Base station, beam transmission method, and mobile communication system

Abstract

To provide a base station, a beam transmission method, and a mobile communication system capable of improving power efficiency. The present invention also provides a base station, a beam transmission method, and a mobile communication system that enable initial access using a certain number of beams or less in a user apparatus.SOLUTION: A base station according to one aspect is a base station in a mobile communication system. The base station includes a receiver configured to receive, from a user apparatus, reception qualities for M (M <N) beams, where M is smaller than N (N> 2) beams that can be transmitted from the base station. An inference unit configured to infer the N received qualities from the M received qualities using a learned AI model; The base station further includes a beam selecting unit configured to select M beams based on the N reception qualities. The base station further includes a transmitter configured to transmit the selected M beams.SELECTED DRAWING: Figure 10

Description

[Technical Field]

[0001] The present invention relates to a base station, a beam transmission method, and a mobile communication system. [Background technology]

[0002] In a mobile communication system, a user equipment can establish a connection with a base station by performing initial access. Specifically, initial access can be performed using a cell search or a random access procedure. By performing a cell search, the user equipment establishes downlink synchronization with a base station (cell) and acquires a cell ID (PCI: Physical Cell ID). Furthermore, the user equipment establishes uplink synchronization with a base station (cell) by performing a random access procedure.

[0003] Meanwhile, base stations use beamforming to transmit synchronization signal blocks (SSB: Synchronization Signal / Physical Broadcast Channel Blocks). Beamforming is a technology that controls the phase and amplitude of radio waves for each antenna to transmit or receive radio waves (beams) in a specific direction. In particular, 5G (5th Generation), a technical specification of 3GPP (registered trademark; the same applies hereinafter), can use higher frequency bands than 4G (4th Generation), so the transmission direction of radio waves is narrowed and propagation loss is compensated for by beam gain.

[0004] In addition, base stations transmit SSBs using beam sweeping. Beam sweeping is a technology that switches the transmission direction of SSBs at predetermined time intervals. Using beam sweeping, base stations transmit each SSB in an SS burst in a time-division manner, with each SSB in a different transmission beam direction. This allows the base station to distribute the SSBs within the SS burst throughout the entire coverage area of ​​the cell.

[0005] The user equipment performs cell search using SSB and selects, for example, the beam with the best reception quality. Then, the user equipment notifies the base station of information about the selected beam in a random access procedure. Thereafter, the user equipment can transmit or receive user data or control data using the selected beam.

[0006] The following technology is available for such mobile communication systems: A base station transmits multiple search beams in different directions, and a terminal ranks the search beams based on the received power of each of the search beams at the terminal, and calculates the estimated direction of the beam from the base station that maximizes the received power at the terminal based on the comparison result between a direction ranked third or lower and each direction higher in the rank. [Prior art documents] [Patent documents]

[0007] [Patent Document 1] JP 2018-11150 A Summary of the Invention [Problem to be solved by the invention]

[0008] The beam used for initial access may not necessarily be transmitted in all directions. For example, if a base station is installed near a structure such as a building, the direction in which user devices are located may be biased toward a specific direction. In such a situation, if the base station transmits beams in all directions by beam sweeping, it will increase the number of SSB transmissions and reduce the power efficiency of the base station.

[0009] The above-described technology does not disclose calculating the estimated direction of a beam using a certain number or fewer search beams when using multiple search beams. Therefore, in the above-described technology, even when there is a bias in the direction where the user equipment exists, the base station transmits beams in all directions, and it is not possible to improve power efficiency.

[0010] Therefore, an object of the present disclosure is to provide a base station, a beam transmission method, and a mobile communication system capable of improving power efficiency. Another object of the present disclosure is to provide a base station, a beam transmission method, and a mobile communication system that enable initial access using a certain number or fewer beams in a user equipment.

Means for Solving the Problems

[0011] The base station according to the first aspect is a base station in a mobile communication system. The base station has a receiving unit that receives, from a user equipment, the reception quality for M (M < N) beams that is less than the number of N (N > 2) beams that can be transmitted from the base station. The base station also has an inference unit that uses a learned AI model to infer the reception quality of N beams from the reception quality of M beams. Further, the base station has a beam selection unit that selects M beams based on the reception quality of N beams. Furthermore, the base station has a transmission unit that transmits the selected M beams.

[0012] The beam transmission method according to the second aspect is a beam transmission method in a mobile communication system. The beam transmission method includes a step in which a base station receives, from a user equipment, the reception quality for M (M < N) beams that is less than the number of N (N > 2) beams that can be transmitted from the base station. The beam transmission method also includes a step in which the base station uses a learned AI model to infer the reception quality of N beams from the reception quality of M beams. Further, the beam transmission method includes a step in which the base station selects M beams based on the reception quality of N beams. Furthermore, the beam transmission method includes a step in which the base station transmits the selected M beams.

[0013] A mobile communication system according to a third aspect is a mobile communication system having a base station and a user device. In the mobile communication system, the base station receives from the user device reception qualities for M (N>M) beams, which is less than the number of N (N>2) beams that can be transmitted from the base station. Also, in the mobile communication system, the base station uses a trained AI model to infer N reception qualities from the M reception qualities. Furthermore, in the mobile communication system, the base station selects M beams based on the N reception qualities. Furthermore, in the mobile communication system, the base station transmits the selected M beams. [Brief explanation of the drawings]

[0014] [Figure 1] FIG. 1 is a diagram illustrating an example of the configuration of a mobile communication system according to the first embodiment. [Figure 2] FIG. 2 is a diagram illustrating an example of the configuration of a protocol stack related to a user plane according to the first embodiment. [Figure 3] FIG. 3 is a diagram illustrating an example of the configuration of a protocol stack related to the control plane according to the first embodiment. [Figure 4] FIG. 4 is a diagram illustrating an example of the configuration of a base station according to the first embodiment. [Figure 5] FIG. 5 is a diagram illustrating an example of the configuration of a UE according to the first embodiment. [Figure 6] FIG. 6 is a diagram for explaining the initial access according to the first embodiment. [Figure 7] FIG. 7 is a diagram for explaining the initial access according to the first embodiment. [Figure 8] FIG. 8 is a diagram illustrating an example of an RSRP map according to the first embodiment. [Figure 9] FIG. 9 is a diagram illustrating an example of operation according to the first embodiment. [Figure 10] FIG. 10 is a diagram illustrating an example of the configuration of a base station according to the first embodiment. [Figure 11]FIG. 11(A) is a diagram showing an example of input data according to the first embodiment, FIG. 11(B) is a diagram showing correct answer data according to the first embodiment, and FIG. 11(C) is a diagram showing an example of predicted data. [Figure 12] FIG. 12 is a diagram illustrating an example of operation according to the first embodiment. [Figure 13] FIG. 13 is a diagram illustrating an example of the operation of the beam selection process according to the first embodiment. [Figure 14] FIG. 14 is a diagram illustrating an example of the configuration of a base station 200 according to the second embodiment. [Figure 15] FIG. 15 is a diagram illustrating an example of operation according to the second embodiment. [Figure 16] FIG. 16 is a diagram illustrating an example of the operation of the beam selection process according to the first embodiment. DETAILED DESCRIPTION OF THE INVENTION

[0015] A mobile communication system according to an embodiment will be described with reference to the drawings. In the description of the drawings, the same or similar parts are denoted by the same or similar reference numerals.

