary by terminal. Taking this into account, the base station may instruct the terminal to operate only within a portion of the wideband CC's bandwidth rather than the entire bandwidth; for convenience, this portion is defined as the Bandwidth Part (BWP). A BWP can be composed of consecutive resource blocks (RBs) along the frequency axis and can correspond to a single numerology (e.g., sub-carrier spacing, CP length, slot / mini-slot duration).

[0074] Meanwhile, the base station can configure multiple BWPs within a single CC configured for the terminal. For example, a BWP occupying a relatively small frequency range can be configured in the PDCCH monitoring slot, and the PDSCH indicated by the PDCCH can be scheduled on a larger BWP. Alternatively, if UEs are concentrated on a specific BWP, some UEs can be assigned to a different BWP for load balancing. Or, considering frequency domain inter-cell interference cancellation between neighboring cells, a portion of the spectrum in the middle of the total bandwidth can be excluded, and both BWPs can be configured within the same slot. That is, the base station can configure at least one DL / UL BWP for a terminal associated with a wideband CC, and at a specific time, it can activate at least one of the configured DL / UL BWP(s) (by L1 signaling, MAC CE, RRC signaling, etc.), and a switch to another configured DL / UL BWP can be instructed (by L1 signaling, MAC CE, RRC signaling, etc.), or a switch to a designated DL / UL BWP can be made based on a timer when the timer value expires. In this case, the activated DL / UL BWP is defined as the active DL / UL BWP. However, in situations such as when the terminal is in the initial access process or before the RRC connection is set up, the configuration for the DL / UL BWP may not be received; in such situations, the DL / UL BWP assumed by the terminal is defined as the initial active DL / UL BWP.

[0075] FIG. 5 is a diagram illustrating physical channels that can be used in various embodiments and a signal transmission method using them.

[0076] Referring to FIG. 5, when a terminal is turned on again from a powered-off state or newly enters a cell, it performs an initial cell search operation, such as synchronizing with a base station, in step S11. To this end, the terminal receives a Synchronization Signal Block (SSB) from the base station. The SSB includes a Primary Synchronization Signal (PSS), a Secondary Synchronization Signal (SSS), and a Physical Broadcast Channel (PBCH). Based on the PSS / SSS, the terminal synchronizes with the base station and obtains information such as a cell ID. Additionally, the terminal can obtain in-cell broadcast information based on the PBCH. Meanwhile, during the initial cell search phase, the terminal can receive a Downlink Reference Signal (DL RS) to check the downlink channel status.

[0077] After completing the initial cell search, the terminal can obtain more specific system information by receiving the Physical Downlink Control Channel (PDCCH) and the Physical Downlink Control Channel (PDSCH) based on the Physical Downlink Control Channel information (S12).

[0078] Subsequently, the terminal may perform a Random Access Procedure to complete the connection to the base station (S13 ~ S16). To this end, the terminal transmits a preamble through a Physical Random Access Channel (PRACH) (S13) and receives a Random Access Response (RAR) for the preamble through a Physical Uplink Control Channel and a corresponding Physical Uplink Shared Channel (S14). The terminal transmits a Physical Uplink Shared Channel (PUSCH) using scheduling information within the RAR (S15) and may perform a Conflict Resolution Procedure, such as receiving a Physical Uplink Control Channel signal and a corresponding Physical Uplink Shared Channel signal (S16).

[0079] Meanwhile, in addition to the random access process performed in 4 steps as described above (4-step RACH, Type-1 random access procedure), if the random access process is performed in 2 steps (2-step RACH, Type-2 random access procedure), S13 / S15 may be performed as a single operation in which the terminal performs transmission (e.g., a transmission operation of message A including a PRACH preamble and / or PUSCH), and S14 / S16 may be performed as a single operation in which the base station performs transmission (e.g., a transmission operation of message B including RAR and / or collision resolution information).

[0080] A terminal that has performed the procedure described above may subsequently perform the reception (S17) of a physical downlink control channel signal and / or a physical downlink shared channel signal and the transmission (S18) of a physical uplink shared channel (PUSCH) signal and / or a physical uplink control channel (PUCCH) signal as a general uplink / downlink signal transmission procedure.

[0081] Control information transmitted by a terminal to a base station is collectively referred to as Uplink Control Information (UCI). UCI includes HARQ-ACK / NACK (Hybrid Automatic Repeat and reQuest Acknowledgement / Negative-ACK), SR (Scheduling Request), CQI (Channel Quality Indication), PMI (Precoding Matrix Indication), RI (Rank Indication) information, etc.

[0082] UCI is generally transmitted periodically via PUCCH, but it may be transmitted via PUSCH if control information and data need to be transmitted simultaneously. Additionally, the terminal may transmit UCI non-periodically via PUSCH in response to network requests or instructions.

[0083] FIG. 6 is a diagram showing the structure of an SSB (Synchronization Signal Block) to which various examples of the present disclosure are applicable.

[0084] The UE can perform cell search, system information acquisition, beam alignment for initial connection, DL measurement, etc. based on the SSB. The SSB is used in combination with the SS / PBCH (Synchronization Signal / Physical Broadcast channel) block.

[0085] Referring to Fig. 6, the SSB consists of the PSS, SSS, and PBCH. The SSB is composed of four consecutive OFDM symbols, and the PSS, PBCH, SSS / PBCH, and PBCH are transmitted for each OFDM symbol. The PSS and SSS each consist of one OFDM symbol and 127 subcarriers, while the PBCH consists of three OFDM symbols and 576 subcarriers. Polar coding and QPSK (Quadrature Phase Shift Keying) are applied to the PBCH. The PBCH consists of a Data RE and a DMRS (Demodulation Reference Signal) RE for each OFDM symbol. There are three DMRS REs for each RB, and three Data REs exist between the DMRS REs.

[0086] Cell search refers to the process in which the UE acquires the time / frequency synchronization of a cell and detects the cell ID (Identifier) ​​of said cell (e.g., Physical layer Cell ID, PCID). PSS is used to detect a cell ID within a cell ID group, and SSS is used to detect a cell ID group. PBCH is used for detecting the SSB (time) index and half-frame detection.

[0087] The cell search process of the UE can be summarized as shown in Table 5 below.

[0088] Type of SignalOperations1 st StepPSS*SS / PBCH block (SSB) symbol timing acquisition*Cell ID detection within a cell ID group (3 hypothesis)2 nd StepSSS*Cell ID group detection (336 hypothesis)3 rd StepPBCH DMRS*SSB index and Half frame (HF) index (Slot and frame boundary detection)4th StepPBCH*time information (80ms, System Frame Number (SFN, SSB index, HF)*Remaining Minimum System Information (RMSI) Control resource set (CORESET) / Search space configuration5 th StepPDCCH and PDSCH* Cell access information* RACH configuration

[0089] There are 336 cell ID groups, and there are 3 cell IDs for each cell ID group. There are a total of 1008 cell IDs. Information regarding the cell ID group to which a cell ID belongs is provided / obtained through the cell's SSS, and information regarding the cell ID among the 336 cells within the cell ID is provided / obtained through the PSS.

[0090] FIG. 7 is a diagram illustrating an example of a transmission method of an SSB to which various examples of the present disclosure are applicable.

[0091] Referring to FIG. 7, the SSB is transmitted periodically in accordance with the SSB periododicity. During initial cell search, the default SSB period assumed by the UE is defined as 20ms. After cell connection, the SSB period can be set by the network (e.g., base station) to one of {5ms, 10ms, 20ms, 40ms, 80ms, 160ms}. An SSB burst set is configured at the beginning of the SSB period. The SSB burst set consists of a 5ms time window (i.e., a half-frame), and the SSB can be transmitted up to L times within the SSB burst set. The maximum number of SSB transmissions L can be given as follows, depending on the carrier frequency band. One slot contains up to 2 SSBs.

[0092] - For frequency range up to 3 GHz, L = 4

[0093] - For frequency range from 3GHz to 6 GHz, L = 8

[0094] - For frequency range from 6 GHz to 52.6 GHz, L = 64

[0095] The time positions of SSB candidates within an SS burst set can be defined according to the SCS as follows. The time positions of SSB candidates are indexed from 0 to L-1 in chronological order within the SSB burst set (i.e., half-frame) (SSB index).

[0096] - Case A : 15 kHz SCS: The index of the starting symbol of a candidate SSB is given as {2, 8} + 14*n. If the carrier frequency is 3 GHz or less, n=0, 1. If the carrier frequency is 3 GHz to 6 GHz, n=0, 1, 2, 3.

[0097] - Case B : 30 kHz SCS: The index of the starting symbol of a candidate SSB is given by {4, 8, 16, 20} + 28*n. If the carrier frequency is 3 GHz or less, n=0. If the carrier frequency is 3 GHz to 6 GHz, n=0 or 1.

[0098] - Case C : 30 kHz SCS: The index of the starting symbol of a candidate SSB is given as {2, 8} + 14*n. If the carrier frequency is 3 GHz or less, n=0, 1. If the carrier frequency is 3 GHz to 6 GHz, n=0, 1, 2, 3.

[0099] - Case D : 120 kHz SCS: The index of the starting symbol of the candidate SSB is given as {4, 8, 16, 20} + 28*n. When the carrier frequency is greater than 6 GHz, n=0, 1, 2, 3, 5, 6, 7, 8, 10, 11, 12, 13, 15, 16, 17, 18.

[0100] - Case E : 240 kHz SCS: The index of the starting symbol of the candidate SSB is given by {8, 12, 16, 20, 32, 36, 40, 44} + 56*n. When the carrier frequency is greater than 6 GHz, n=0, 1, 2, 3, 5, 6, 7, 8.

[0101] FIG. 8 illustrates multi-beam transmission of an SSB applicable to various examples of the present disclosure.

[0102] Beam sweeping refers to a Transmission Reception Point (TRP) (e.g., base station / cell) changing the beam (direction) of a radio signal over time (hereafter, beam and beam direction may be used interchangeably). SSBs can be transmitted periodically using beam sweeping. In this case, the SSB index is implicitly linked to the SSB beam. The SSB beam can be changed on an SSB (index) basis or on an SSB (index) group basis. In the latter case, the SSB beam remains the same within the SSB (index) group. That is, the transmission beam reflection of the SSB is repeated across multiple consecutive SSBs. The maximum number of SSB transmissions L within an SSB burst set has a value of 4, 8, or 64 depending on the frequency band to which the carrier belongs. Therefore, the maximum number of SSB beams within an SSB burst set can also be given as follows, depending on the carrier's frequency band.

