Method and device for performing communication by using artificial intelligence and machine learning

Abstract

The present embodiments relate to a method and a device for performing communication by using artificial intelligence and machine learning, and provide a method comprising the steps of: receiving, from a first base station, a handover command message including at least one associated ID supported by a second base station with respect to an AI / ML model; performing a handover to the second base station on the basis of the handover command message; when the at least one associated ID matches an associated ID for the AI / ML model stored in a terminal, performing performance monitoring on the stored AI / ML model; and transmitting, to the second base station, a message including a monitoring result.

Description

Method and device for performing communication using artificial intelligence and machine learning

[0001] The present embodiments propose a method and apparatus for performing communication using artificial intelligence and machine learning in a next-generation wireless access network (in this disclosure, “5G,” “NR [New Radio],” “5G-Advanced,” “6G,” or a subsequent 3GPP wireless access network).

[0002] Next-generation wireless communication technology is evolving beyond 5G to 6G, aiming to achieve faster data transmission speeds and ultra-low latency compared to 5G in ultra-high frequency ranges such as the terahertz (THz) band. Therefore, technology is advancing toward incorporating artificial intelligence (AI) and machine learning (ML) technologies from the communication system design stage. Consequently, wireless communication systems are establishing a technological foundation to support new services and applications in ultra-high-performance, ultra-low latency, and hyper-connected environments.

[0003] In particular, AI / ML technologies are being introduced in wireless communication networks to optimize network operations and ensure real-time quality. AI / ML can perform a variety of roles, including situational awareness through big data analysis, adaptive utilization of network resources and data, and intelligent, data-driven system optimization. These capabilities can enable efficient resource management and improved service quality in complex wireless environments, where conventional methods have proven limited.

[0004] As part of this aspect, a specific design is needed to enable wireless communication using AI / ML models.

[0005] Embodiments of the present disclosure can provide a method and device for performing communication using artificial intelligence and machine learning in a next-generation wireless access network.

[0006] In one aspect, the present embodiments may provide a method for a user equipment (UE) to perform communication using artificial intelligence and machine learning (AI / ML), the method comprising: receiving, from a first base station, a handover command message including at least one associated ID supported by a second base station for an AI / ML model; performing a handover to the second base station based on the handover command message; performing performance monitoring for the stored AI / ML model when the associated ID for the AI / ML model stored in the UE matches the at least one associated ID; and transmitting a message including the monitoring result to the second base station.

[0007] In another aspect, the present embodiments may provide a method for a base station to perform communication using artificial intelligence and machine learning (AI / ML), the method comprising: transmitting, to a target base station, a handover request message including at least one associated ID for an AI / ML model stored in a terminal; receiving, from the target base station, a handover request ACK message including at least one associated ID supported by the target base station for the AI / ML model; and transmitting, to the terminal, a handover command message including at least one associated ID supported by the target base station.

[0008] In another aspect, the present embodiments provide a user equipment (UE) that performs communication using artificial intelligence and machine learning (AI / ML), the UE including a transmitter, a receiver, and a control unit that controls operations of the transmitter and the receiver, wherein the control unit receives, from a first base station, a handover command message including at least one associated ID supported by a second base station for an AI / ML model, performs a handover to the second base station based on the handover command message, and performs performance monitoring for the stored AI / ML model when the associated ID for the AI / ML model stored in the UE matches the at least one associated ID, and transmits a message including the monitoring result to the second base station.

[0009] In another aspect, the present embodiments may provide a base station that performs communication using artificial intelligence and machine learning, including a transmitter, a receiver, and a control unit that controls operations of the transmitter and the receiver, wherein the control unit transmits, to a target base station, a handover request message including at least one associated ID for an AI / ML model stored in a terminal, receives, from the target base station, a handover request ACK message including at least one associated ID supported by the target base station for the AI / ML model, and transmits, to the terminal, a handover command message including at least one associated ID supported by the target base station.

[0010] According to the present embodiments, a method and device for performing communication using artificial intelligence and machine learning in a next-generation wireless access network can be provided.

[0011] Additionally, according to the present embodiments, the overall system performance can be improved by minimizing the overhead of performing model inference through data collection and model training each time a terminal moves between cells.

[0012] FIG. 1 is a schematic diagram illustrating the structure of an NR wireless communication system to which the present embodiment can be applied.

[0013] FIG. 2 is a drawing for explaining a frame structure in an NR system to which the present embodiment can be applied.

[0014] FIG. 3 is a diagram for explaining a resource grid supported by a wireless access technology to which the present embodiment can be applied.

[0015] FIG. 4 is a diagram for explaining a bandwidth part supported by a wireless access technology to which the present embodiment can be applied.

[0016] FIG. 5 is a diagram illustrating an example of a synchronization signal block in a wireless access technology to which the present embodiment can be applied.