[0016] [First embodiment]

[0017] (1) Example of a mobile communication system configuration 1 is a diagram showing an example of the configuration of a mobile communication system according to this embodiment. The mobile communication system according to this embodiment is a system that complies with the 3GPP standard. For example, the mobile communication system according to this embodiment may be a fifth generation (5G) system or a sixth generation (6G) system.

[0018] The mobile communication system includes a network (NW) 10 and a user equipment (UE) 100. The UE 100 is a mobile communication device that performs wireless communication with the NW 10. The UE 100 may be any device used by a user, and may be, for example, a mobile phone terminal (including a smartphone), a tablet terminal, a notebook PC (Personal Computer), a communication module (including a communication card or a chipset), a sensor or a device provided in a sensor, a vehicle or a device provided in a vehicle (Vehicle UE), or an aircraft or a device provided in an aircraft (Aerial UE).

[0019] The NW 10 includes a radio access network (RAN) 20 and a core network (CN) 30. When the mobile communication system is a 5th generation system (5GS), the RAN 20 is referred to as a next generation radio access network (NG-RAN), and the CN 30 is referred to as a 5G core network (5GC).

[0020] The RAN 20 includes a plurality of base stations 200 (base stations 200a to 200c in the example of FIG. 1). The base stations 200 are connected to each other via an inter-base station interface. The base station 200 is an example of a network node. The base station 200 may be configured (i.e., functionally divided) with a CU (Central Unit) and a DU (Distributed Unit), and the two units may be connected via a fronthaul interface. When the mobile communication system is 5GS, the base station 200 is called a gNB, the inter-base station interface is called an Xn interface, and the fronthaul interface is called an F1 interface.

[0021] Each base station 200 manages one or more cells. The base station 200 performs wireless communication with the UE 100 that has established a connection with its own cell. Each base station 200 has a radio resource management (RRM) function, a user data (also simply referred to as "data") routing function, a measurement control function for mobility control and scheduling, and the like. Note that the term "cell" is used to indicate the smallest unit of a wireless communication area. The term "cell" is also used to indicate a function or resource for performing wireless communication with the UE 100. One cell belongs to one carrier frequency. One downlink component carrier and one uplink component carrier may be associated with one cell. The bandwidth (system bandwidth) corresponding to one cell may be divided into multiple bandwidth parts (BWP: Bandwidth Parts).

[0022] The CN 30 includes a CN (Core Network) device 300. The CN device 300 may include a C-plane device corresponding to the control plane (C-plane) and a U-plane device corresponding to the user plane (U-plane). The C-plane device performs various mobility controls and paging for the UE 100. The C-plane device communicates with the UE 100 using NAS (Non-Access Stratum) signaling. The U-plane device performs data transfer control. When the mobile communication system is 5GS, the C-plane device is called an AMF (Access and Mobility Management Function), the U-plane device is called a UPF (User Plane Function), and the interface between the base station 200 and the CN device 300 is called an NG interface.

[0023] FIG. 2 is a diagram showing an example of the configuration of a protocol stack of a U-plane radio interface that handles data.

[0024] The U-plane radio interface protocol includes, for example, a physical (PHY) layer, a medium access control (MAC) layer, a radio link control (RLC) layer, a packet data convergence protocol (PDCP) layer, and a service data adaptation protocol (SDAP) layer.

[0025] The PHY layer performs encoding / decoding, modulation / demodulation, antenna mapping / demapping, and resource mapping / demapping. Data and control information are transmitted between the PHY layer of the UE 100 and the PHY layer of the base station 200 via a physical channel. The PHY layer of the UE 100 receives downlink control information (DCI) transmitted from the base station 200 on a physical downlink control channel (PDCCH). Specifically, the UE 100 performs blind decoding of the PDCCH using a radio network temporary identifier (RNTI) and acquires successfully decoded DCI as DCI addressed to the UE. The DCI transmitted from the base station 200 has CRC parity bits scrambled by the RNTI added.

[0026] The MAC layer performs data priority control and retransmission processing using Hybrid ARQ (HARQ). Data and control information are transmitted between the MAC layer of UE 100 and the MAC layer of base station 200 via a transport channel. The MAC layer of base station 200 includes a scheduler. The scheduler determines the uplink and downlink transport format (transport block size, modulation and coding scheme (MCS)) and the resources to be allocated to UE 100.

[0027] The RLC layer transmits data to the RLC layer on the receiving side using the functions of the MAC layer and PHY layer. Data and control information are transmitted between the RLC layer of the UE 100 and the RLC layer of the base station 200 via logical channels.

[0028] The PDCP layer performs header compression / decompression, encryption / decryption, etc.

[0029] The SDAP layer maps IP flows, which are units for QoS control by the CN 30, to radio bearers, which are units for QoS control by the AS (Access Stratum). Note that if the RAN is connected to the EPC, the SDAP may not be necessary.

[0030] FIG. 3 is a diagram showing an example of the configuration of a protocol stack of a C-plane radio interface that handles signaling (control signals).

[0031] The protocol stack of the C-plane radio interface includes, for example, a Radio Resource Control (RRC) layer and a Non-Access Stratum (NAS) layer instead of the SDAP layer shown in FIG.

[0032] RRC signaling for various settings is transmitted between the RRC layer of the UE 100 and the RRC layer of the base station 200. The RRC layer controls logical channels, transport channels, and physical channels in accordance with the establishment, re-establishment, and release of radio bearers. When there is a connection (RRC connection) between the RRC of the UE 100 and the RRC of the base station 200, the UE 100 is in an RRC connected state. When there is no connection (RRC connection) between the RRC of the UE 100 and the RRC of the base station 200, the UE 100 is in an RRC idle state. When the connection between the RRC of the UE 100 and the RRC of the base station 200 is suspended, the UE 100 is in an RRC inactive state.

[0033] The NAS layer (also simply referred to as "NAS") located above the RRC layer performs session management, mobility management, etc. NAS signaling is transmitted between the NAS layer of UE 100 and the NAS layer of CN device 300. Note that UE 100 has an application layer and the like in addition to the radio interface protocol. Also, the layer below the NAS layer is referred to as the AS layer (also simply referred to as "AS").