[0103] - For frequency range up to 3 GHz, Max number of beams = 4

[0104] - For frequency range from 3GHz to 6 GHz, Max number of beams = 8

[0105] - For frequency range from 6 GHz to 52.6 GHz, Max number of beams = 64

[0106] * If multi-beam transmission is not applied, the number of SSB beams is 1.

[0107] When a UE attempts an initial connection to a base station, the UE can align beams with the base station based on SSBs. For example, the UE performs SSB detection and identifies the best SSB. Subsequently, the UE can send a RACH preamble to the base station using the PRACH resource linked to / corresponding to the index (i.e., beam) of the best SSB. SSBs can also be used to align beams between the base station and the UE even after the initial connection.

[0108] CSI-related operations

[0109] Figure 9 shows an example of a CSI-related procedure.

[0110] The terminal receives configuration information related to CSI from the base station via RRC signaling (710). The configuration information related to CSI may include at least one of CSI-IM (interference management) resource information, CSI measurement configuration information, CSI resource configuration information, CSI-RS resource information, or CSI report configuration information.

[0111] - A CSI-IM resource may be configured for interference measurement (IM) of a terminal. In the time domain, the CSI-IM resource set may be configured periodic, semi-permanent, or non-periodic. The CSI-IM resource may be configured as Zero Power (ZP)-CSI-RS for the terminal. ZP-CSI-RS may be configured separately from Non-Zero Power (NZP)-CSI-RS.

[0112] - UE can assume that the CSI-RS resource(s) for channel measurement set for one CSI reporting and the CSI-IM / NZP CSI-RS resource(s) for interference measurement (when NZP CSI-RS resource(s) are used for interference measurement) have a QCL relationship with respect to 'QCL-TypeD' on a resource-by-resource basis.

[0113] - The CSI resource configuration may include at least one of a CSI-IM resource for interference measurement, an NZP CSI-RS resource for interference measurement, and an NZP CSI-RS resource for channel measurement. The CMR (channel measurement resource) may be an NZP CSI-RS for CSI acquisition, and the IMR (Interference measurement resource) may be an NZP CSI-RS for CSI-IM and IM.

[0114] - CSI-RS may be configured for one or more terminals. Different CSI-RS configurations may be provided for each terminal, or the same CSI-RS configuration may be provided for multiple terminals. CSI-RS may support up to 32 antenna ports. CSI-RS corresponding to N (N is 1 or more) antenna ports may be mapped to N RE positions within a time-frequency unit corresponding to one slot and one RB. If N is 2 or more, N-port CSI-RS may be multiplexed using CDM, FDM, and / or TDM methods. CSI-RS may be mapped to the remaining REs, excluding the REs to which CORESET, DMRS, and SSB are mapped. In the frequency domain, CSI-RS may be configured for the entire bandwidth, a portion of the bandwidth (BWP), or a portion of the bandwidth. CSI-RS may be transmitted at each RB within the configured bandwidth (i.e., density=1), or at every second RB (e.g., even or odd RB) (i.e., density=1 / 2). When CSI-RS is used as a Tracking Reference Signal (TRS), a single-port CSI-RS may be mapped to three subcarriers in each resource block (i.e., density=3). In the time domain, one or more CSI-RS resource sets may be configured for the terminal. Each CSI-RS resource set may include one or more CSI-RS configurations. Each CSI-RS resource set may be configured periodicly, semipersistently, or non-periodically.

[0115] - CSI report configuration may include settings for feedback type, measurement resources, report type, etc. NZP-CSI-RS resource sets may be used for the CSI report configuration of the terminal. NZP-CSI-RS resource sets may be associated with CSI-RS or SSB. Additionally, multiple periodic NZP-CSI-RS resource sets may be configured as TRS resource sets. (i) Feedback types may include Channel Quality Indicator (CQI), Precoding Matrix Indicator (PMI), CRI (CSI-RS Resource Indicator), SSBRI (SSB Resource block Indicator), LI (Layer Indicator), Rank Indicator (RI), Layer 1-Reference Signal Received Strength (RSRP), etc. (ii) Measurement resources may include settings for downlink signals and / or downlink resources for which the terminal performs measurements to determine feedback information. Measurement resources may be set as ZP and / or NZP CSI-RS resource sets associated with CSI reporting settings. NZP CSI-RS resource sets may include CSI-RS sets or SSB sets. For example, L1-RSRP may be measured against CSI-RS sets or against SSB sets. (iii) Report type may include settings for the timing and uplink channel, etc. for which the terminal performs reporting. Report timing may be set to periodic, semi-persistent, or non-periodic. Periodic CSI reporting may be transmitted over PUCCH. Semi-persistent CSI reporting may be transmitted over PUCCH or PUSCH based on MAC CE indicating activation / deactivation. Non-periodic CSI reporting may be indicated by DCI signaling.For example, the CSI request field of an uplink grant can specify one of various report trigger sizes. Non-periodic CSI reports can be transmitted over PUSCH.

[0116] The terminal measures the CSI based on configuration information related to the CSI. The CSI measurement may include a procedure for receiving the CSI-RS (720) and acquiring the CSI (730) by computationing the received CSI-RS.

[0117] The terminal can transmit CSI reports to the base station (740). For CSI reporting, the time and frequency resources available to the UE are controlled by the base station. The channel state information (CSI) may include at least one of a channel quality indicator (CQI), a precoding matrix indicator (PMI), a CSI-RS resource indicator (CRI), an SS / PBCH block resource indicator (SSBRI), a layer indicator (LI), a rank indicator (RI), an L1-RSRP, and / or an L-SINR.

[0118] The time domain behavior of CSI reporting supports periodic, semi-persistent, and aperiodic. i) Periodic CSI reporting is performed on short PUCCH or long PUCCH. The periodicity and slot offset of periodic CSI reporting can be set to RRC; refer to CSI-ReportConfig IE. ii) Semi-periodic (SP) CSI reporting is performed on short PUCCH, long PUCCH, or PUSCH. In the case of SP CSI on short / long PUCCH, the periodicity and slot offset are set to RRC, and CSI reporting is activated / deactivated by separate MAC CE / DCI. In the case of SP CSI on PUSCH, the periodicity of SP CSI reporting is set to RRC, but the slot offset is not set to RRC, and SP CSI reporting is activated / deactivated by DCI (format 0_1). For SP CSI reporting over PUSCH, a separate RNTI (SP-CSI C-RNTI) is used. The timing of the initial CSI report follows the PUSCH time domain allocation value specified in the DCI, while subsequent CSI reporting timing follows the period set by the RRC. DCI format 0_1 ​​includes a CSI request field and can activate / deactivate specific configured SP-CSI trigger states. SP CSI reporting has the same or similar activation / deactivation mechanisms as those used for data transmission over SPS PUSCH.iii) aperiodic CSI reporting is performed on PUSCH and triggered by DCI. In this case, information related to the trigger of aperiodic CSI reporting can be transmitted / instructed / set via MAC-CE. For an AP CSI with AP CSI-RS, the AP CSI-RS timing is set by RRC, and the timing for AP CSI reporting is dynamically controlled by DCI.

[0119] CSI codebooks (e.g., PMI codebooks) defined in NR standards can be broadly classified into Type I codebooks and Type II codebooks. Type I codebooks are primarily targeted at Single User (SU)-MIMO, which supports both high and low orders. Type II codebooks can primarily support MI-MIMO, which supports up to two layers. While Type II codebooks can provide more accurate CSI compared to Type I, this may result in increased signaling overhead. Meanwhile, the Enhanced Type II codebook was introduced to address the CSI overhead drawbacks associated with the existing Type II codebook; it reduces the codebook payload by considering the correlation along the frequency axis.

[0120] CSI reporting via PUSCH can be configured into Part 1 and Part 2. Part 1 has a fixed payload size and is used to identify the number of information bits in Part 2. Part 1 is transmitted in its entirety prior to Part 2.

[0121] - For Type I CSI feedback, Part 1 includes RI (if reported), CRI (if reported), and the CQI of the first code word. Part 2 includes PMI, and when RI > 4, Part 2 includes CQI.

[0122] - For Type II CSI feedback, Part 1 includes indications for the number of non-zero WB amplitude coefficients per layer of RI (if reported), CQI, and Type II CSI. Part 2 includes the PMI of Type II CSI.

[0123] - For Enhanced Type II CSI feedback, Part 1 includes indications of the total number of non-zero WB amplitude coefficients for the RI (if reported), CQI, and total layers of Enhanced Type II CSI. Part 2 includes the PMI of Enhanced Type II CSI.

[0124] In PUSCH, CSI reporting includes two parts, and if the CSI payload to be reported is less than the payload size provided by the PUSCH resources allocated for CSI reporting, the terminal may omit part of Part 2 CSI.

[0125] Meanwhile, semi-persistent CSI reporting performed in PUCCH format 3 or 4 supports Type II CSI feedback, but only Part 1 of Type II CSI feedback.

[0126] QCL (quasi-co location)

[0127] Two antenna ports are quasi-co-located if the channel properties of one antenna port can be inferred from the channel of another antenna port. Channel properties may include one or more of Delay spread, Doppler spread, Frequency / Doppler shift, Average received power, Received Timing / average delay, and Spatial RX parameters.

[0128] A list of multiple TCI-State configurations can be set in the terminal through the upper layer parameter PDSCH-Config. Each TCI-State is associated with a QCL configuration parameter between one or two DL reference signals and the DM-RS port of the PDSCH. The QCL may include qcl-Type1 for the first DL RS and qcl-Type2 for the second DL RS. The QCL type may correspond to one of the following.