[0017] FIG. 6 is a diagram for explaining a random access procedure in a wireless access technology to which the present embodiment can be applied.

[0018] Figure 7 is a drawing for explaining CORESET.

[0019] FIG. 8 is a diagram illustrating a procedure in which a terminal performs communication using artificial intelligence and machine learning according to one embodiment.

[0020] FIG. 9 is a diagram illustrating a procedure in which a base station performs communication using artificial intelligence and machine learning according to one embodiment.

[0021] FIG. 10 and FIG. 11 are diagrams for explaining an operation when a target base station determines an associated ID according to one embodiment.

[0022] FIG. 12 and FIG. 13 are diagrams for explaining an operation when a serving base station determines an association ID according to one embodiment.

[0023] FIG. 14 and FIG. 15 are diagrams for explaining the operation when a terminal determines an associated ID according to one embodiment.

[0024] FIG. 16 and FIG. 17 are diagrams for explaining an operation of receiving an association ID of a target base station according to one embodiment.

[0025] Fig. 18 is a drawing showing the configuration of a terminal according to another embodiment.

[0026] Fig. 19 is a drawing showing the configuration of a base station according to another embodiment.

[0027] Hereinafter, some embodiments of the present disclosure will be described in detail with reference to exemplary drawings. When adding reference numerals to components in each drawing, identical components may have the same numerals as much as possible even if they are shown in different drawings. In addition, when describing the present embodiments, if it is determined that a detailed description of a related known configuration or function may obscure the gist of the technical idea of ​​the present invention, the detailed description may be omitted. When "includes," "has," "consists of," etc. are used in this specification, other parts may be added unless "only" is used. When a component is expressed in the singular, it may include a case in which the plural is included unless specifically stated otherwise.

[0028] Additionally, terms such as first, second, A, B, (a), (b), etc. may be used to describe components of the present disclosure. These terms are only intended to distinguish the components from other components, and the nature, order, sequence, or number of the components are not limited by the terms.

[0029] In a description of the positional relationship of components, when it is described that two or more components are "connected," "combined," or "connected," it should be understood that the two or more components may be directly "connected," "combined," or "connected," but that the two or more components may also be further "interposed" with another component to be "connected," "combined," or "connected." Here, the other component may be included in one or more of the two or more components that are "connected," "combined," or "connected" to each other.

[0030] In the description of the temporal flow relationship related to components, operation methods, or manufacturing methods, for example, when the temporal or flow relationship is described as “after”, “following”, “next to”, “before”, etc., it may also include cases where it is not continuous, unless “immediately” or “directly” is used.

[0031] Meanwhile, when numerical values ​​or corresponding information (e.g., levels, etc.) for components are mentioned, even without separate explicit description, the numerical values ​​or corresponding information may be interpreted as including an error range that may occur due to various factors (e.g., process factors, internal or external impact, noise, etc.).

[0032] The wireless communication system in this specification refers to a system for providing various communication services such as voice, data packets, etc. using wireless resources, and may include a terminal, a base station, or a core network.

[0033] The embodiments disclosed below can be applied to wireless communication systems using various wireless access technologies. For example, the embodiments can be applied to various wireless access technologies such as 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), or NOMA (non-orthogonal multiple access). In addition, the wireless access technology may not only refer to a specific access technology, but also to each generation of communication technology established by various communication agreement organizations such as 3GPP, 3GPP2, WiFi, Bluetooth, IEEE, and ITU. For example, CDMA can be implemented with wireless technologies such as UTRA (universal terrestrial radio access) or CDMA2000. TDMA can be implemented with wireless technologies such as GSM (global system for mobile communications) / GPRS (general packet radio service) / EDGE (enhanced data rates for GSM evolution). OFDMA can be implemented in 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) that uses E-UTRA (evolved-UMTSterrestrial radio access), employing OFDMA in the downlink and SC-FDMA in the uplink. Thus, the present embodiments can be applied to currently disclosed or commercialized wireless access technologies, as well as wireless access technologies currently under development or to be developed in the future.

[0034] Meanwhile, the term "terminal" in this specification is a comprehensive concept that refers to a device that includes a wireless communication module that performs communication with a base station in a wireless communication system, and should be interpreted as a concept that includes not only UE (User Equipment) in WCDMA, LTE, NR, HSPA, and IMT-2020 (5G or New Radio), but also MS (Mobile Station), UT (User Terminal), SS (Subscriber Station), and wireless device in GSM. In addition, the terminal may be a user portable device such as a smartphone depending on the usage type, and in a V2X communication system, it may mean a vehicle, a device including a wireless communication module in the vehicle, etc. In addition, in the case of a Machine Type Communication system, it may mean an MTC terminal, M2M terminal, URLLC terminal, etc. that is equipped with a communication module to perform machine type communication.