[0034] (2) Example of base station configuration FIG. 4 is a diagram showing an example of the configuration of the base station 200 (network node) according to this embodiment.

[0035] The base station 200 includes a transmitting unit 210, a receiving unit 220, a control unit 230, and a NW communication unit 240. The transmitting unit 210 and the receiving unit 220 configure a wireless communication unit 250 that performs wireless communication with the UE 100.

[0036] The transmitting unit 210 performs various transmissions under the control of the control unit 230. The transmitting unit 210 includes an antenna and a transmitter. The transmitter converts (up-converts) a baseband signal (transmission signal) output by the control unit 230 into a radio signal and transmits it from the antenna. The receiving unit 220 performs various receptions under the control of the control unit 230. The receiving unit 220 includes an antenna and a receiver. The receiver converts (down-converts) a radio signal received by the antenna into a baseband signal (reception signal) and outputs it to the control unit 230.

[0037] The control unit 230 performs various controls and processes in the base station 200. The operations of the base station 200 described below may be operations under the control of the control unit 230. The control unit 230 includes at least one processor and at least one memory. The memory stores programs executed by the processor and information used in the processing by the processor. The processor may include a baseband processor and a CPU. The baseband processor performs modulation / demodulation and encoding / decoding of baseband signals. The CPU executes programs stored in the memory to perform various processes.

[0038] The NW communication unit 240 is connected to adjacent base stations via an inter-base station interface, and is also connected to the CN device 300 via a base station-CN interface.

[0039] The transmitter 210 of the base station 200 configured in this manner transmits SSBs using beam sweeping, which switches the transmission beam direction at predetermined time intervals, and also transmits SSBs using beamforming for each transmission beam. Details of SSBs and beamforming will be described later.

[0040] (3) Example of user device configuration FIG. 5 is a diagram showing an example of the configuration of the UE 100 (user equipment) according to this embodiment.

[0041] The UE 100 includes a receiving unit 110, a transmitting unit 120, and a control unit 130. The receiving unit 110 and the transmitting unit 120 configure a wireless communication unit 140 that performs wireless communication with the base station 200.

[0042] The receiving unit 110 performs various reception operations under the control of the control unit 130. The receiving unit 110 includes an antenna and a receiver. The receiver converts (down-converts) a radio signal received by the antenna into a baseband signal (received signal) and outputs the signal to the control unit 130. The transmitting unit 120 performs various transmission operations under the control of the control unit 130. The transmitting unit 120 includes an antenna and a transmitter. The transmitter converts (up-converts) a baseband signal (transmitted signal) output by the control unit 130 into a radio signal and transmits the signal from the antenna.

[0043] The control unit 130 performs various controls and processes in the UE 100. The operations of the UE 100 described below may be operations under the control of the control unit 130. The control unit 130 includes at least one processor and at least one memory. The memory stores programs executed by the processor and information used in the processing by the processor. The processor may include a baseband processor and a CPU (Central Processing Unit). The baseband processor performs modulation / demodulation and encoding / decoding of baseband signals. The CPU executes programs stored in the memory to perform various processes.

[0044] The UE 100 configured in this manner performs wireless communication with a base station 200 that manages a cell in a mobile communication system. The receiver 110 receives an SSB transmitted from the base station 200 using beamforming and to which beam sweeping, which switches the transmission beam direction at predetermined time intervals, has been applied. The controller 130 performs a cell search based on the SSB received by the receiver 110. Details of the cell search will be described later.

[0045] (4) Initial Access Next, initial access will be explained. First, cell search, which is one of the initial accesses, will be explained. Specifically, SSB, beam sweeping, and cell search will be explained.

[0046] 6 and 7 are diagrams for explaining SSB, beam sweeping, and cell search according to the first embodiment.

[0047] The transmitter 210 of the base station 200 that manages the cell transmits an SSB (SS / PBCH block) used for cell search by the UE 100. As shown in FIG. 6, each SSB is composed of four symbols in the time axis direction and 240 consecutive subcarriers (i.e., 20 RBs) in the frequency axis direction. These subcarriers are numbered in ascending order from 0 to 239 within the SSB, from the lowest frequency side to the highest frequency side. The subcarrier on the lowest frequency side within an SSB is also referred to as subcarrier 0. Each of the PSS and SSS consists of one symbol and 127 subcarriers. The PBCH consists of three symbols and 240 subcarriers.

[0048] Each SSB includes a synchronization signal (SS) and a physical broadcast channel (PBCH). The SS includes a primary synchronization signal (PSS) and a secondary synchronization signal (SSS). The PSS and SSS are used for synchronization at least in the time axis direction. The combination of the PSS and SSS signal sequences indicates the cell ID (PCI) of the transmitting cell. The PBCH includes a master information block (MIB) and a demodulation reference signal (DMRS). The MIB includes parameters for decoding the system information block type 1 (SIB1). The DMRS is a reference signal for decoding the PBCH.

[0049] The SSBs are allocated continuously in the time domain at specific frequencies within the cell band. Specifically, in the current 3GPP technical specifications, the SSBs are allocated to a unique frequency predetermined by the base station 200, and the SSBs are repeatedly transmitted in the time domain. The positions of the SSBs on the frequency domain can be notified to the UE 100 by RRC signaling from the base station 200.

[0050] The transmitter 210 of the base station 200 periodically transmits SS bursts (also referred to as "SS burst sets" or "synchronization signal bursts") consisting of multiple SSBs arranged in the time direction. The transmission period of the SS bursts can be selected from 5 ms, 10 ms, 20 ms, 40 ms, 80 ms, and 160 ms, with 20 ms being the most common. Figure 6 shows an example in which the transmission period of the SS bursts is 20 ms and the number of SSBs in the SS burst is 8. Note that the SS bursts are specified to be set within a half-frame time (5 ms).

[0051] Control unit 230 of base station 200 assigns an SSB index, which is an identifier for each SSB in an SS burst, to the SSB. As shown in FIG. 6, the SSB index may be a unique number that starts from 0 and increments by 1, or may be a unique number that starts from "1" and increments by 1. This number is reset to 0 in the next SS burst. Base station 200 notifies UE 100 of the SSB index via the PBCH in the SSB. UE 100 that receives an SSB can identify the SSB index of the SSB based on the PBCH in the received SSB. The index of each beam transmitted from base station 200 may be the SSB index of the SSB included in the beam.

[0052] As shown in Figure 7, transmitter 210 of base station 200 performs beam sweeping within the period of each SS burst (SS burst period) in order to transmit SSBs throughout the entire coverage area of ​​the cell. By using beam sweeping, base station 200 transmits each SSB within the SS burst in a time-division manner, with each SSB in a different transmission beam direction. This makes it possible to transmit SSBs throughout the entire coverage area of ​​the cell (cell coverage) within the SS burst, even when transmitting SSBs using beamforming.