[0129] - 'QCL-TypeA': {Doppler shift, Doppler spread, average delay, delay spread}

[0130] - 'QCL-TypeB': {Doppler shift, Doppler spread}

[0131] - 'QCL-TypeC': {Doppler shift, average delay}

[0132] - 'QCL-TypeD': {Spatial Rx parameter}

[0133] Beam Management (BM)

[0134] The BM process is a process for acquiring and maintaining a set of BS (or transmission and reception point (TRP)) and / or UE beams that can be used for downlink (DL) and uplink (UL) transmission / reception, and may include the following processes and terms.

[0135] - Beam measurement: An operation in which a BS or UE measures the characteristics of a received beamforming signal.

[0136] - Beam determination: The operation in which a BS or UE selects its transmit beam (Tx beam) / receive beam (Rx beam).

[0137] - Beam sweeping: An operation that covers a spatial domain using a transmitted and / or received beam for a set time interval in a predetermined manner.

[0138] - Beam report: An operation in which the UE reports information about the beamformed signal based on beam measurements.

[0139] The BM process can be divided into (1) a DL BM process using SSB or CSI-RS and (2) a UL BM process using SRS (sounding reference signal). Additionally, each BM process may include Tx beam sweeping to determine the Tx beam and Rx beam sweeping to determine the Rx beam.

[0140] At this time, the DL BM process may include (1) transmission of beam-formed DL RSs (e.g., CSI-RS or SSB) by the BS and (2) beam reporting by the UE.

[0141] Here, the beam report may include preferred DL RS ID(s) and the corresponding reference signal received power (RSRP). The DL RS ID may be an SSBRI (SSB Resource Indicator) or a CRI (CSI-RS Resource Indicator).

[0142] AI / ML (Artificial intelligence / machine learning)

[0143] With the advancement of AI / ML technology, the node(s) and terminal(s) constituting wireless communication networks are becoming more intelligent and sophisticated. In particular, due to the intelligence of networks and base stations, it is expected that various network / base station decision parameter values ​​(e.g., transmit / receive power of each base station, transmit power of each terminal, precoder / beam of base stations / terminals, time / frequency resource allocation for each terminal, duplex method of each base station, etc.) can be rapidly optimized, derived, and applied according to various environmental parameters (e.g., distribution / location of base stations, distribution / location / material of buildings / furniture, location / movement direction / speed of terminals, weather information, etc.). In line with this trend, many standardization organizations (e.g., 3GPP, O-RAN) are considering its adoption, and research on this is actively underway.

[0144] AI / ML can be easily referred to as deep learning-based artificial intelligence in a narrow sense, but conceptually it is as shown in Fig. 8.

[0145] - Artificial Intelligence: This can refer to any automation where machines can take over tasks that humans would otherwise have to perform.

[0146] - Machine Learning: Machines can learn patterns for decision-making from data on their own without explicitly programming rules.

[0147] - Deep Learning: An AI / ML model based on artificial neural networks in which machines perform everything from feature extraction to judgment from unstructured data in a single step. The algorithm relies on multilayer networks composed of interconnected nodes for feature extraction and transformation, inspired by biological nervous systems, or neural networks. Common deep learning network architectures may include deep neural networks (DNN), recurrent neural networks (RNN), and convolutional neural networks (CNN).

[0148] Classification of AI / ML types based on various criteria

[0149] 1. Offline vs. Online

[0150] (1) Offline Learning: This follows a sequential procedure of database collection, learning, and prediction. In other words, collection and learning are performed offline, and the completed program can be installed on-site and utilized for prediction tasks. This offline learning method is used in most situations. In offline learning, the system does not learn incrementally; instead, learning is performed using all available collected data, and the data is applied to the system without further learning. If learning on new data becomes necessary, learning can be restarted using the entire new dataset.

[0151] (2) Online Learning: Online learning is a method of gradually improving performance by incrementally learning with additional data, utilizing the fact that data available for recent learning is continuously generated through the internet. Learning is performed in real-time in units of specific data (batch) collected online, and as a result, the system can quickly adapt to changing data.

[0152] To build an AI system, only online learning may be used so that learning is performed using only real-time data, or offline learning may be performed using a predetermined dataset, followed by additional learning using real-time data that is subsequently generated (online + offline learning).

[0153] 2. Classification based on AI / ML Framework concepts

[0154] (1) Centralized Learning: Training data collected from multiple different nodes is reported to a centralized node, and all data resources, storage, and learning (e.g., supervised, unsupervised, reinforcement learning) are performed at a single centralized node.

[0155] (2) Federated Learning: A collective AI / ML model is constructed based on data distributed across data owners. Instead of bringing data into the AI / ML model, the AI / ML model is brought as a data source, allowing local nodes / individual devices to collect data and train their own copies of the AI / ML model, thus eliminating the need to report source data to a central node. In Federated learning, the parameters / weights of the AI / ML model simply need to be sent back to a centralized node to support general AI / ML model training. The advantages of Federated learning include increased computational speed and superiority in terms of information security. In other words, the process of uploading personal data to a central server is unnecessary, which can prevent the leakage and misuse of personal information.

[0156] (3) Distributed Learning: This refers to the concept where the machine learning process is extended and distributed across a node cluster. To accelerate the training of AI / ML models, the models are partitioned and shared across multiple nodes operating simultaneously.

[0157] 3. Classification by Learning Method

[0158] (1) Supervised Learning: Supervised learning is a machine learning task that aims to learn a mapping function from input to output given a labeled dataset. The input data is called training data and has known labels or outcomes. Examples of supervised learning include (i) Regression: Linear Regression, Logistic Regression, (ii) Instance-based Algorithms: k-Nearest Neighbor (KNN), (iii) Decision Tree Algorithms: CART, (iv) Support Vector Machines: SVM, (v) Bayesian Algorithms: Naive Bayes, and (vi) Ensemble Algorithms: Extreme Gradient Boosting, Bagging: Random Forest. Supervised learning can be further grouped into regression and classification problems; classification involves predicting labels, while regression involves predicting quantities.

[0159] (2) Unsupervised Learning: A machine learning task that aims to learn a function that explains a hidden structure in unlabeled data. The input data is unlabeled and has no known results. Some examples of unsupervised learning include K-means clustering, principal component analysis (PCA), nonlinear independent component analysis (ICA), and LSTM.

[0160] (3) Reinforcement Learning: In reinforcement learning (RL), agents aim to optimize long-term goals by interacting with the environment through a trial-and-error process; it is goal-oriented learning based on interaction with the environment. Examples of RL algorithms include (i) Q-learning, (ii) Multi-armed bandit learning, (iii) Deep Q Network, State-Action-Reward-State-Action (SARSA), (iv) Temporal Difference Learning, (v) Actor-critic reinforcement learning, (vi) Deep deterministic policy gradient, and (vii) Monte-Carlo tree search. Reinforcement learning can be further grouped into AI / ML model-based reinforcement learning and AI / ML model-free reinforcement learning. Model-based reinforcement learning is an RL algorithm that uses predictive AI / ML models to obtain transition probabilities between states by utilizing various dynamic states of the environment and AI / ML models where these states lead to rewards. Model-free reinforcement learning is an RL algorithm based on values ​​or policies to achieve maximum future rewards; in multi-agent environments / states, it is computationally less complex and does not require an accurate representation of the environment. Meanwhile, RL algorithms can also be classified into value-based RL versus policy-based RL, policy-based RL versus non-policy RL, etc.

[0161] AI / ML models

[0162] Figure 11 illustrates a Feed-Forward Neural Network (FFNN) AI / ML model. Referring to Figure 11, the FFNN AI / ML model includes an input layer, a hidden layer, and an output layer.

[0163] Figure 12 illustrates a Recurrent Neural Network (RNN) AI / ML model. Referring to Figure 12, the RNN AI / ML model is a type of artificial neural network in which hidden nodes are connected by directed edges to form a directed cycle structure; it is an AI / ML model suitable for processing sequentially appearing data such as speech and text. One type of RNN is the Long Short-Term Memory (LSTM), which is a structure in which a cell state is added to the hidden state of the RNN. Specifically, in the LSTM, an input gate, a forget gate, and an output gate are added to the RNN cell, and a cell state is added. In Figure 12, A is the neural network, x t is the input value, h t represents the output value. Here, h t can mean a state value representing the current based on time, and h t-1 can represent the previous state value.

[0164] Figure 13 illustrates a Convolutional Neural Network (CNN) AI / ML model. CNNs are used for two purposes: to reduce the complexity of AI / ML models and to extract useful features, by applying convolution operations commonly used in image processing. Referring to Figure 13, a kernel or filter refers to a unit or structure that applies weights to inputs within a specific range or unit. The kernel (or filter) can be changed through training. A stride refers to the range of movement of the kernel within the input. A feature map refers to the result of applying the kernel to the input. Padding refers to values ​​added to adjust the size of the feature map. Multiple feature maps may be extracted to induce robustness against distortion, alteration, etc. Pooling refers to operations (e.g., max pooling, average pooling) to reduce the size of the feature map by downsampling it.

[0165] Figure 14 illustrates an auto-encoder AI / ML model. Referring to Figure 14, an auto-encoder is a neural network that takes a feature vector x as input and outputs an identical or similar vector x', where the input and output nodes share the same features and is a type of unsupervised learning. Since the auto-encoder reconstructs the input, the output can be referred to as a reconstruction. The loss function can be expressed as shown in Equation 1.

[0166] [Mathematical Formula 1]

[0167]

[0168] Figure 15 is a diagram illustrating segmented AI inference.

[0169] Figure 15 illustrates a case in which, among split AI operations, the Model Inference function is performed in cooperation between an end device such as a UE and a network AI / ML endpoint.

[0170] In addition to the Model Inference function, the Model Training function, Actor, and Data Collection function can each be split into multiple parts depending on the current task and environment, and can be performed through the cooperation of multiple entities.

[0171] For example, computationally intensive and energy-intensive parts may be performed at a network endpoint, while privacy-sensitive and latency-sensitive parts may be performed at an end device. In this case, the end device can execute a task / model up to a specific part / layer from the input data and then transmit intermediated data to the network endpoint. The network endpoint executes the remaining part / layer and provides the inference outputs to one or more devices performing the operation / task.

[0172] The following describes a functional framework for AI operation.

[0173] Below, to provide a more specific explanation of AI (or AI / ML), terms may be defined as follows.