[0035] The base station or cell in this specification refers to an end that communicates with a terminal in terms of a network, and includes various coverage areas such as Node-B, eNB (evolved Node-B), gNB (gNode-B), LPN (Low Power Node), Sector, Site, various types of antennas, BTS (Base Transceiver System), Access Point, Point (e.g., Transmission Point, Reception Point, Transmission / Reception Point), Relay Node, Mega Cell, Macro Cell, Micro Cell, Pico Cell, Femto Cell, RRH (Remote Radio Head), RU (Radio Unit), and Small Cell. In addition, a cell may mean including a BWP (Bandwidth Part) in the frequency domain. For example, a serving cell may mean an Activation BWP of a terminal.

[0036] Since the various cells listed above have a base station that controls one or more cells, the base station can be interpreted in two meanings. 1) It can be a device itself that provides a mega cell, macro cell, micro cell, pico cell, femto cell, or small cell in relation to a wireless area, or 2) it can indicate the wireless area itself. In 1), all devices that provide a given wireless area are controlled by the same entity or that interact to cooperatively configure the wireless area are all indicated as a base station. Depending on how the wireless area is configured, a point, a transceiver point, a transmission point, a reception point, etc. can be an embodiment of a base station. In 2), the wireless area itself that receives or transmits a signal from the perspective of a user terminal or a neighboring base station can also be indicated as a base station.

[0037] In this specification, a cell may mean a component carrier having coverage of a signal transmitted from a transmission / reception point or a transmission / reception point itself.

[0038] Uplink (UL, or uplink) refers to a method of transmitting and receiving data from a terminal to a base station, and downlink (DL, or downlink) refers to a method of transmitting and receiving data from a base station to a terminal. Downlink may refer to communication or a communication path from multiple transmission / reception points to a terminal, and uplink may refer to communication or a communication path from a terminal to multiple transmission / reception points. In this case, in the downlink, the transmitter may be part of the multiple transmission / reception points, and the receiver may be part of the terminal. In addition, in the uplink, the transmitter may be part of the terminal, and the receiver may be part of the multiple transmission / reception points.

[0039] Uplink and downlink transmit and receive control information through control channels such as PDCCH (Physical Downlink Control CHannel) and PUCCH (Physical Uplink Control CHannel), and transmit and receive data by configuring data channels such as PDSCH (Physical Downlink Shared CHannel) and PUSCH (Physical Uplink Shared CHannel). Hereinafter, the situation in which signals are transmitted and received through channels such as PUCCH, PUSCH, PDCCH, and PDSCH is also expressed in the form of 'transmitting and receiving PUCCH, PUSCH, PDCCH, and PDSCH'.

[0040] For clarity of explanation, the technical idea of ​​this invention is described below mainly with reference to the 3GPP LTE / LTE-A / NR (New RAT) communication system, but the technical features of this invention are not limited to the communication system.

[0041] After researching 4G (4th-Generation) communication technology, 3GPP develops 5G (5th-Generation) communication technology to meet the requirements of the next-generation wireless access technology of the ITU-R. Specifically, 3GPP develops LTE-A pro, which enhances LTE-Advanced technology to meet the requirements of the ITU-R, and NR, a new communication technology separate from 4G communication technology. Both LTE-A pro and NR refer to 5G communication technology, and in the following, 5G communication technology will be explained with NR as the focus, unless a specific communication technology is specifically mentioned.

[0042] The operating scenario in NR defines various operation scenarios by adding considerations for satellites, automobiles, and new verticals to the existing 4G LTE scenario, and in terms of service, it supports the eMBB (Enhanced Mobile Broadband) scenario, the mMTC (Massive Machine Communication) scenario that has high terminal density but is deployed over a wide area and requires low data rate and asynchronous access, and the URLLC (Ultra Reliability and Low Latency) scenario that requires high responsiveness and reliability and can support high-speed mobility.

[0043] To meet these scenarios, NR introduces a wireless communication system that incorporates new waveform and frame structure technologies, low latency technologies, support for ultra-high frequency bands (mmWave), and forward compatibility technologies. In particular, NR systems offer various technological changes in terms of flexibility to ensure forward compatibility. The key technical features of NR are described below with reference to the drawings.

[0044]

[0045] <NR 시스템 일반>

[0046] Figure 1 is a schematic diagram illustrating the structure of an NR system to which the present embodiment can be applied.

[0047] Referring to Fig. 1, the NR system is divided into 5GC (5G Core Network) and NR-RAN parts, and the NG-RAN is composed of gNBs and ng-eNBs that provide user plane (SDAP / PDCP / RLC / MAC / PHY) and control plane (RRC) protocol termination for UE (User Equipment). gNBs or gNBs and ng-eNBs are interconnected via the Xn interface. gNBs and ng-eNBs are each connected to the 5GC via the NG interface. The 5GC can be configured to include an AMF (Access and Mobility Management Function) that is responsible for the control plane such as terminal access and mobility control functions, and an UPF (User Plane Function) that is responsible for the control function for user data. NR includes support for both frequency bands below 6 GHz (FR1, Frequency Range 1) and frequency bands above 6 GHz (FR2, Frequency Range 2).