[0053] Meanwhile, the control unit 130 of the UE 100 performs a cell search based on the SSBs (specifically, the PSS, SSS, and DMRS in the SSBs). The cell search is a procedure in which the UE 100 obtains time and frequency synchronization with a cell and detects the cell ID of the cell. The receiving unit 110 of the UE 100 performs an SSB scan (cell search) on a synchronization raster, which is a position on the frequency axis where the SSBs can be located. The receiving unit 110 (or the control unit 130) of the UE 100 measures the reception quality of each received SSB and identifies the SSB index of the SSB whose reception quality satisfies a predetermined condition, thereby identifying an appropriate beam. Here, the reception quality may be RSRP (Reference Signal Received Power) in the SSB. The predetermined condition may be that the reception quality exceeds a threshold. The predetermined condition may be that the reception quality is the highest among the SSBs received within a predetermined period (e.g., an SS burst period). In the example of FIG. 7, since the reception quality of SSB#1 is the highest, UE 100 can identify SSB#1 as the SSB whose reception quality satisfies a predetermined condition.

[0054] Next, the random access (RA) procedure for initial access will be described.

[0055] The control unit 130 of the UE 100 performs an RA procedure for initial access to the NW 10 (base station 200). Specifically, the transmission unit 120 of the UE 100 transmits an RA preamble (message 1: Msg1) to the base station 200 to perform RA. RA occasions, which are timings at which RA preambles can be transmitted, are prepared as many times as the number of transmission beams of the base station 200 (i.e., the number of SSBs in an SS burst). The UE 100 transmits the RA preamble to the base station 200 in the RA occasion corresponding to an SSB (SSB index) whose reception quality satisfies a predetermined condition. The control unit 230 of the base station 200 that receives the RA preamble determines a transmission beam preferable for the UE 100 (i.e., the direction in which the UE 100 is located) based on the correspondence between the beam (SSB index) and the RA occasion.

[0056] The RA occasion is notified to UE 100 in the system information provided by base station 200. There is a one-to-one relationship between SSB and RA occasion. When base station 200 directs a transmission beam in a certain direction and receives a signal at the RA occasion corresponding to that SSB, it receives the signal using a reception beam directed in the same direction as that transmission beam.

[0057] In addition, the control unit 130 of the UE 100 can perform time synchronization in the uplink direction with the base station 200 by performing time adjustment based on the timing advance value included in the RA response (message 2: Msg2) transmitted from the base station 200 after the RA preamble.

[0058] In the mobile communication system, initial access, specifically cell search and random access procedure, is performed as described above.

[0059] (5) RSRP map according to the first embodiment Next, the RSRP map according to the first embodiment will be described.

[0060] FIG. 8 is a diagram illustrating an example of an RSRP map.

[0061] As shown in Fig. 8, the RSRP map is a two-dimensional array of RSRPs representing reception quality. The RSRPs represent the reception quality at UE 100 when UE 100 receives the SSB included in each beam. Specifically, the RSRP map represents, in a two-dimensional array, the RSRPs for each beam identified by the azimuth angle and the elevation angle. In Fig. 8, the horizontal axis represents the azimuth angle, and the vertical axis represents the elevation angle. Also, in Fig. 8, the RSRP for each beam is represented by different shading. Fig. 8 shows an example where the number of beams is 15.

[0062] In the first embodiment, the base station 200 inputs the RSRP map into an AI model (trained AI model) and predicts, for example, the RSRP for all beams from the RSRPs obtained from a small number of beams equal to or less than a certain number. Then, the base station 200 selects beams to be used for initial access from the predicted RSRPs of all beams. For example, the control unit 230 of the base station 200 selects fewer than 15 beams from the (predicted) 15 beams shown in FIG. 8 as beams to be used for initial access. This makes it possible to select beams according to the bias, even if there is a bias in the distribution of the UEs 100, for example, and improve the efficiency of the selected beams.

[0063] Fig. 9 is a diagram illustrating an example of operation according to the first embodiment. However, the example of operation illustrated in Fig. 9 illustrates an example in which an AI model is not used.

[0064] As shown in FIG. 9 , in step S1, base station 200 performs data collection. Specifically, transmitter 210 of base station 200 transmits N beams (e.g., N=15) using beamforming and beam sweeping. "N" may represent the total number of beams that can be transmitted from base station 200, or may represent a number of beams less than the total number of beams. Control unit 130 of UE 100 measures the RSRP of the SSB included in each of the N beams. Transmitter 120 of UE 100 transmits the RSRP of each beam to base station 200 as a measurement result. Transmitter 120 of UE 100 may transmit the RSRP of each beam (e.g., "ssb-Index-RSRP") to base station 200 using a CSI report. Receiver 220 of base station 200 receives the RSRP of each beam for each UE 100. The control unit 230 of the base station 200 stores the collected RSRPs in a memory as an RSRP map (FIG. 8). The control unit 230 may store an RSRP map for each UE 100 in a memory. Note that data collection may be performed, for example, when the base station 200 is installed (or when measurements are performed before operation).

[0065] In step S2, the control unit 230 of the base station 200 selects M (M < N) beams using the RSRP map. The control unit 230 will select M beams from among the N beams. The beam selection may also be performed, for example, at the time of installation (or measurement) of the base station 200.

[0066] In step S3, the transmission unit 210 of the base station 200 transmits the M beams selected in step S2 by beamforming and beam sweeping. In the UE100, initial access is performed using the M beams.

[0067] (6) Problems Related to the First Embodiment The UE100 may move over time. Therefore, during operation, even if the M beams (step S2) selected by the base station 200 are appropriate beams for connecting to the base station 200 at a certain time in the UE100, they may not be appropriate beams for connecting to the base station 200 at other times. Therefore, the base station 200 also performs data collection using the N beams (step S1) during operation, updates the M beams used for connection with the UE100 (step S2), and there is a need to perform initial access with the UE100 again using the updated M beams (step S3).

[0068] However, during operation, even though initial access is performed with M beams (step S3), having the base station 200 perform initial access using all beams to collect the RSRP of each UE (step S1) in order to perform data collection (step S1) does not lead to an improvement in the power efficiency of the base station 200. Also, assuming that the UE100 is moving over time, if data collection using the N beams (step S1) is also performed during operation of the base station 200, even when the UE100 is not moving, the base station 200 will reset the initial access established between the base station 200 and the UE100 in order to transmit the N beams during data collection, which does not lead to an improvement in power efficiency.

[0069] Therefore, in the first embodiment, an AI model is used to infer the RSRPs of N beams (more than M) from the RSRPs for M beams (used for initial access), and M beams are selected from the inferred N RSRPs to update the beams to be used for initial access.