[0174] - Data collection: Data collected from network nodes, management entities, or UEs, etc., as a basis for AI model training, data analysis, and inference.

[0175] - AI Model: A data-driven algorithm that applies AI technology to generate a set of outputs containing predictive information and / or decision parameters based on a set of inputs.

[0176] - AI / ML Training: An online or offline process of training an AI model by learning features and patterns that best represent data and acquire an AI / ML model trained for inference.

[0177] - AI / ML Inference: A process of making predictions or deriving decisions based on collected data and AI models using trained AI models.

[0178] Figure 16 is a diagram illustrating the framework for 3GPP RAN Intelligence.

[0179] Referring to FIG. 16, the data collection function (10) is a function that collects input data and provides processed input data to the model training function (20) and the model inference function (30).

[0180] Examples of input data may include measurements from UEs or other network entities, feedback from actors, and outputs from AI models.

[0181] The Data Collection function (10) performs data preparation based on input data and provides the input data processed through data preparation. Here, the Data Collection function (10) does not perform specific data preparation (e.g., data pre-processing and cleaning, forming and transformation) for each AI algorithm, and can perform data preparation common to AI algorithms.

[0182] After the data preparation process is performed, the Model Training function (10) provides the training data (11) to the Model Training function (20) and provides the inference data (12) to the Model Inference function (30). Here, the training data (11) is the data required as input for the AI ​​Model Training function (20). The inference data (12) is the data required as input for the AI ​​Model Inference function (30).

[0183] The Data Collection function (10) may be performed by a single entity (e.g., UE, RAN node, network node, etc.) but may also be performed by multiple entities. In this case, Training Data (11) and Inference Data (12) from multiple entities may be provided to the Model Training function (20) and Model Inference function (30), respectively.

[0184] The Model Training function (20) is a function that performs AI model training, validation, and testing, which can generate model performance metrics as part of the AI ​​model testing procedure. If necessary, the Model Training function (20) also handles data preparation (e.g., data pre-processing and cleaning, forming and transformation) based on the Training Data (11) provided by the Data Collection function (10).

[0185] Here, Model Deployment / Update (13) is used to initially deploy a trained, validated, and tested AI model to the Model Inference function (30) or to provide an updated model to the Model Inference function (30).

[0186] The Model Inference function (30) is a function that provides AI model inference output (16) (e.g., prediction or decision). The Model Inference function (30) can provide model performance feedback (14) to the Model Training function (20) where applicable. Additionally, the Model Inference function (30) is responsible for data preparation (e.g., data pre-processing and cleaning, forming and transformation) based on the inference data (12) provided by the Data Collection function (10) if necessary.

[0187] Here, Output (16) refers to the inference output of the AI ​​model generated by the Model Inference function (30), and the details of the inference output may vary depending on the use case.

[0188] Model Performance Feedback (14) can be used to monitor the performance of the AI ​​model if available, and this feedback may be omitted.

[0189] The Actor function (40) is a function that receives an output (16) from the Model Inference function (30) and triggers or performs a corresponding operation / action. The Actor function (40) can trigger an operation / action on another entity (e.g., one or more UEs, one or more RAN nodes, one or more network nodes, etc.) or on itself.

[0190] Feedback (15) can be used to derive training data (11) and inference data (12), or to monitor the performance of the AI ​​model, the impact on the network, etc.

[0191] Meanwhile, the definitions of training, validation, and testing in data sets used in AI / ML can be distinguished as follows.

[0192] - Training data: Refers to the data set used to train a model.

[0193] - Validation data: Refers to a data set used to validate a model that has already been trained. In other words, it refers to a data set typically used to prevent overfitting of the training data set.

[0194] Furthermore, it refers to a data set used to select the best among various models learned during the learning process. Therefore, it can also be viewed as a type of learning.

[0195] - Test data: Refers to the data set for final evaluation. This data is unrelated to training.

[0196] In the case of the above data set, generally, if the training set is divided, the training data and validation data within the entire training set can be divided in a ratio of about 8:2 or 7:3, and if the test is included, it can be divided in a ratio of 6:2:2 (training: validation: test).

[0197] Depending on the capability of the AI / ML function between the base station and the terminal, the cooperation level can be defined as follows, and variations resulting from the combination of multiple levels or the separation of any one level are also possible.

[0198] Cat 0a) No collaboration framework: AI / ML algorithms are based on pure implementation and do not require changes to the wireless interface.

[0199] Cat 0b) This level corresponds to a framework that involves a wireless interface modified to fit efficient implementation-based AI / ML algorithms but without cooperation.

[0200] Cat 1) Inter-node support is involved to improve the AI / ML algorithms of each node. This applies when a UE receives support from a gNB (for training, adaptation, etc.) and vice versa. At this level, model exchange between network nodes is not required.

[0201] Cat 2) Joint ML operations can be performed between the UE and gNB. This level requires the exchange of AI / ML model commands or network nodes.

[0202] The functions previously exemplified in FIG. 16 may be implemented in RAN nodes (e.g., base station, TRP, central unit (CU) of the base station, etc.), network nodes, OAM (operation administration maintenance) of the network operator, or UE.

[0203] Alternatively, two or more entities among a RAN, a network node, a network operator's OAM, or a UE may cooperate to implement the functions exemplified in FIG. 16. For example, one entity may perform some of the functions of FIG. 16, and another entity may perform the remaining functions. As such, some of the functions exemplified in FIG. 16 are performed by a single entity (e.g., UE, RAN node, network node, etc.), the transmission / provision of data / information between each function may be omitted. For example, if the Model Training function (20) and the Model Inference function (30) are performed by the same entity, the transmission / provision of Model Deployment / Update (13) and Model Performance Feedback (14) may be omitted.

[0204] Alternatively, any one of the functions exemplified in FIG. 16 may be performed by two or more entities among the RAN, network node, network operator's OAM, or UE in collaboration. This may be referred to as a split AI operation.

[0205] Meanwhile, in 5G / 6G environments, there is a trend to improve system performance by introducing MIMO (massive multiple-input-multiple-output) to support UEs through efficient resource utilization. Massive MIMO-based systems can concentrate signals in a specific direction using tens or hundreds of antennas, but this can significantly increase the amount of channel feedback information reported by the UE to the base station. Considering this, it is important to reduce the overhead generated during the process of transmitting channel state information to the base station.

[0206] In addition, active technological development is underway to improve data and other throughput by applying near-field beamforming in massive MIMO systems and allocating a larger number of ranks to each UE. Near-field beamforming requires measuring beam sets formed according to distance and angle. Therefore, compared to existing far-field beamforming systems that form beam sets according to angle, more beams must be measured, and technological development to reduce the overhead generated during the beam sweeping process is important.

[0207] In this regard, various artificial intelligence (AI)-based communication technologies are currently being developed by standard organizations such as IEEE and 3GPP to improve the communication performance of wireless communication systems. Discussions are underway at 3GPP RAN WG1 (working group 1) regarding the introduction of AI technologies for channel state information feedback enhancement, beam management, and positioning. Among the three use cases mentioned earlier, AI technology for channel state information feedback enhancement is being discussed as a technique to reduce overhead and improve throughput during the transmission of channel state information. Additionally, AI technology for beam management is being discussed as a technique to predict the ID of a beam that provides the maximum RSRP (reference signal received power) while reducing the number of SSBs (synchronization signal blocks) or CSI-RS (channel state information reference signals) transmitted during the process of a base station allocating a beam to a user on a downlink channel.

[0208] Furthermore, to support UEs through near-field beamforming, a technique for measuring near-field beams with minimal beam sweeping overhead may be required. Codebook techniques that hierarchically measure far-field and near-field beam sets can be considered as a method to reduce near-field beam sweeping overhead. However, since such codebook techniques predict the near-field beam after measuring the far-field beam, beam sweeping overhead may actually increase significantly. Additionally, existing standard-based beam prediction techniques may present a problem in that it is difficult to accurately focus the beam on the UE because they select a beam from a fixed set and assign a fixed beam to the UE.

[0209] On the other hand, when calculating the downlink channel value between the UE and the base station by sweeping a small number of beams, it can be difficult to calculate the accurate channel. Conversely, if the entire beam is swept to calculate the accurate channel and feed the calculated channel back to the UE, the overhead for channel feedback can increase significantly.

[0210] Therefore, it may be necessary to develop a new technology capable of predicting, for each UE, the appropriate beamforming weight for a specific field (e.g., near or far) to maximize cell throughput based on low beam sweeping overhead, depending on the channel environment. Below, we describe in detail a technology that reduces both beam sweeping overhead and channel feedback overhead by predicting the optimal near or far beamforming weight and channel for each UE with a small number of beam sweeping cycles.

[0211] Beam Prediction

[0212] Figure 17 is a diagram illustrating a method for performing beam prediction based on AI / ML.

[0213] Regarding conventional technology, AI-based beam management technology being discussed by the 3GPP standardization body can predict k beam IDs that provide the maximum L1-RSRP among the entire narrow beam set. Here, L1-RSRP may refer to SSB-RSRP calculated through SSB-based sweeping operations, or CSI-RSRP calculated through CSI-RS-based sweeping. The base station can predict the optimal beam by sweeping SSB or CSI-RS to the UE, receiving L1-RSRP as feedback, and inputting it into an NW-side AI model.

[0214] For example, referring to FIG. 17 (a), the SSB-RSRP for a wide beam set can be measured by sweeping the SSB through a wide beam during the initial access process, and the L1-RSRP for the entire narrow beam set can be predicted based on the measured SSB-RSRP for the wide beam set. Alternatively, referring to FIG. 17 (b), the CSI-RSRP for a narrow beam set can be measured by sweeping the CSI-RS through a portion of the narrow beam during the beam measurement process, and the L1-RSRP for the entire narrow beam set can be predicted based on the measured CSI-RSRP for the narrow beam set. In FIG. 17 (b), the narrow beam set B sweeping the CSI-RS is not specifically defined and can be defined based on the CSI-RSRP or SSB-RSRP measured during the previous beam sweeping process. For example, the narrow beam set B can be defined / determined based on the CSI-RSRP measured by beam sweeping of the CSI-RS through some narrow beams during the beam measurement process.