[0048] gNB refers to a base station that provides NR user plane and control plane protocol termination to terminals, and ng-eNB refers to a base station that provides E-UTRA user plane and control plane protocol termination to terminals. The base station described in this specification should be understood to encompass both gNB and ng-eNB, and may also be used to refer to gNB or ng-eNB separately as needed.

[0049] <NR 웨이브 폼, 뉴머롤러지 및 프레임 구조>

[0050]

[0051] *NR uses the CP-OFDM waveform with a cyclic prefix for downlink transmission, and CP-OFDM or DFT-s-OFDM for uplink transmission. OFDM technology is easily combined with MIMO (Multiple Input Multiple Output) and has the advantage of enabling the use of low-complexity receivers with high frequency efficiency.

[0052] Meanwhile, in NR, the requirements for data rates, latency, and coverage differ across the three scenarios mentioned above. Therefore, it is necessary to efficiently satisfy these requirements across the frequency bands that comprise any NR system. To this end, technologies have been proposed to efficiently multiplex radio resources based on multiple different numerologies.

[0053] Specifically, the NR transmission numerator is determined based on the sub-carrier spacing and the cyclic prefix (CP), and is changed exponentially with the μ value being an exponent value of 2 based on 15 kHz, as shown in Table 1 below.

[0054] μ서브캐리어 간격Cyclic prefixSupported for dataSupported for synch015NormalYesYes130NormalYesYes260Normal, ExtendedYesNo3120NormalYesYes4240NormalNoYes

[0055] As shown in Table 1 above, the numerology of NR can be divided into five types according to the subcarrier spacing. This is different from the fixed subcarrier spacing of LTE, one of the 4G communication technologies, at 15 kHz. Specifically, the subcarrier spacing used for data transmission in NR is 15, 30, 60, and 120 kHz, and the subcarrier spacing used for synchronization signal transmission is 15, 30, 12, and 240 kHz. In addition, the extended CP is applied only to the 60 kHz subcarrier spacing. Meanwhile, the frame structure in NR is defined as a 10 ms frame consisting of 10 subframes with the same length of 1 ms. One frame can be divided into half frames of 5 ms, and each half frame contains 5 subframes. In the case of a 15 kHz subcarrier spacing, one subframe consists of one slot, and each slot consists of 14 OFDM symbols. FIG. 2 is a diagram for explaining the frame structure in an NR system to which the present embodiment can be applied. Referring to FIG. 2, a slot is fixedly composed of 14 OFDM symbols in the case of a normal CP, but the length of the slot in the time domain may vary depending on the subcarrier spacing. For example, in the case of a numerology with a 15 kHz subcarrier spacing, a slot is composed of 1 ms, which is the same length as a subframe. In contrast, in the case of a numerology with a 30 kHz subcarrier spacing, a slot is composed of 14 OFDM symbols, but two slots may be included in one subframe with a length of 0.5 ms. That is, a subframe and a frame are defined with a fixed time length, and a slot is defined by the number of symbols, so the time length may vary depending on the subcarrier spacing.Meanwhile, NR defines slots as the basic scheduling unit and also introduces mini-slots (or sub-slots, or non-slot-based scheduling) to reduce transmission delay in the wireless section. Using wider subcarrier spacing reduces transmission delay in the wireless section by shortening the length of each slot inversely. Mini-slots (or sub-slots) are designed to efficiently support URLLC scenarios and allow scheduling in units of 2, 4, or 7 symbols.

[0056] Furthermore, unlike LTE, NR defines uplink and downlink resource allocation at the symbol level within a single slot. To reduce HARQ delay, a slot structure was defined that allows HARQ ACK / NACKs to be transmitted directly within the transmission slot. This slot structure is referred to as a self-contained structure and will be described in detail.

[0057] NR is designed to support a total of 256 slot formats, of which 62 are used in 3GPP Rel-15. It also supports a common frame structure that configures FDD or TDD frames through various combinations of slots. For example, it supports a slot structure in which all symbols in a slot are set to downlink, a slot structure in which all symbols are set to uplink, and a slot structure in which downlink and uplink symbols are combined. NR also supports data transmission being distributed and scheduled across one or more slots. Therefore, a base station can use a slot format indicator (SFI) to inform a UE whether a slot is a downlink slot, an uplink slot, or a flexible slot. The base station can indicate the slot format by indicating an index of a table configured through UE-specific RRC signaling using the SFI, and can also indicate it dynamically through DCI (Downlink Control Information) or statically or semi-statically through RRC.

[0058] <NR 물리 자원 >

[0059] In relation to physical resources in NR, antenna ports, resource grids, resource elements, resource blocks, and bandwidth parts are considered.