[0070] Specifically, first, base station 200 receives from UE 100 reception qualities for M (N>M) beams, which is less than the number of N (N>2) beams that base station 200 can transmit. Second, base station 200 uses a trained AI model to infer N reception qualities from the M reception qualities. Third, base station 200 selects M beams based on the N reception qualities. Fourth, base station 200 transmits the selected M beams.

[0071] As a result, for example, base station 200 can acquire RSRP from UE 100 using M beams. Compared to acquiring RSRP from UE 100 using all beams (e.g., N beams), the number of transmission beams can be reduced, and power efficiency can be improved. UE 100 can perform initial access using a certain number of beams or less. Furthermore, in a situation where UE 100 is not moving and there is no need to perform initial access again, it is possible to determine that UE 100 is not moving based on the inference result of the AI ​​model, and maintain the already established initial access. This prevents unnecessary resetting of initial access once established, and makes it possible to improve power efficiency.

[0072] (7) Configuration example of the base station 200 according to the first embodiment Next, a configuration example of the base station 200 according to the first embodiment will be described.

[0073] FIG. 10 is a diagram illustrating an example of the configuration of the base station 200. As shown in FIG.

[0074] 10, base station 200 includes a digital signal processing unit (hereinafter may be referred to as a "processing unit") 231, an RSRP map AI learning unit (hereinafter may be referred to as a "learning unit") 232, an RSRP AI inference unit (hereinafter may be referred to as an "inference unit") 233, an RSRP map storage database 234, an initial access beam selection unit (hereinafter may be referred to as a "beam selection unit") 235, and an initial access beam storage memory (hereinafter may be referred to as a "beam storage memory") 236. Base station 200 also includes a beam forming unit (BF unit) 211 and multiple antennas 212.

[0075] The processing unit 231, the learning unit 232, the inference unit 233, the RSRP map storage database 234, the beam selection unit 235, and the beam storage memory 236 are included in the control unit 230 in Fig. 4. The BF unit 211 and the multiple antennas 212 are included in the transmission unit 210 in Fig. 4.

[0076] The processing unit 231 outputs various data to the learning unit 232 and the inference unit 233, and instructs the BF unit 211 to perform various processes. Specifically, the following occurs.

[0077] First, processing unit 231 outputs RSRPs for N beams to learning unit 232. Processing unit 231 transmits N beams in advance (when the base station is installed or during measurement) via BF unit 211 and antenna 212 by beamforming and beam sweeping, and further acquires the RSRP of each beam from UE 100 as a CSI report. Processing unit 231 outputs the (N) RSRPs for the N beams acquired from UE 100 to learning unit 232. Processing unit 231 can acquire N RSRPs for each UE 100, and in this case outputs the N RSRPs for each UE 100 to learning unit 232. Note that the N RSRPs can be correct data in the learning model.

[0078] Second, processing unit 231 outputs the RSRPs for the M beams to inference unit 233. The M RSRPs may be obtained by processing unit 231 transmitting M beams in advance when the base station is installed or measured, and further acquiring the RSRPs for each beam from UE 100 as a CSI report. Alternatively, the M RSRPs may be obtained by base station 200 transmitting M beams selected for initial access, and receiving the RSRPs for the beams from UE 100 as a CSI report. Processing unit 231 can acquire M RSRPs for each UE 100, and outputs the M RSRPs for each UE 100 to inference unit 233. The M RSRPs are used as input data for the trained AI model.

[0079] Third, the processing unit 231 instructs the BF unit 211 to perform beamforming using setting values ​​for beamforming (hereinafter, sometimes referred to as "BF setting values").

[0080] The learning unit 232 trains an AI model using N RSRPs (correct answer data) and M RSRPs to generate a trained AI model. Specifically, the learning unit 232 trains an AI model using an RSRP map including N RSRPs and an RSRP including M RSRPs to generate a trained AI model. FIG. 11(B) is a diagram showing an example of correct answer data. In the example of FIG. 11(B), an RSRP map including N=15 RSRPs is shown as correct answer data. As shown in FIGS. 11(A) and 11(C), the trained AI model is an AI model that infers N RSRPs (N=15 in the example of FIG. 11(C)) from M RSRPs (M=4 in the example of FIG. 11(A)). The learning unit 232 outputs the trained AI model to the inference unit 233.

[0081] As described above, the N RSRPs used for learning are RSRPs that the processing unit 231 has acquired in advance from the UE 100. The M RSRPs used for learning may be acquired by selecting M RSRPs from the N RSRPs.

[0082] The inference unit 233 infers (or predicts) N RSRPs from M RSRPs using the trained AI model. Specifically, the inference unit 233 infers an RSRP map including N RSRPs from an RSRP map including M RSRPs using the trained AI model. For example, the inference unit 233 receives an RSRP map including M=4 RSRPs as input data as shown in FIG. 11(A), and infers an RSRP map including N=15 RSRPs (all beams) as shown in FIG. 11(C). The inference unit 233 outputs the inferred RSRP map including N RSRPs to the RSRP map storage database 234. Note that the inference unit 233 also infers an RSRP map for each UE 100 (each RSRP map including M RSRPs) using the RSRP map for each UE 100 (each RSRP map including N RSRPs).

[0083] The RSRP map storage database 234 stores an RSRP map that is the inference result of the inference unit 233. The RSRP map storage database 234 stores an RSRP map for each UE 100. The RSRP map storage database 234 may be included in the memory in the control unit 230.

[0084] The beam selection unit 235 reads out an RSRP map from the RSRP map storage database 234, and selects M beams based on the N reception qualities included in the RSRP map. Specifically, the beam selection unit 235 selects M beams to be used for initial access by using a set of RSRP maps (hereinafter, may be referred to as a "RSRP map data set") in which the RSRP maps existing for each UE 100 are compiled in the number corresponding to the number of UEs 100. The selection process will be described in an operation example. The beam selection unit 235 stores the beam indexes of the selected M beams in the beam storage memory 236.

[0085] The beam storage memory 236 stores the beam index of the beam selected by the beam selection unit 235. The beam storage memory 236 holds a setting value (BF setting value) related to beam forming of the beam for each beam index, and can read out the BF setting value corresponding to the beam index selected by the beam selection unit 235. The beam storage memory 236 may be included in the memory within the control unit 230.

[0086] The BF unit 211 reads the BF setting values ​​stored in the beam storage memory 236 in accordance with instructions from the processing unit 231, and performs beamforming processing on a beam including SSB. The BF unit 211 may be configured using analog beamforming with a phase shifter and amplifier for each antenna element. Alternatively, the BF unit 211 may be configured using digital beamforming that achieves beamforming using a digital signal processing circuit. Alternatively, the BF unit 211 may be configured using hybrid beamforming that combines digital beamforming and analog beamforming and performs beam control using a phase controller, which is an analog circuit. Furthermore, the BF unit 211 performs beam sweeping by switching a beam formed in a fixed direction to a different direction at predetermined time intervals in accordance with instructions from the processing unit 231.