[0215] As shown in FIGS. 17 (a) and (b), beam set A and beam set B may be changed according to the network operator's settings and may be determined based on data swept through the beams. Two cases considered in a given scenario related to this may be as follows.

[0216] (1) Case 1 (SSB Sweeping)

[0217] - Beam Set A: Wide Area Beam Set

[0218] - Beam Set B: Full narrow beam set

[0219] (2) Case 2 (CSI-RS Sweeping)

[0220] - Beam Set A: Some narrow beam sets

[0221] - Beam Set B: Full narrow beam set

[0222] In addition, the input and output data of the NW-side AI model in FIG. 17 (a) and (b) may be as follows.

[0223] - Input data: L1-RSRP of Beam Set B

[0224] - Output data: Top-k beam ID of beam set A

[0225] In the case of a method that predicts the beam ID by providing the maximum L1-RSRP through beam sweeping based on a beam set as in the example above, there may be a disadvantage in that it is difficult to accurately focus the beam to the user because a beam is selected from within a fixed beam set and a fixed beam is assigned to the UE.

[0226] In Case 2 of the AI-based beam management technology described above, the beam ID is predicted using CSI-RSRP from the CSIs fed back through CSI-RS sweeping. The CSI feedback information is defined to include PMI (Precoding matrix indicator), RI (rank indicator), CQI (channel quality indicator), CSI-RSRP, etc., and a technology that enables the base station to service the UE using accurate CSI data while reducing the amount of such feedback information may also be important.

[0227] AI-based CSI feedback enhancement technology can predict appropriate channel data that maintains or improves throughput while reducing the overhead incurred when a user transmits channel state information to a base station. In Case 2 of the beam management AI technology described above, CSI data measured through some narrow beams is fed back, and the optimal beam ID is predicted based on CSI-RSRP. However, when CSI data is measured through a small number of beams, there is a disadvantage in that it is difficult to accurately predict the channel value between the UE and the base station.

[0228] In the case of near-field beamforming systems, this problem becomes more severe as the beam set size increases, and precise channel and beam prediction techniques capable of reducing overhead may be required. Below, we propose a precise channel and beam prediction technique that reduces overhead, such as feedback information.

[0229] The proposed method described below aims to provide short-range and long-range beamforming system technology that reduces pilot and channel state information feedback overhead and improves throughput in next-generation wireless communication systems.

[0230] FIGS. 18 to 21 are diagrams illustrating a method for a base station to transmit a data stream to a UE based on a hybrid beamforming system.

[0231] To construct a hybrid beamforming system that provides services to a UE using both near-field and far-field beams, a sub-array-based communication system as illustrated in FIG. 18 may be considered. Referring to FIG. 18, beam sweeping based on CSI-RS can be performed in a system where an RF-chain is assigned to each sub-array.

[0232] Additionally, referring to FIG. 19, N CSI-RS can be mapped to far beams #1 to #N and transmitted sequentially or simultaneously. The proposed method may be one in which a base station sweeps the CSI-RS using some of the far beam sets, predicts the entire channel based on the feedback CSI, and assigns a beamforming weight of the optimal field (e.g., near or far) to each UE.

[0233] Specifically, with reference to FIG. 20, the base station can calculate an analog beamforming matrix and a digital precoding matrix through the following processes (1) to (6), and transmit data to the UE based on the calculated analog beamforming matrix and digital precoding matrix.

[0234] (1) The base station maps CSI-RS to some of the antenna sub-arrays among all antenna sub-arrays and sequentially sweeps or simultaneously transmits the far beams to which CSI-RS is mapped.

[0235] (2) The UE calculates the CSI for the received CSI-RS and reports the calculated CSI to the base station.

[0236] (3) The base station inputs the precoding matrix calculated based on the PMI included in the CSI reported by the UE and the analog beamforming matrix used for CSI-RS beam sweeping (or the weights of the analog beamforming matrix) into the NW-side AI model, and can obtain the following data from the NW-side AI model (hereinafter referred to as the AI ​​model).

[0237] 1) An overall precoding matrix H' that maximizes cell throughput (e.g., a precoding matrix H' for a digital precoder)

[0238] 2) Near-field and far-field beamforming matrix B' that maximizes cell throughput (e.g., near-field and far-field beamforming matrix B' for an analog beamformer)

[0239] (4) The base station maps the data stream to at least one antenna sub-array based on the precoding matrix output from the AI ​​model.

[0240] (5) The ABF (analog beam former) of the base station adjusts / regulates the beam direction based on the analog beamforming matrix output from the AI ​​model.

[0241] (6) The base station transmits a data stream to the UE using the beam generated through the process of (1) to (5).

[0242] For example, as illustrated in FIG. 20, a base station may transmit three CSI-RS to a UE using three far-field beams. In this case, the UE may calculate a 3x3 precoding matrix and a PMI based on the received CSI-RS, and feed back channel state information (CSI) containing the calculated PMI to the base station. The base station may calculate a precoding matrix (3x3 size) based on the feedback PMI. The base station may predict a precoding matrix and an analog beamforming matrix for the entire channel based on the 3x3 precoding matrix corresponding to the feedback PMI. At this time, the prediction of the precoding matrix and the analog beamforming matrix for the entire channel may be performed using a prediction model based on an artificial neural network trained to predict them as described above. The base station can transmit data / signals to the UE using beamformed far beams and / or near beams based on the precoding matrix for the entire predicted channel and the analog beamforming matrix.

[0243] In this case, the UE can receive the transmitted data stream based on the precoding matrix output from the base station's AI model. At this time, the UE can receive the data stream using the receiving beam selected during the initial connection process to the base station. If additional beam refinement is required for the UE's receiving beam, the base station can predict the base station's optimal beam (transmit or receive beam) through the base station's AI model and perform additional beam refinement operations on the UE's receiving beam.

[0244] Alternatively, even if information regarding the digital precoder and analog precoder output from the AI ​​model of the base station is not provided by the base station, the UE may directly predict the digital precoder and analog precoder.

[0245] Furthermore, referring to FIG. 21, the base station can reduce the pilot overhead used for CSI-RS transmission and the overhead used for CSI-RS feedback by using an AI model to perform the following ① and ② processes.

[0246] - ① Predict the entire channel based on channel data measured by sweeping the far beam

[0247] - ② Predict the optimal analog beamforming matrix based on channel conditions / channel environments by considering both near and far beamforming together

[0248] In addition, the input and output data of the AI ​​model can be organized as follows.

[0249] - Input data: Feedbacked PMI for some far beams (or precoding matrix or digital precoder calculated based on the said PMI) and beamforming matrix used for some far beam sweeping (or beamforming matrix weights)

[0250] - Output data: Full precoding matrix (for digital precoder) and / or beamforming matrix considering both near and far beamforming (for analog beamformer)

[0251] Below, the performance increase and overhead reduction according to the proposed method are explained in detail.

[0252] FIG. 22 is a diagram illustrating the performance of a beam setting or determination method based on a hybrid beamforming system.

[0253] Meanwhile, as described above, the base station transmits CSI-RS to the UE via far-field beam sweeping and can receive feedback on some channel data from the UE. After receiving the channel data feedback, the base station can derive a precoding matrix for the entire channel and a beamforming matrix that considers both near-field beamforming and far-field beamforming. In this case, the overhead O occurring during the beam measurement process using CSI-RS BSO It can be defined as shown in the following mathematical formula 2.

[0254] [Mathematical Formula 2]

[0255] O BSO = N SB ХN CSI-RS / N RE

[0256] Here, N SB is the number of beams used for beam sweeping, and N CSI-RS is the number of resource elements used for CSI-RS transmission, and N RE can be the total number of resource elements.

[0257] Figures 22 (a) and 22 (b) illustrate the performance of the proposed method and a method based on a 'near-field beam codebook' that sweeps the existing entire near-field beam set-based CSI-RS in a simulation environment set as shown in Table 6 below.

[0258] Simulation parametersvalueSub-array size of base station (BS)5 by 10, 5 by 50, 5 by 100Number of BS antennas500Number of UE antennas4Carrier frequency150 GHzScan angle (azimuth)[0, 180]Scan angle (elevation)[-90, 90]SNR0 dBNumber of symbols per slot14Number of resource block for CSI-RS25

[0259] Referring to FIG. 22 (a), the change in CSI-RS sweeping overhead with increasing transmission sub-array size is illustrated, and FIG. 21 (b) illustrates the change in combination performance with increasing transmission sub-array size.

[0260] Specifically, the beam width may decrease as the Tx sub-array size increases. In this case, for the 'near-field beam codebook,' the overhead for CSI-RS sweeping increases exponentially as the Tx (base station) sub-array size grows. In contrast, since the proposed method performs CSI-RS sweeping using only the far-field beam, the overhead can be maintained at a constant value even if the sub-array size increases.

[0261] In addition, the total spectral efficiency (η) can be calculated based on the maximum number of supported ranks depending on the channel situation / state, and the combined performance (P) considering overhead and total spectral efficiency together. combined ) can be calculated / derived using the following mathematical formula 3.

[0262] [Mathematical Formula 3]

[0263]

[0264] Referring to Fig. 22 (b), when the sub-array size of the base station is 5 by 100, the proposed method has significantly higher performance in terms of SNR than the near-field beam codebook method. This is because the conventional near-field beam codebook method incurs significant overhead due to the need to measure a large number of beams.

[0265] As such, the proposed invention can maintain or improve throughput while reducing pilot overhead and overhead used for CSI-RS feedback during the CSI-RS transmission and feedback process through beam and channel prediction using a hybrid beamforming technique. Specifically, the proposed invention predicts the entire channel using an AI model based on CSI obtained by sweeping CSI-RS using some beams from a far-field beam set, and effectively predicts the optimal field (near or far) analog beamforming matrix based on the predicted channel, thereby effectively reducing pilot overhead and CSI feedback overhead.

[0266] Figure 23 is a diagram illustrating how a base station transmits data to a UE.