[0060] Antenna ports are defined such that the channel through which a symbol on an antenna port is carried can be inferred from the channel through which another symbol on the same antenna port is carried. Two antenna ports are said to be quasi co-located (or quasi co-located) if the large-scale properties of the channel through which a symbol on one antenna port is carried can be inferred from the channel through which a symbol on the other antenna port is carried. Here, the large-scale properties include one or more of delay spread, Doppler spread, frequency shift, average received power, and received timing.

[0061] FIG. 3 is a diagram for explaining a resource grid supported by a wireless access technology to which the present embodiment can be applied.

[0062] Referring to Figure 3, a resource grid may exist for each numeral, as NR supports multiple numerals on the same carrier. Furthermore, resource grids may exist based on antenna ports, subcarrier spacing, and transmission direction.

[0063] A resource block (RB) consists of 12 subcarriers and is defined solely in the frequency domain. Furthermore, a resource element (RE) consists of one OFDM symbol and one subcarrier. Therefore, as shown in Figure 3, the size of a single RB can vary depending on the subcarrier spacing. NR also defines "Point A," which serves as a common reference point for the RB grid, as well as common RBs and virtual RBs.

[0064] FIG. 4 is a diagram for explaining a bandwidth part supported by a wireless access technology to which the present embodiment can be applied.

[0065] Unlike LTE, where the carrier bandwidth is fixed at 20 MHz, NR sets the maximum carrier bandwidth from 50 MHz to 400 MHz for each subcarrier interval. Therefore, it is not assumed that all terminals will use the entire carrier bandwidth. Accordingly, NR allows terminals to designate bandwidth parts (BWPs) within the carrier bandwidth, as illustrated in Figure 4. Furthermore, bandwidth parts are associated with a single numerology, consist of a subset of consecutive common resource blocks, and can be dynamically activated over time. Each terminal is configured with up to four bandwidth parts for both the uplink and downlink, and data is transmitted and received using the bandwidth parts activated at a given time.

[0066] In the case of a paired spectrum, the uplink and downlink bandwidth parts are set independently, and in the case of an unpaired spectrum, the downlink and uplink bandwidth parts are set in pairs so that they can share a center frequency to prevent unnecessary frequency re-tuning between downlink and uplink operations.

[0067] <NR 초기 접속>

[0068] In NR, a terminal performs cell search and random access procedures to connect to a base station and perform communication.

[0069] Cell search is a procedure in which a terminal synchronizes to the cell of a corresponding base station, obtains a physical layer cell ID, and obtains system information using a synchronization signal block (SSB) transmitted by the base station.

[0070] FIG. 5 is a diagram illustrating an example of a synchronization signal block in a wireless access technology to which the present embodiment can be applied.

[0071] Referring to FIG. 5, SSB is composed of a primary synchronization signal (PSS) and a secondary synchronization signal (SSS), each occupying 1 symbol and 127 subcarriers, and a PBCH spanning 3 OFDM symbols and 240 subcarriers.

[0072] The terminal receives SSB by monitoring SSB in the time and frequency domain.

[0073] SSB can be transmitted up to 64 times in 5ms. Multiple SSBs are transmitted in different transmission beams within 5ms, and the terminal performs detection assuming that SSBs are transmitted every 20ms based on a specific beam used for transmission. The number of beams that can be used for SSB transmission within 5ms can increase as the frequency band increases. For example, up to 4 SSB beams can be transmitted below 3GHz, up to 8 in the frequency band between 3GHz and 6GHz, and up to 64 different beams can be used for SSB transmission in the frequency band above 6GHz.

[0074] SSB contains two symbols in one slot, and the starting symbol and number of repetitions within the slot are determined as follows depending on the subcarrier spacing.

[0075] Meanwhile, unlike SS in conventional LTE, SSB is not transmitted at the center frequency of the carrier bandwidth. This means that SSB can be transmitted even in locations other than the center of the system bandwidth, and when supporting wideband operation, multiple SSBs can be transmitted in the frequency domain. Accordingly, the terminal monitors SSB using the synchronization raster, which is a candidate frequency location for monitoring SSB. The carrier raster, which is the center frequency location information of the channel for initial access, and the synchronization raster are newly defined in NR. The synchronization raster has a wider frequency interval than the carrier raster, which can support the terminal's fast SSB search.

[0076] A UE can obtain the MIB through the PBCH of the SSB. The MIB (Master Information Block) includes the minimum information required for the UE to receive the remaining system information (RMSI, Remaining Minimum System Information) broadcast by the network. In addition, the PBCH may include information on the position of the first DM-RS symbol in the time domain, information for the UE to monitor SIB1 (e.g., SIB1 numerology information, information related to SIB1 CORESET, search space information, PDCCH-related parameter information, etc.), offset information between the common resource block and the SSB (the absolute position of the SSB within the carrier is transmitted through SIB1), etc. Here, the SIB1 numerology information is also applied equally to some messages used in the random access procedure for the UE to access the base station after completing the cell search procedure. For example, the numerology information of SIB1 may be applied to at least one of messages 1 to 4 for the random access procedure.