[0087] The antenna 212 transmits the beam formed by the BF unit 211 .

[0088] (8) Operational Example of the First Embodiment Next, an example of operation according to the first embodiment will be described.

[0089] FIG. 12 is a diagram illustrating an example of operation.

[0090] As shown in FIG. 12, in step S10, the base station 200 starts the process.

[0091] In step S11, processing unit 231 of base station 200 performs initial access using N beams and acquires RSRP of each UE 100. Step S11 may be performed when base station 200 is installed or when measurement is performed before operation. Step S11 may be the same process as step S1 shown in FIG. 9.

[0092] In step S12, base station 200 selects M beams. Here, learning unit 232 of base station 200 may select M RSRPs to be used for learning from the N RSRPs received from processing unit 231.

[0093] In step S13, the learning unit 232 of the base station 200 trains an AI model using the N RSRPs (correct data) collected in step S11 and the M RSRPs selected in step S12, and generates a trained AI model. The trained AI model is a model that infers N RSRPs from M RSRPs.

[0094] In step S14, the inference unit 233 of the base station 200 receives the M RSRPs acquired by the processing unit 231 from the processing unit 231, and infers N RSRPs from the M RSRPs using the trained AI model generated in step S13.

[0095] The process again proceeds to step S12, where the beam selection unit 235 of the base station 200 selects M beams to be used for initial access using the N RSRPs (specifically, a set of RSRP maps) inferred in step S14. Details of the beam selection process will be described later. Note that if the M beams selected in the previous execution of step S12 and the M beams selected in the current execution of step S12 are the same, in step S15, the base station 200 may obtain the RSRPs of the M beams and then proceed to step S14.

[0096] In step S15, base station 200 transitions to the operation phase, and BF unit 211 transmits the M beams selected in step S12 by beamforming and beam sweeping. UE 100 performs initial access using the M beams.

[0097] After starting operation, the base station 200 regenerates the trained AI model each time the beam index of the beam used for initial access changes. For example, if a different beam is used for initial access among the M=4 beams shown in FIG. 11(A), the RSRP map arrangement will also be different. Even if the RSRP map with a different arrangement is input to the trained AI model, an inference result significantly different from the correct data may be output. Therefore, in the first embodiment, after starting operation, step S13 is performed again each time the beam index of the beam used for initial access changes (step S13). The base station 200 then inputs an RSRP map including M RSRPs acquired based on the changed beam index to the updated trained AI model, and performs inference (step S14). The base station 200 selects a beam using the N RSRP maps that are the inference results (step S12).

[0098] (9) Beam Selection Processing in the First Embodiment Next, a specific example of the beam selection process (step S12) will be described.

[0099] FIG. 13 is a diagram illustrating an example of the operation of the beam selection process according to the first embodiment.

[0100] In step S120, the beam selector 235 of the base station 200 starts processing.

[0101] In step S121, the beam selection unit 235 sets "1" to P. P represents the number of selected beams.

[0102] In step S122, the beam selection unit 235 defines as a data set of RSRP maps aggregating the RSRPs for each beam in each UE 100. For example, the predicted data of the RSRP maps for each UE 100 shown in Fig. 11(C) may be aggregated for the target UE, and the resulting RSRP map may be defined as a data set of RSRP maps.

[0103] 13, in step S123, the beam selection unit 235 counts the number of UEs 100 whose RSRP is equal to or greater than threshold X in each beam for the data set of the RSRP map. Threshold X represents, for example, the RSRP threshold at which UEs 100 can connect to the base station 200. The count value represents, for example, the number of UEs 100 that can connect to the base station 200 in each beam.

[0104] In step S124, the beam selection unit 235 selects the beam with the most UEs 100. For example, the beam selection unit 235 selects the beam with the most UEs 100 whose RSRP is equal to or greater than the threshold X, that is, the most UEs 100 that are expected to connect to the base station 200. This beam becomes the beam used for initial access (the beam for P=1).

[0105] In step S125, the beam selection unit 235 selects Y UEs 100 in the beam selected in step S124 in descending order of RSRP. Then, the beam selection unit 235 excludes the RSRPs of the selected UEs 100 from the data set of the RSRP map. "Y" may represent the number of UEs 100 that the base station 200 is to accommodate.

[0106] In step S126, the beam selection unit 235 increments P.

[0107] In step S127, the beam selection unit 235 determines whether or not P = M. If P = M (Yes in step S127), the process proceeds to step S128. On the other hand, if P = M is not true (No in step S127), the process proceeds to step S123.

[0108] If P=M is ​​not satisfied, the processes from step S123 to step S127 are performed, and the processes are repeated until the number of beams selected in step S124 becomes M.

[0109] In step S128, the beam selection unit 235 outputs the beam indices of the selected M beams to the beam storage memory 236.

[0110] Then, in step S129, the beam selection is completed, and the process proceeds to step S15 in FIG.

[0111] 13, M beams are selected in descending order of the number of UEs 100 that have an RSRP equal to or greater than threshold X (step S123) (step S124). The base station 200 uses the beam selection process to select beams that can be transmitted to the maximum number of UEs 100 that can be expected to connect to the base station.

[0112] [Second embodiment] Next, a second embodiment will be described, focusing on the differences from the first embodiment.

[0113] In the first embodiment, an example is described in which a new trained AI model is generated in the base station 200 each time the beam used for initial access changes. In the second embodiment, an example is described in which a trained AI model is generated using RSRP with K (M>K) fixed beams.

[0114] In this case, the K beams are always included in the M beams selected by the beam selection unit 235. Therefore, even if the beam used for initial access changes, the K beams included therein remain fixed. Therefore, once the base station 200 generates a trained AI model using RSRP using the K beams, even if the M beams change, the K beams that are the target of the input data for the trained AI model do not change, and there is no need to update the trained AI model. Therefore, in the second embodiment, even if the beam used for initial access changes, there is no need to update the trained AI model.

[0115] (1) Configuration example of base station according to the second embodiment 14 is a diagram illustrating a configuration example of the base station 200 according to the second embodiment. In the base station 200 according to the second embodiment, the learning unit 232 and the inference unit 233 perform learning and inference using an RSRP map including K RSRPs.

[0116] That is, the learning unit 232 receives, for example, N RSRPs (correct data) from the processing unit 231 and selects RSRPs for K fixed beams from the N RSRPs. The learning unit 232 trains an AI model using an RSRP map including N RSRPs and an RSRP map including K RSRPs to generate a trained AI model. The trained AI model uses the RSRP map including K RSRPs as input data and infers (or predicts) an RSRP map including N RSRPs.