[0267] As described above, the base station may include a prediction model based on AI / ML. The prediction model may be a model pre-trained to predict channel state information for the entire channel through channel state information for a partial channel. Here, the channel state information for a partial channel is channel state information for some of the far-field beams among a plurality of far-field beams included in the far-field beam set for the base station, and the channel state information for the entire channel may be channel state information for all far-field beams included in the far-field beam set. For example, the prediction model may be pre-trained to predict channel state information for the entire channel by analyzing the spatial correlation between the far-field beams (e.g., in the case of adjacent beams, they may share spatially similar patterns and similar paths), spatial continuity and similarity, and channel patterns caused by obstacles, etc., from the channel state information for the partial channel.

[0268] Referring to FIG. 23, a base station can transmit a reference signal using at least one far-field beam selected within a far-field beam set (S231). For example, as described with reference to FIG. 20, the base station can map a reference signal (e.g., CSI-RS (Channel State Information - Reference Signal) or SSB (Synchronization Signal Block)) to only some sub-arrays among a plurality of antenna sub-arrays, and perform a beam sweeping operation for the reference signal using the at least one far-field beam to which a predetermined beamforming matrix (hereinafter, a first beamforming matrix) is applied. For example, the base station can select M far-field beams (where M is smaller than N) from among N far-field beams included in the far-field beam set as the at least one far-field beam, and perform a beam sweeping operation to sequentially transmit the reference signal using the selected at least one far-field beam.

[0269] For example, a base station may select a far-field beam corresponding to a predefined beam index from among a plurality of far-field beams included in a far-field beam set as the at least one far-field beam. Alternatively, the base station may select beams whose beam indices have a predefined multiple relationship based on the number of the plurality of far-field beams as the at least one far-field beam. Alternatively, the prediction model may be pre-trained to output the at least one far-field beam indices that are optimal for predicting the overall channel state information among the plurality of far-field beams. In this case, the base station may select at least one beam corresponding to the far-field beam index output by the prediction model from among the plurality of far-field beams as the at least one far-field beam. Alternatively, the base station may select far-field beams with beam indices having a specific interval from among the plurality of far-field beams as the at least one far-field beam. Here, the specific interval may be determined based on the surrounding channel congestion of the base station or the number of UEs to be serviced. For example, the specific interval may be determined as a relatively larger value as the surrounding channel congestion (or the number of UEs) of the base station increases. In addition, the plurality of far beams and / or at least one far beam may be a far beam formed by applying the first beamforming matrix described above, and the first beamforming matrix may be an analog beamforming matrix for forming far beams (only).

[0270] Next, the base station may receive channel state information of a partial channel corresponding to the at least one far-field beam, measured based on the reference signal, from the UE (user equipment) (S233). For example, the UE may perform a measurement on each of the at least one reference signals transmitted using the at least one far-field beam to calculate channel state information (e.g., partial channel state information) for the partial channel corresponding to the at least one far-field beam, and transmit the channel state information for the partial channel to the base station. Meanwhile, the channel state information of the partial channel may include only a PMI (precoding matrix indicator).

[0271] Next, the base station inputs channel state information of the partial channel and information related to the first beamforming matrix (e.g., weights of the first beamforming matrix applied to at least one far-field beam) into a prediction model based on the artificial neural network, and can obtain channel state information of the entire channel corresponding to the set of far-field beams predicted by the prediction model (S235). For example, the prediction model can predict the channel state information of the entire channel based on the PMI (or a precoding matrix calculated based on the PMI) included in the channel state information of the partial channel and / or the weights of the first beamforming matrix. As described above, the channel state information of the entire channel may include information on an optimal precoding matrix and a beamforming matrix (hereinafter referred to as the second beamforming matrix) that can maximize the throughput of the base station for the entire channel. Additionally, the second beamforming matrix may be an analog beamforming matrix that simultaneously supports near-field beamforming and far-field beamforming.

[0272] For example, a base station may input channel state information of the partial channel into a prediction model based on the artificial neural network. Based on the channel state information of the partial channel, the prediction model may predict channel state information of the entire channel for the entire far-field beam set. For example, based on the channel state information of the partial channel, the prediction model may output a precoding matrix for the entire channel that maximizes the throughput of the base station, and a second beamforming matrix that supports both far-field and near-field beamforming for the entire channel that maximizes the throughput of the base station. In this case, the base station may apply the precoding matrix to a data stream for data to be transmitted to the UE to convert the data stream into an antenna signal, and transmit a first signal in which the second beamforming matrix is ​​applied to the antenna signal. Here, the first signal may be beamformed into a near-field beam and / or a far-field beam by applying the second beamforming matrix and transmitted.

[0273] Next, the base station may transmit a first signal to the UE using a beam-formed far-field beam and / or near-field beam based on channel state information for the entire channel (S237). For example, if the base station obtains channel state information for the entire channel including a precoding matrix and a second beam-forming matrix for the entire channel from the prediction model, the base station may map a data stream for the UE to at least one antenna sub-array based on the precoding matrix to convert the data stream into an antenna signal, and transmit a first signal by applying the second beam-forming matrix to the antenna signal. The first signal may be transmitted to the UE via a near-field beam or a far-field beam through the application of the second beam-forming matrix.

[0274] Thus, the proposed invention can effectively predict the entire channel state information through an AI model trained to acquire the entire channel state information based on partial channel state information, even if only partial channel state information for some beams is acquired by sweeping a reference signal using some beams from a set of far-field beams. Alternatively, since the proposed invention only requires feedback on the partial channel information, it can effectively reduce pilot overhead related to the reference signal and feedback overhead of the channel information. Alternatively, the proposed invention can effectively calculate a hybrid beamforming matrix supporting near-field and far-field beams based on the predicted entire channel state information.

[0275] Example of a communication system to which the invention is applied

[0276] Although not limited thereto, the various descriptions, functions, procedures, proposals, methods, and / or flowcharts of the invention disclosed in this document may be applied to various fields requiring wireless communication / connection (e.g., 5G) between devices.

[0277] Examples are provided in more detail below with reference to the drawings. In the following drawings and descriptions, the same reference numerals may represent the same or corresponding hardware blocks, software blocks, or function blocks unless otherwise described.

[0278] FIG. 24 illustrates a communication system to which the present invention is applied.

[0279] Referring to FIG. 24, the communication system (1) to which the present invention applies includes a wireless device, a base station, and a network. Here, the wireless device refers to a device that performs communication using wireless access technology (e.g., 5G NR (New RAT), LTE (Long Term Evolution)) and may be referred to as a communication / wireless / 5G device. Although not limited thereto, the wireless device may include a robot (100a), a vehicle (100b-1, 100b-2), an XR (eXtended Reality) device (100c), a hand-held device (100d), a home appliance (100e), an IoT (Internet of Thing) device (100f), and an AI device / server (400). For example, the vehicle may include a vehicle equipped with wireless communication functions, an autonomous vehicle, a vehicle capable of performing inter-vehicle communication, etc. Here, the vehicle may include an Unmanned Aerial Vehicle (UAV) (e.g., a drone). XR devices include AR (Augmented Reality) / VR (Virtual Reality) / MR (Mixed Reality) devices and can be implemented in the form of HMDs (Head-Mounted Devices), HUDs (Head-Up Displays) equipped in vehicles, televisions, smartphones, computers, wearable devices, home appliances, digital signage, vehicles, robots, etc. Portable devices may include smartphones, smartpads, wearable devices (e.g., smartwatches, smart glasses), computers (e.g., laptops, etc.). Home appliances may include TVs, refrigerators, washing machines, etc. IoT devices may include sensors, smart meters, etc. For example, base stations and networks may be implemented as wireless devices, and a specific wireless device (200a) may operate as a base station / network node to other wireless devices.

[0280] Wireless devices (100a to 100f) can be connected to a network (300) through a base station (200). Artificial Intelligence (AI) technology may be applied to the wireless devices (100a to 100f), and the wireless devices (100a to 100f) can be connected to an AI server (400) through the network (300). The network (300) can be configured using a 3G network, a 4G (e.g., LTE) network, or a 5G (e.g., NR) network. The wireless devices (100a to 100f) may communicate with each other through the base station (200) / network (300), but they may also communicate directly (e.g., sidelink communication) without going through the base station / network. For example, vehicles (100b-1, 100b-2) can communicate directly (e.g., V2V (Vehicle to Vehicle) / V2X (Vehicle to everything) communication). Also, IoT devices (e.g., sensors) can communicate directly with other IoT devices (e.g., sensors) or other wireless devices (100a to 100f).

[0281] Wireless communication / connection (150a, 150b, 150c) can be established between wireless devices (100a~100f) / base station (200) and base station (200) / base station (200). Here, wireless communication / connection can be achieved through various wireless access technologies (e.g., 5G NR), such as uplink / downlink communication (150a), sidelink communication (150b) (or D2D communication), and inter-base station communication (150c) (e.g., relay, IAB (Integrated Access Backhaul)). Through wireless communication / connection (150a, 150b, 150c), wireless devices and base stations / wireless devices, and base stations and base stations can transmit / receive wireless signals to / from each other. For example, wireless communication / connection (150a, 150b, 150c) can transmit / receive signals through various physical channels. To this end, based on various proposals of the present invention, at least some of the following may be performed: various configuration information setting processes for transmitting / receiving wireless signals, various signal processing processes (e.g., channel encoding / decoding, modulation / demodulation, resource mapping / demapping, etc.), resource allocation processes, etc.

[0282] Example of a wireless device to which the present invention is applied

[0283] FIG. 25 illustrates a wireless device that can be applied to the present invention.

[0284] Referring to FIG. 25, the first wireless device (100) and the second wireless device (200) can transmit and receive wireless signals through various wireless access technologies (e.g., LTE, NR). Here, {the first wireless device (100), the second wireless device (200)} may correspond to {wireless device (100x), base station (200)} and / or {wireless device (100x), wireless device (100x)} of FIG. 24.