[0077] The aforementioned RMSI may refer to SIB1 (System Information Block 1), and SIB1 is broadcast periodically (e.g., every 160 ms) in the cell. SIB1 contains information necessary for the UE to perform the initial random access procedure and is periodically transmitted via PDSCH. In order for the UE to receive SIB1, it must receive numerology information used for SIB1 transmission and CORESET (Control Resource Set) information used for SIB1 scheduling via PBCH. The UE checks scheduling information for SIB1 using SI-RNTI in CORESET and acquires SIB1 on PDSCH according to the scheduling information. The remaining SIBs, excluding SIB1, may be transmitted periodically or upon request of the UE.

[0078] FIG. 6 is a diagram for explaining a random access procedure in a wireless access technology to which the present embodiment can be applied.

[0079] Referring to FIG. 6, once cell search is complete, the terminal transmits a random access preamble for random access to the base station. The random access preamble is transmitted via the PRACH. Specifically, the random access preamble is transmitted to the base station via the PRACH, which consists of consecutive radio resources in a specific slot that is periodically repeated. Generally, when a terminal initially accesses a cell, a contention-based random access procedure is performed, and when performing random access for beam failure recovery (BFR), a non-contention-based random access procedure is performed.

[0080] The terminal receives a random access response to the transmitted random access preamble. The random access response may include a random access preamble identifier (ID), an UL Grant (uplink radio resource), a temporary C-RNTI (Temporary Cell - Radio Network Temporary Identifier), and a TAC (Time Alignment Command). Since one random access response may include random access response information for one or more terminals, the random access preamble identifier may be included to indicate which terminal the included UL Grant, temporary C-RNTI, and TAC are valid for. The random access preamble identifier may be an identifier for the random access preamble received by the base station. The TAC may be included as information for the terminal to adjust uplink synchronization. The random access response may be indicated by a random access identifier on the PDCCH, i.e., an RA-RNTI (Random Access - Radio Network Temporary Identifier).

[0081] Upon receiving a valid random access response, the terminal processes the information contained in the random access response and performs scheduled transmission to the base station. For example, the terminal applies TAC and stores a temporary C-RNTI. Furthermore, using the UL Grant, the terminal transmits data stored in its buffer or newly generated data to the base station. In this case, information that identifies the terminal must be included.

[0082] Finally, the terminal receives a downlink message for contention resolution.

[0083] <NR CORESET>

[0084] The downlink control channel in NR is transmitted in a CORESET (Control Resource Set) with a length of 1 to 3 symbols, and transmits uplink / downlink scheduling information, SFI (Slot format Index), and TPC (Transmit Power Control) information.

[0085] To ensure system flexibility, NR introduced the CORESET concept. CORESET (Control Resource Set) refers to time-frequency resources for downlink control signals. A terminal can decode control channel candidates using one or more search spaces within the CORESET time-frequency resources. A QCL (Quasi CoLocation) assumption is established for each CORESET, which is used to inform the characteristics of analog beam direction in addition to the delay spread, Doppler spread, Doppler shift, and average delay assumed by the conventional QCL.

[0086] Figure 7 is a drawing for explaining CORESET.

[0087] Referring to Figure 7, a CORESET can exist in various forms within the carrier bandwidth within a single slot, and in the time domain, a CORESET can consist of up to three OFDM symbols. In addition, a CORESET is defined as a multiple of six resource blocks up to the carrier bandwidth in the frequency domain.

[0088] The first CORESET is indicated via the MIB as part of the initial bandwidth part configuration, allowing the terminal to receive additional configuration and system information from the network. After establishing a connection with the base station, the terminal can receive and configure one or more CORESET information via RRC signaling.

[0089] In this specification, the terms frequency, frame, subframe, resource, resource block, region, band, subband, control channel, data channel, synchronization signal, various reference signals, various signals or various messages related to NR (New Radio) may be interpreted in the past or present meaning or in various meanings used in the future.

[0090] Wider bandwidth operations

[0091] Existing LTE systems supported scalable bandwidth operation for any LTE Component Carrier (CC). That is, depending on the deployment scenario, any LTE operator could configure a single LTE CC with a bandwidth ranging from a minimum of 1.4 MHz to a maximum of 20 MHz, and a normal LTE terminal supported transmission and reception capabilities of 20 MHz bandwidth for a single LTE CC.

[0092] However, in the case of NR, the design is made to support NR terminals with different transmission and reception bandwidth capabilities through a single wideband NR CC, and accordingly, it is required to configure one or more bandwidth parts (BWP, bandwidth part(s)) consisting of segmented bandwidths for any NR CC, and to support flexible wider bandwidth operation through different bandwidth part configurations and activations for each terminal.