[0117] The inference unit 233 uses the trained AI model to infer an RSRP map including N RSRPs from an RSRP map including K RSRPs.

[0118] The beam selection unit 235 selects M beams including K fixed beams from the RSRP map data set. The beam selection process will be explained in an operation example. Even if the arrangement of the M beams selected by the beam selection unit 235 differs from the previously selected arrangement, the M beams always include K beams, so there is no need for the learning unit 232 to generate a new trained AI model.

[0119] (2) Operation example according to the second embodiment FIG. 15 is a diagram illustrating an example of operation according to the second embodiment.

[0120] In FIG. 15, steps S11 and S15 are the same as those in the first embodiment.

[0121] In step S20, processing unit 231 of base station 200 determines K fixed beams. Processing unit 231 outputs beam indices of the determined K beams to learning unit 232. Processing unit 231 also acquires RSRPs for the K beams from UE 100, and outputs the acquired K RSRPs to inference unit 233. Processing unit 231 may cause UE 100 to perform initial access using K beams when base station 200 is installed or during measurement before operation, and acquire K RSRPs from UE 100. Processing unit 231 also outputs beam indices of the determined K beams to learning unit 232.

[0122] In step S21, the learning unit 232 acquires K RSRPs corresponding to the beam indexes of K beams from the N RSRPs. The learning unit 232 of the base station 200 trains an AI model using an RSRP map including the N RSRPs and an RSRP map including the K RSRPs, and generates a trained AI model.

[0123] In step S22, the inference unit 233 infers an RSRP map including N RSRPs from the RSRP map including K RSRPs, using the trained AI model generated in step S21.

[0124] In step S23, the beam selector 235 selects M beams including K fixed beams using an RSRP map including N RSRPs. The beam selection process will be described later.

[0125] Then, in step S15, base station 200 enters the operation stage and transmits the beam selected in step S23 by beamforming and beam sweeping. UE 100 performs initial access. After this, even if the M beams selected in step S23 are different from the previously selected M beams, the selected M beams include the fixedly selected K beams, so that the learning itself (step S21) does not need to be performed again to create a new trained AI / ML model.

[0126] (3) Beam selection processing according to the second embodiment FIG. 16 is a diagram illustrating an example of the operation of the beam selection process (step S23) according to the second embodiment.

[0127] In FIG. 16, steps S231 to S235 are the same as steps S121 to S125 in the first embodiment, respectively.

[0128] In step S236, the beam selection unit 235 increments P when the selected beam is other than the K beams. P in the second embodiment represents the number of beams other than the K beams out of the M beams that are finally selected. The beam selection unit 235 selects beams ((MK) beams) to be used for initial access from among the beams other than the K beams by the operation shown in FIG. 16.

[0129] Therefore, in step S237, when the number of P beams that have been selected reaches (MK), the beam selection unit 235 proceeds to step S238. On the other hand, in step S237, if the number of P beams that have been selected is (MK) or less, the beam selection unit 235 repeats steps S233 to S237 until the number reaches (MK).

[0130] In step S238, the beam selection unit 235 determines M beams, which are a combination of the selected (MK) beams and the K beams, as beams to be used for initial access, and outputs the beam index of the beams.

[0131] Then, in step S239, the beam selector 235 ends the beam selection process, and proceeds to step S15 in FIG.

[0132] [Other embodiments] The learning unit 232 in the first and second embodiments may create a trained model using, for example, a convolutional neural network (CNN). CNN is a machine learning technique that extracts features of input data in a convolutional layer, extracts the maximum value in a pooling layer, and combines the outputs of the pooling layers in a connection layer. CNN may be used for super-resolution, which obtains an output image with a higher resolution than the input image, such as converting a 2K image into a 4K image. Inferring the predicted data shown in FIG. 11(C) from the input data shown in FIG. 11(A) can be considered to be the same as super-resolution. The learning unit 232 may generate a trained model using CNN by calculating parameters of an AI model that minimizes the loss function (i.e., the inference result is closest to the correct data).

[0133] Furthermore, in the first and second embodiments, RSRP has been used as an example of reception quality, but reception quality is not limited to RSRP. For example, SINR (Signal-to-Interference-plus-Noise Ratio) may be used as reception quality. SINR represents the ratio between the power of a desired signal and the power of signals other than the desired signal (interference signals or noise signals).

[0134] Furthermore, in the first and second embodiments, an example has been described in which the base station is an NR base station (gNB), but the base station may be an LTE base station (eNB) or a 6G base station. Also, the base station may be a relay node such as an IAB (Integrated Access and Backhaul) node. The base station may be a DU of the IAB node. Also, the UE 100 may be an MT (Mobile Termination) of the IAB node.

[0135] Furthermore, in the first and second embodiments, an example in which inference is performed in the base station 200 has been described, but inference may be performed in an external device such as a server device, for example.

[0136] In this case, the external device may have an inference unit 233. The base station 200 transmits the CSI report received from the UE 100 to the external device, and the external device may obtain the RSRP value from the CSI report. Alternatively, the base station 200 may transmit the RSRP value of the UE 100 obtained based on the CSI report received from the UE 100 to the external device, so that the external device may obtain the RSRP value of the UE 100. Then, in the inference unit 233 of the external device, the inference described in the first embodiment and the second embodiment may be performed. That is, in the inference unit 233 of the external device, the RSRP (received quality) (or RSRP map) for M (M < N) beams with a smaller number of beams than N (N > 2) beams that can be transmitted from the base station 200 is input, and the RSRP (or RSRP map) for N beams is inferred. The external device may transmit the N RSRPs that are the inference results to the base station 200. In the beam selection unit 235 of the base station 200, M beam indexes used for initial access may be selected from the N RSRPs. Alternatively, the inference described in the first embodiment and the second embodiment may be performed by the cooperation of the base station 200 and the external device. In this way, the base station 200 or the external device may perform the inference described in the first embodiment and the second embodiment by cooperating with other devices.

[0137] Alternatively, the external device may include an inference unit 233, an RSRP map storage database 234, and a beam selection unit 235. In this case, the RSRP map storage database 234 and the beam selection unit 235 may perform the processes described in the first and second embodiments. That is, the RSRP map storage database 234 stores the N RSRPs inferred by the inference unit 233, and the beam selection unit 235 reads the N RSRPs from the RSRP map storage database 234 and selects M beams to be used for initial selection. The external device may transmit beam indexes of the selected M beams to the base station 200. Alternatively, the external device may instruct the base station 200 to transmit the beams together with the beam indexes. The instruction may be given by a message or a signal. The base station 200 performs initial access processing using the beams instructed by the instruction.

[0138] Furthermore, the term "network node" primarily refers to a base station, but may also refer to a part (CU, DU, or RU) of the CN device 300 or the base station 200. Also, a network node may be configured by a combination of at least a part of the CN device 300 and at least a part of the base station 200.