[0285] The first wireless device (100) includes one or more processors (102) and one or more memories (104), and may additionally include one or more transceivers (106) and / or one or more antennas (108). The processor (102) controls the memory (104) and / or transceivers (106) and may be configured to implement the descriptions, functions, procedures, proposals, methods and / or flowcharts of operation disclosed in this document. For example, the processor (102) may process information within the memory (104) to generate a first information / signal and then transmit a wireless signal containing the first information / signal through the transceiver (106). Additionally, the processor (102) may receive a wireless signal containing a second information / signal through the transceiver (106) and then store information obtained from the signal processing of the second information / signal in the memory (104). The memory (104) may be connected to the processor (102) and may store various information related to the operation of the processor (102). For example, the memory (104) may store software code containing instructions for performing some or all of the processes controlled by the processor (102) or for performing the descriptions, functions, procedures, proposals, methods, and / or operation sequence diagrams disclosed in this document. Here, the processor (102) and the memory (104) may be part of a communication modem / circuit / chipset designed to implement wireless communication technology (e.g., LTE, NR). The transceiver (106) may be connected to the processor (102) and may transmit and / or receive wireless signals through one or more antennas (108). The transceiver (106) may include a transmitter and / or receiver. The transceiver (106) may be combined with an RF (Radio Frequency) unit. In the present invention, the wireless device may refer to a communication modem / circuit / chipset.

[0286] According to one example, the first wireless device (100) or UE may include a processor (102) connected to the RF transceiver and a memory (104). The memory (104) may include at least one program capable of performing operations related to the embodiments described in FIGS. 16 to 23.

[0287] Specifically, the processor (102) controls the RF transceiver (106) to receive a reference signal transmitted using at least one far-field beam selected within a far-field beam set, transmits channel state information of a partial channel corresponding to the at least one far-field beam measured based on the reference signal, and can receive a first signal transmitted using a far-field beam or a near-field beam determined based on the channel state information of the entire channel. Here, the channel state information of the entire channel may be channel state information for all far-field beams included in the far-field beam set predicted based on the channel state information of the partial channel using a prediction model based on an artificial neural network included in the base station.

[0288] Alternatively, a non-transient computer-readable storage medium may be configured that records instructions for performing the proposed methods described with reference to FIGS. 14 to 23.

[0289] The second wireless device (200) includes one or more processors (202) and one or more memories (204), and may additionally include one or more transceivers (206) and / or one or more antennas (208). The processor (202) controls the memory (204) and / or transceivers (206) and may be configured to implement the descriptions, functions, procedures, proposals, methods and / or operation sequences disclosed in this document. For example, the processor (202) may process information within the memory (204) to generate a third information / signal and then transmit a wireless signal containing the third information / signal through the transceiver (206). Additionally, the processor (202) may receive a wireless signal containing a fourth information / signal through the transceiver (206) and then store information obtained from the signal processing of the fourth information / signal in the memory (204). The memory (204) may be connected to the processor (202) and may store various information related to the operation of the processor (202). For example, the memory (204) may store software code containing instructions for performing some or all of the processes controlled by the processor (202) or for performing the descriptions, functions, procedures, proposals, methods, and / or operation sequence diagrams disclosed in this document. Here, the processor (202) and the memory (204) may be part of a communication modem / circuit / chip designed to implement wireless communication technology (e.g., LTE, NR). The transceiver (206) may be connected to the processor (202) and may transmit and / or receive wireless signals through one or more antennas (208). The transceiver (206) may include a transmitter and / or receiver. The transceiver (206) may be interchangeable with an RF unit. In the present invention, the wireless device may refer to a communication modem / circuit / chip.

[0290] According to one example, the second wireless device or base station (200) may include a processor (202) and a memory (204) connected to the RF transceiver. The memory (204) may include at least one program capable of performing operations related to the embodiments described in FIGS. 16 through 23.

[0291] Specifically, the processor (202) controls the RF transceiver (206) to transmit a reference signal using at least one far-field beam selected within a far-field beam set, receives channel state information of a partial channel corresponding to the at least one far-field beam measured based on the reference signal from the UE (user equipment), inputs the channel state information of the partial channel into a prediction model based on an artificial neural network, obtains channel state information of the entire channel corresponding to the far-field beam set predicted by the prediction model, and transmits a first signal beamformed into a near-field beam or a far-field beam based on the channel state information of the entire channel to the UE.

[0292] Alternatively, a processing device comprising a processor (202) and a memory (204) may be configured. In this case, the processing device comprises at least one processor; and at least one memory connected to the at least one processor and storing instructions, wherein the instructions are executed by the at least one processor and, based on this, the base station (200) may: transmit a reference signal using at least one far-field beam selected within a far-field beam set, receive channel state information of a partial channel corresponding to the at least one far-field beam measured based on the reference signal from a UE (user equipment), input the channel state information of the partial channel into a prediction model based on an artificial neural network, obtain channel state information of a whole channel corresponding to the far-field beam set predicted by the prediction model, and transmit a first signal beamformed into a near-field beam or a far-field beam based on the channel state information of the whole channel to the UE.

[0293] Hereinafter, hardware elements of the wireless device (100, 200) will be described in more detail. Although not limited thereto, one or more protocol layers may be implemented by one or more processors (102, 202). For example, one or more processors (102, 202) may implement one or more layers (e.g., functional layers such as PHY, MAC, RLC, PDCP, RRC, SDAP). One or more processors (102, 202) may generate one or more Protocol Data Units (PDUs) and / or Service Data Units (SDUs) according to the descriptions, functions, procedures, proposals, methods, and / or flowcharts of operation disclosed in this document. One or more processors (102, 202) may generate messages, control information, data, or information according to the descriptions, functions, procedures, proposals, methods, and / or flowcharts of operation disclosed in this document. One or more processors (102, 202) may generate a signal (e.g., baseband signal) containing a PDU, SDU, message, control information, data, or information according to the functions, procedures, proposals, and / or methods disclosed in this document and provide it to one or more transceivers (106, 206). One or more processors (102, 202) may receive a signal (e.g., baseband signal) from one or more transceivers (106, 206) and may obtain a PDU, SDU, message, control information, data, or information according to the descriptions, functions, procedures, proposals, methods, and / or flowcharts disclosed in this document.

[0294] One or more processors (102, 202) may be referred to as a controller, microcontroller, microprocessor, or microcomputer. One or more processors (102, 202) may be implemented by hardware, firmware, software, or a combination thereof. For example, one or more Application Specific Integrated Circuits (ASICs), one or more Digital Signal Processors (DSPs), one or more Digital Signal Processing Devices (DSPDs), one or more Programmable Logic Devices (PLDs), or one or more Field Programmable Gate Arrays (FPGAs) may be included in one or more processors (102, 202). The descriptions, functions, procedures, proposals, methods, and / or flowcharts disclosed in this document may be implemented using firmware or software, and the firmware or software may be implemented to include modules, procedures, functions, etc. Firmware or software configured to perform the descriptions, functions, procedures, proposals, methods, and / or operation sequences disclosed in this document may be contained in one or more processors (102, 202) or stored in one or more memories (104, 204) and driven by one or more processors (102, 202). The descriptions, functions, procedures, proposals, methods, and / or operation sequences disclosed in this document may be implemented using firmware or software in the form of code, instructions, and / or sets of instructions.

[0295] One or more memories (104, 204) may be connected to one or more processors (102, 202) and may store various forms of data, signals, messages, information, programs, code, instructions, and / or commands. One or more memories (104, 204) may be composed of ROM, RAM, EPROM, flash memory, hard drive, registers, cache memory, computer read storage media, and / or combinations thereof. One or more memories (104, 204) may be located inside and / or outside of one or more processors (102, 202). Additionally, one or more memories (104, 204) may be connected to one or more processors (102, 202) through various technologies such as wired or wireless connections.

[0296] One or more transceivers (106, 206) may transmit user data, control information, wireless signals / channels, etc., as mentioned in the methods and / or operation flowcharts, etc., of this document to one or more other devices. One or more transceivers (106, 206) may receive user data, control information, wireless signals / channels, etc., as mentioned in the descriptions, functions, procedures, proposals, methods and / or operation flowcharts, etc., disclosed in this document from one or more other devices. For example, one or more transceivers (106, 206) may be connected to one or more processors (102, 202) and may transmit and receive wireless signals. For example, one or more processors (102, 202) may control one or more transceivers (106, 206) to transmit user data, control information, or wireless signals to one or more other devices. Additionally, one or more processors (102, 202) may control one or more transceivers (106, 206) to receive user data, control information, or wireless signals from one or more other devices. Additionally, one or more transceivers (106, 206) may be connected to one or more antennas (108, 208), and one or more transceivers (106, 206) may be configured to transmit and receive user data, control information, wireless signals / channels, etc., as described in the descriptions, functions, procedures, proposals, methods, and / or flowcharts of operation disclosed in this document through one or more antennas (108, 208). In this document, one or more antennas may be multiple physical antennas or multiple logical antennas (e.g., antenna ports). One or more transceivers (106, 206) can convert the received wireless signal / channel, etc. from an RF band signal to a baseband signal in order to process the received user data, control information, wireless signal / channel, etc. using one or more processors (102, 202).One or more transceivers (106, 206) can convert user data, control information, wireless signals / channels, etc. processed using one or more processors (102, 202) from baseband signals to RF band signals. To this end, one or more transceivers (106, 206) may include (analog) oscillators and / or filters.

[0297] Examples of wireless device applications to which the present invention is applied

[0298] FIG. 26 illustrates another example of a wireless device to which the present invention applies. The wireless device may be implemented in various forms depending on the use-example / service.

[0299] Referring to FIG. 26, the wireless device (100, 200) corresponds to the wireless device (100, 200) of FIG. 25 and may be composed of various elements, components, units / parts, and / or modules. For example, the wireless device (100, 200) may include a communication unit (110), a control unit (120), a memory unit (130), and additional elements (140). The communication unit may include a communication circuit (112) and transceiver(s) (114). For example, the communication circuit (112) may include one or more processors (102, 202) and / or one or more memories (104, 204) of FIG. 26. For example, the transceiver(s) (114) may include one or more transceivers (106, 206) and / or one or more antennas (108, 208) of FIG. 25. The control unit (120) is electrically connected to the communication unit (110), the memory unit (130), and additional elements (140) and controls the general operation of the wireless device. For example, the control unit (120) may control the electrical / mechanical operation of the wireless device based on a program / code / command / information stored in the memory unit (130). Additionally, the control unit (120) may transmit information stored in the memory unit (130) to an external (e.g., another communication device) via a wireless / wired interface through the communication unit (110), or store information received from an external (e.g., another communication device) via a wireless / wired interface through the communication unit (110) in the memory unit (130).