[0093] Specifically, in NR, one or more bandwidth parts can be configured through one serving cell configured from the terminal's perspective, and the terminal is defined to activate one downlink bandwidth part (DL bandwidth part) and one uplink bandwidth part (UL bandwidth part) in the serving cell to use them for uplink / downlink data transmission and reception. In addition, when multiple serving cells are configured in the terminal, that is, for the terminal to which CA is applied, it is defined to activate one downlink bandwidth part and / or uplink bandwidth part for each serving cell to use the radio resources of the serving cell to use them for uplink / downlink data transmission and reception.

[0094] Specifically, an initial bandwidth part for an initial access procedure of a terminal in an arbitrary serving cell is defined, one or more UE-specific bandwidth part(s) are configured for each terminal through dedicated RRC signaling, and a default bandwidth part for a fallback operation can also be defined for each terminal.

[0095] However, it can be defined that multiple downlink and / or uplink bandwidth parts can be activated and used simultaneously depending on the capability and bandwidth part(s) configuration of the terminal in any serving cell, but in NR rel-15, it is defined that only one downlink bandwidth part (DL bandwidth part) and one uplink bandwidth part (UL bandwidth part) can be activated and used in any terminal at any time.

[0096]

[0097] The present disclosure relates to a method for ensuring the performance of an AI / ML model of a terminal in one or more cells. The present disclosure proposes a method for ensuring consistency between conditions between training and inference when applying a model trained based on data collection provided by a network (NW) to inference in a system in which a terminal performs wireless communication using an AI / ML model. In particular, based on an associated ID associated with data collection, if this is guaranteed, even if the terminal moves to a different cell other than the cell in which training was performed, the trained model can be applied to the new cell as long as the new cell provides the same data and network configuration conditions as the trained cell.

[0098] The following terms can be defined for AI / ML-based wireless communications:

[0099] Data collection refers to the process by which network nodes, management entities, or UEs collect data for the purpose of AI / ML model training, data analysis, and inference.

[0100] An AI / ML model (hereinafter also referred to as a "model") is a data-driven algorithm that applies AI / ML technology to generate a series of outputs based on a series of inputs. AI / ML model training refers to the process of training an AI / ML model in a data-driven manner by learning input / output relationships and obtaining a trained AI / ML model for inference. AI / ML model inference refers to the process of using a trained AI / ML model to generate a series of outputs based on a series of inputs. AI / ML model validation refers to the subprocess of training that evaluates the quality of an AI / ML model using a dataset different from the one used for model training. AI / ML model testing refers to the subprocess of training that evaluates the performance of the final AI / ML model using a dataset different from the one used for model training and validation. Unlike AI / ML model validation, testing does not assume subsequent adjustments to the model.

[0101] The UE-side (AI / ML) model refers to an AI / ML model where inference is performed entirely in the UE. The Network-side (AI / ML) model refers to an AI / ML model where inference is performed entirely in the network. The One-sided (AI / ML) model refers to either the UE-side (AI / ML) model or the network-side (AI / ML) model. The Two-sided (AI / ML) model refers to a pair of AI / ML models where joint inference is performed. Here, joint inference consists of AI / ML inference where inference is performed jointly across the UE and the network. That is, the first part of inference is performed first by the UE and the remaining part by the gNB or vice versa.

[0102] AI / ML model transfer refers to the transmission of an AI / ML model over a wireless interface, either with parameters of a model structure known to the receiver or with a new model with parameters. The transfer may include a complete model or a partial model. Model download refers to the transmission of a model from the network to the UE. Model upload refers to the transmission of a model from the UE to the network.

[0103] Federated learning / federated training refers to a machine learning technique that trains AI / ML models on multiple distributed edge nodes (e.g., UEs, gNBs), each performing local model training using local data samples. This requires multiple model interactions but does not require the exchange of local data samples. Offline field data refers to data collected in the field and used for offline training of AI / ML models. Online field data refers to data collected in the field and used for online training of AI / ML models.

[0104] Model monitoring refers to the process of monitoring the inference performance of AI / ML models.

[0105] Supervised learning refers to the process of training a model using inputs and their corresponding labels. Unsupervised learning refers to the process of training a model without labeled data. Semi-supervised learning refers to the process of training a model using a mixture of labeled and unlabeled data. Reinforcement learning (RL) refers to the process of training an AI / ML model from inputs (i.e., states) and feedback signals (i.e., rewards) resulting from the model's outputs (i.e., actions) in an environment in which the model interacts.

[0106] Model activation refers to activating an AI / ML model for a specific function. Model deactivation refers to deactivating an AI / ML model for a specific function. Model switching refers to deactivating the currently activated AI / ML model and activating a different AI / ML model for a specific function.

[0107] When applying AI / ML models, the following network-UE collaboration levels are considered.