[0139] Furthermore, a program may be provided that causes a computer to execute each process performed by the UE 100 or the base station 200. The program may be recorded on a computer-readable medium. Using the computer-readable medium, the program can be installed on a computer. Here, the computer-readable medium on which the program is recorded may be a non-transitory recording medium. The non-transitory recording medium is not particularly limited, and may be, for example, a recording medium such as a CD-ROM or a DVD-ROM. Furthermore, circuits that execute each process performed by the UE 100 or the base station 200 may be integrated, and at least a part of the UE 100 or the base station 200 may be configured as a semiconductor integrated circuit (chip set, SoC).

[0140] Furthermore, the operation flows in the first and second embodiments can be implemented not only separately and independently but also by combining two or more operation flows. For example, some steps of one operation flow may be added to another operation flow, or some steps of one operation flow may be replaced with some steps of another operation flow. In each flow, it is not necessary to execute all the steps, and only some steps may be executed.

[0141] As described above, one embodiment has been described in detail with reference to the drawings. However, the specific configuration is not limited to the above, and various design changes and the like can be made without departing from the gist. Also, each embodiment, each operation example, or each process can be appropriately combined within a non - conflicting range.

[0142] (Supplementary Note) Summarizing the above, it becomes as in the supplementary note, but the supplementary note does not limit the embodiments.

[0143] (Supplementary Note 1) A base station in a mobile communication system, a receiving unit that receives from a user device the reception quality for M (M < N) beams, which is less than the number of N (N>2) beams that can be transmitted from the base station; an inference unit that uses a learned AI model to infer the reception quality of the N beams from the reception quality of the M beams; a beam selection unit that selects the M beams based on the reception quality of the N beams; and a transmission unit that transmits the selected M beams. Base station.

[0144] (Supplementary Note 2) The receiving unit receives from the user device the reception quality for the N beams, and further has a learning unit that learns an AI model using the reception quality of the N beams and the reception quality of the M beams to generate the learned AI model. The base station according to Supplementary Note 1.

[0145] (Appendix 3) The M reception qualities input to the trained AI model and the N reception qualities output as inference results from the trained AI model are arranged in a two-dimensional array. 1. A base station according to claim 1 or 2.

[0146] (Appendix 4) the inference unit infers the N reception qualities for each of the user devices; The beam selector selects the M beams in descending order of the number of user devices that have the reception quality equal to or greater than a threshold. 4. The base station according to claim 1, wherein the base station is a base station having a plurality of sub-units.

[0147] (Appendix 5) The inference unit infers the N reception qualities from the reception qualities of K (M>K) fixed beams using the trained AI model. 5. The base station according to claim 1, wherein the base station is a base station having a plurality of sub-units.

[0148] (Appendix 6) The learning unit trains the AI ​​model using the N reception qualities and the reception qualities of K (M>K) fixed beams to generate the trained AI model. 6. A base station according to any one of Supplementary Note 1 to Supplementary Note 5.

[0149] (Appendix 7) The beam selector selects the M beams including the K fixed beams. 7. The base station according to claim 1,

[0150] (Appendix 8) The reception quality is the RSRP (Reference Signal Received Power) of the signal included in the beam. 8. The base station according to claim 1,

[0151] (Appendix 9) A beam transmission method in a mobile communication system, comprising: A step in which the base station receives from the user equipment reception qualities for M (N>M) beams, the number of which is less than the number of N (N>2) beams that can be transmitted from the base station; The base station infers the N reception qualities from the M reception qualities using a trained AI model; the base station selecting the M beams based on the N reception qualities; the base station transmitting the selected M beams. Beam transmission method.

[0152] (Appendix 10) A mobile communication system having a base station and a user equipment, The base station receives from the user equipment reception qualities for M (N>M) beams, which is less than the number of N (N>2) beams that can be transmitted from the base station; The base station uses a trained AI model to infer the N reception qualities from the M reception qualities; the base station selects the M beams based on the N reception qualities; The base station transmits the selected M beams. Mobile communication system. [Explanation of symbols]

[0153] 10: Network (NW) 20: RAN 30: Core network (CN) 100: User equipment (UE) 110: Receiving section 120: Transmitting section 130: Control unit 200: Base station 210: Transmitting unit 211: BF unit 212: Antenna 220: Receiving unit 230: Control unit 231: Digital signal processing unit 232: AI learning part of RSRP map 233: AI inference part of RSRP map 234: RSRP map storage database 235: Initial access beam selection unit 236: Beam storage memory for initial access 300:CN device

Claims

1. A base station in a mobile communication system, a receiving unit that receives, from a user device, reception qualities for M (M<N) beams, the number of which is less than the number N (N>2) beams that can be transmitted from the base station; an inference unit that infers the N reception qualities from the M reception qualities using a trained AI model; a beam selection unit that selects the M beams based on the N reception qualities; a transmitting unit for transmitting the selected M beams; Base station.

2. The receiver receives reception qualities for the N beams from the user equipment; Further, a learning unit that trains an AI model using the N reception qualities and the M reception qualities to generate the trained AI model. The base station of claim 1.

3. The M reception qualities input to the trained AI model and the N reception qualities output as inference results from the trained AI model are arranged in a two-dimensional array. The base station of claim 1.

4. the inference unit infers the N reception qualities for each of the user devices; The beam selector selects the M beams in descending order of the number of user devices that have the reception quality equal to or greater than a threshold. The base station of claim 1.

5. The inference unit infers the N reception qualities from the reception qualities of K (M>K) fixed beams using the trained AI model. The base station of claim 1.

6. The learning unit trains the AI ​​model using the N reception qualities and the reception qualities of K (M>K) fixed beams to generate the trained AI model. The base station according to claim 2.

7. The beam selection unit selects the M beams including the K fixed beams. The base station according to claim 5.

8. The reception quality is the RSRP (Reference Signal Received Power) of the signal included in the beam. The base station of claim 1.

9. A beam transmission method in a mobile communication system, comprising: A step in which the base station receives from the user equipment reception qualities for M (N>M) beams, the number of which is less than the number of N (N>2) beams that can be transmitted from the base station; The base station infers the N reception qualities from the M reception qualities using a trained AI model; the base station selecting the M beams based on the N reception qualities; the base station transmitting the selected M beams. Beam transmission method.

10. A mobile communication system having a base station and a user equipment, The base station receives from the user equipment reception qualities for M (N>M) beams, the number of which is less than the number of N (N>2) beams that can be transmitted from the base station; The base station uses a trained AI model to infer the N reception qualities from the M reception qualities; the base station selects the M beams based on the N reception qualities; The base station transmits the selected M beams. Mobile communication system.

Images