[0300] The additional element (140) can be configured in various ways depending on the type of wireless device. For example, the additional element (140) may include at least one of a power unit / battery, an input / output unit (I / O unit), a driving unit, and a computing unit. Although not limited thereto, the wireless device may be implemented in the form of a robot (Fig. 24, 100a), a vehicle (Fig. 24, 100b-1, 100b-2), an XR device (Fig. 24, 100c), a portable device (Fig. 24, 100d), a home appliance (Fig. 24, 100e), an IoT device (Fig. 24, 100f), a digital broadcasting terminal, a hologram device, a public safety device, an MTC device, a medical device, a fintech device (or financial device), a security device, a climate / environment device, an AI server / device (Fig. 24, 400), a base station (Fig. 24, 200), a network node, etc. Wireless devices can be used in a movable or fixed location depending on the use—e.g., service.

[0301] In FIG. 26, various elements, components, units / parts, and / or modules within the wireless device (100, 200) may be entirely interconnected via a wired interface, or at least a portion may be wirelessly connected via a communication unit (110). For example, within the wireless device (100, 200), the control unit (120) and the communication unit (110) may be wired, and the control unit (120) and the first unit (e.g., 130, 140) may be wirelessly connected via the communication unit (110). Additionally, each element, component, unit / part, and / or module within the wireless device (100, 200) may include one or more additional elements. For example, the control unit (120) may be composed of one or more sets of processors. For example, the control unit (120) may be composed of a set of a communication control processor, an application processor, an Electronic Control Unit (ECU), a graphics processing processor, a memory control processor, etc. As another example, the memory unit (130) may be composed of RAM (Random Access Memory), DRAM (Dynamic RAM), ROM (Read Only Memory), flash memory, volatile memory, non-volatile memory and / or a combination thereof.

[0302] Here, the wireless communication technology implemented in the wireless device (XXX, YYY) of this specification may include LTE, NR, and 6G, as well as Narrowband Internet of Things for low-power communication. For example, NB-IoT technology may be an example of LPWAN (Low Power Wide Area Network) technology and may be implemented according to standards such as LTE Cat NB1 and / or LTE Cat NB2, but is not limited to the names mentioned above. Additionally, or generally, the wireless communication technology implemented in the wireless device (XXX, YYY) of this specification may perform communication based on LTE-M technology. For example, LTE-M technology may be an example of LPWAN technology and may be referred to by various names such as eMTC (enhanced Machine Type Communication). For example, LTE-M technology may be implemented in at least one of various standards such as 1) LTE CAT 0, 2) LTE Cat M1, 3) LTE Cat M2, 4) LTE non-BL (non-Bandwidth Limited), 5) LTE-MTC, 6) LTE Machine Type Communication, and / or 7) LTE M, and is not limited to the names mentioned above. Additionally or generally, wireless communication technology implemented in the wireless device (XXX, YYY) of this specification may include at least one of ZigBee, Bluetooth, and Low Power Wide Area Network (LPWAN) with consideration for low-power communication, and is not limited to the names mentioned above. As an example, ZigBee technology can create personal area networks (PANs) related to small / low-power digital communication based on various standards such as IEEE 802.15.4, and may be referred to by various names.

[0303] The embodiments described above are combinations of the components and features of the present invention in a specific form. Each component or feature should be considered optional unless otherwise explicitly stated. Each component or feature may be implemented in a form not combined with other components or features. Additionally, it is possible to construct embodiments of the present invention by combining some components and / or features. The order of operations described in the embodiments of the present invention may be changed. Some components or features of one embodiment may be included in another embodiment, or may be replaced with corresponding components or features of another embodiment. It is obvious that embodiments may be constructed by combining claims that do not have an explicit citation relationship in the claims, or that new claims may be included by amendment after filing.

[0304] In this document, embodiments of the present invention are described primarily with a focus on the signal transmission and reception relationship between a terminal and a base station. This transmission and reception relationship is extended in the same or similar manner to signal transmission and reception between a terminal and a relay or between a base station and a relay. Specific operations described in this document as being performed by a base station may, in some cases, be performed by an upper node. That is, it is self-evident that various operations performed for communication with a terminal in a network consisting of multiple network nodes including a base station may be performed by the base station or other network nodes other than the base station. The base station may be replaced by terms such as fixed station, Node B, eNode B (eNB), and access point. Additionally, the terminal may be replaced by terms such as User Equipment (UE), Mobile Station (MS), and Mobile Subscriber Station (MSS).

[0305] Embodiments according to the present invention may be implemented by various means, for example, hardware, firmware, software, or a combination thereof. In the case of implementation by hardware, one embodiment of the present invention may be implemented by one or more ASICs (application specific integrated circuits), DSPs (digital signal processors), DSPDs (digital signal processing devices), PLDs (programmable logic devices), FPGAs (field programmable gate arrays), processors, controllers, microcontrollers, microprocessors, etc.

[0306] In the case of implementation by firmware or software, an embodiment of the present invention may be implemented in the form of a module, procedure, function, etc., that performs the functions or operations described above. Software code may be stored in a memory unit and executed by a processor. The memory unit may be located inside or outside the processor and may exchange data with the processor by various means already known.

[0307] It is obvious to those skilled in the art that the present invention may be embodied in other specific forms without departing from the features of the invention. Accordingly, the foregoing detailed description should not be interpreted restrictively in all respects but should be considered exemplary. The scope of the invention shall be determined by a reasonable interpretation of the appended claims, and all modifications within the equivalent scope of the invention are included within the scope of the invention.

[0308] The embodiments of the present invention as described above can be applied to various mobile communication systems.

Claims

1. In the method using a base station, A step of transmitting a reference signal using at least one far beam selected within a far beam set based on a first beamforming matrix; A step of receiving channel state information of a partial channel corresponding to at least one far-field beam measured based on the above reference signal from a UE (user equipment); A step of inputting channel state information of the partial channel and information related to the first beamforming matrix into a prediction model based on an artificial neural network, and obtaining channel state information of the entire channel corresponding to the far-field beam set predicted from the prediction model; and A method comprising the step of transmitting to the UE a first signal beamforming into a near beam or a far beam based on channel state information of the entire channel.

2. In Paragraph 1, A method characterized in that the channel state information of the entire channel includes a precoding matrix and a second beamforming matrix for the entire channel.

3. In Paragraph 2, A method characterized in that the second beamforming matrix is ​​a hybrid analog beamforming matrix that simultaneously supports near-field beamforming and far-field beamforming.

4. In Paragraph 1 A method characterized in that the channel state information of the entire channel is predicted based on the PMI (precoding matrix indicator) included in the channel state information of the partial channel.

5. In Paragraph 1, A method characterized in that the channel state information of the above-mentioned partial channel includes only the PMI (precoding matrix indicator) calculated for at least one far beam based on the above-mentioned reference signal.

6. In Paragraph 1, A method characterized in that the base station selects M far-field beams from among N far-field beams included in the far-field beam set as at least one far-field beam, and M is smaller than N.

7. In Paragraph 1, A method characterized in that the prediction model is a model trained to predict channel information of the entire channel based on spatial and frequency characteristic patterns of channel measurement information of the partial channel.

8. In Paragraph 1, A method characterized in that the above reference signal is a CSI-RS (Channel State Information - Reference Signal) or an SSB (Synchronization Signal Block).

9. A non-transient computer-readable storage medium storing instructions for performing the method described in paragraph 1.

10. Regarding base stations, RF (Radio Frequency) transceiver; and It includes a processor connected to the above RF transceiver, and A base station, wherein the processor controls the RF transceiver to transmit a reference signal using at least one far-field beam selected within a far-field beam set based on a first beamforming matrix, receives channel state information of a partial channel corresponding to the at least one far-field beam measured based on the reference signal from a UE (user equipment), inputs the channel state information of the partial channel and information related to the first beamforming matrix into a prediction model based on an artificial neural network, obtains channel state information of the entire channel corresponding to the far-field beam set predicted by the prediction model, and transmits a first signal beamformed into a near-field beam or a far-field beam based on the channel state information of the entire channel to the UE.

11. In Paragraph 10, A base station characterized by the fact that the channel state information of the entire channel includes a precoding matrix and a second beamforming matrix for the entire channel.

12. In Paragraph 11, A base station characterized in that the second beamforming matrix is ​​a hybrid analog beamforming matrix that simultaneously supports near-field beamforming and far-field beamforming.

13. In a processing device for controlling a base station, At least one processor; and The base station includes at least one memory connected to the at least one processor and storing instructions, wherein the instructions are executed by the at least one processor. A processing device that transmits a reference signal using at least one far-field beam selected within a far-field beam set based on a first beamforming matrix, receives channel state information of a partial channel corresponding to the at least one far-field beam measured based on the reference signal from a UE (user equipment), inputs the channel state information of the partial channel and information related to the first beamforming matrix into a prediction model based on an artificial neural network, obtains channel state information of the entire channel corresponding to the far-field beam set predicted by the prediction model, and transmits a first signal beamformed into a near-field beam or a far-field beam based on the channel state information of the entire channel to the UE.

14. In the method using UE (user equipment), A step of receiving a reference signal transmitted using at least one far beam selected within a far beam set; A step of transmitting channel state information of a partial channel corresponding to at least one far-field beam measured based on the above reference signal; and The method includes the step of receiving a first signal transmitted through a far beam or a near beam determined based on channel state information of the entire channel; A method in which the channel state information of the entire channel is channel state information for all far beams included in the far beam set predicted based on the channel state information of the partial channel and a first beamforming matrix applied to at least one far beam using a prediction model based on an artificial neural network.

15. Regarding UE (user equipment), RF (Radio Frequency) transceiver; and It includes a processor connected to the above RF transceiver, and The processor controls the RF transceiver to receive a reference signal transmitted using at least one far-field beam selected within a far-field beam set, transmits channel state information of a partial channel corresponding to the at least one far-field beam measured based on the reference signal, and receives a first signal transmitted using a far-field beam or a near-field beam determined based on the channel state information of the entire channel. The channel state information of the entire channel is channel state information for all far beams included in the far beam set predicted using an artificial neural network-based prediction model, based on the channel state information of the partial channel and a first beamforming matrix applied to at least one far beam.

Images