[0108] 1. Level x: No collaboration.

[0109] 2. Level y: Signaling-based collaboration without model transfer.

[0110] 3. Level z: Signal-based collaboration through model transfer.

[0111] In relation to life cycle management (LCM) procedures for AI / ML models, an AI / ML model may have a model ID with relevant information and / or model functionality for at least some AI / ML operations.

[0112] Model selection, activation, deactivation, switching, and replacement for both UE-side and bilateral models may be initiated by the network, if determined by the network, or initiated by the UE and requested by the network. If determined by the UE, the UE's decision may be reported to the network based on events configured by the network.

[0113] For AI / ML-based features / FGs (feature groups), additional conditions refer to all aspects assumed for model learning, but are not part of the terminal capabilities (UE) for the AI / ML-based features / FGs. This does not necessarily mean that additional conditions are explicitly specified. Additional conditions can be divided into two categories: network-side additional conditions and UE-side additional conditions.

[0114] For the inference of the UE-side model, the following options can be taken as possible approaches to ensure consistency between learning and inference with respect to additional NW-side conditions (if identified):

[0115] - Identification of a model to achieve alignment for additional conditions on the NW side between the NW side and the UE side.

[0116] - The model learned under additional conditions is trained in NW and transferred to UE.

[0117] - Provide information and / or instructions to the UE regarding additional conditions on the NW side.

[0118] - Consistency is supported by monitoring (model / feature selection through performance of candidate models / features on UE side by UE and / or NW).

[0119] - Other approaches are not ruled out.

[0120] - It is not denied that different approaches can achieve the same function.

[0121] In relation to data collection, it can be defined as follows:

[0122] For the UE-side AI / ML model on the UE side, the UE reports to the NW its support / preference configuration for downlink reference signal (DL RS) transmission. Regarding data collection trigger / start, data collection can be initiated / triggered by the NW's configuration or by the UE's request for data collection.

[0123] Signaling aspects for data collection, for example, signaling aspects relate to assistance information (if supported), reference signals, content / type of data collected, configuration related to Set A and / or Set B, and information about the association / mapping of Set A and Set B.

[0124] Support information (if available) provided by the network to the UE for UE data collection to classify data for the purpose of differentiating data characteristics. Support information must protect privacy / proprietary information.

[0125] For NW-side AI / ML models on the NW side, reporting-related mechanisms, additional information about the report content, reporting overhead reduction, signals / configuration / measurement / reporting for data collection, e.g., signal aspects are related to support information (if supported), reference signals.

[0126] Regarding data collection for the NW-side AI / ML models of BM-Case1 and BM-Case2, the following approaches for overhead reduction are identified:

[0127] - Omission / selection of collected data

[0128] - Compression of collected data

[0129] - If the purpose of data collection is different, the overhead reduction mechanism and the resulting specification impact may be different.

[0130] - For each LCM purpose, which mechanisms are supported (if any) and their potential specification implications (if any) are the subject of separate discussion.

[0131] Regarding data collection for the NW-side AI / ML models of BM-Case1 and BM-Case2, the following reporting signals for beam-specific aspects may be applied:

[0132] - L1 signal to report collected data

[0133] - Higher-layer signals to report collected data

[0134] - At least not applicable to AI / ML model inference

[0135] - Existing signaling principles (e.g. RSRP reporting on L1) can be reused.

[0136] RAN1 studies model identification type A, including more details related to use cases.

[0137] RAN1 explores the following options for model identification type B as a starting point, including more details relevant to all use cases:

[0138] - MI-Option 1: Model identification along with data collection related configuration and / or instructions.

[0139] - MI-Option 2: Model Identification with Dataset Transfer

[0140] - MI-Option 3: Model identification in model transfer from NW to UE

[0141] - The names (MI-Option 1, MI-Option 2, MI-Option 3) are used for discussion purposes only.

[0142] - Other options are proposed for model identification type B:

[0143] - MI-Option 4: Model Identification through Standardization of Reference Models (for CSI Compression)

[0144] - MI-Option 5: Model Identification through Model Monitoring

[0145] Regarding MI-Option 1 (Model Identification with Data Collection Related Configuration and / or Instructions) of Model Identification Type B, RAN1 further explores the following aspects:

[0146] - Relationship between model ID and data collection related configuration and / or instructions.

[0147] - Information transmitted from NW to UE (if any)

[0148] - Information transmitted from UE to NW (if any)

[0149] - Related procedures

[0150] - Use cases / applicable uses of MI-Option 1

[0151] For Model Identification Type B of MI-Option 1 (including data collection configuration and / or instructions related to model identification), RAN1 further studies the following aspects:

[0152] - Relationship between model ID and data collection related configuration and / or instructions.

[0153] - Information transmitted from the network (NW) to the UE (if any)

[0154] - Information transmitted from UE to network (NW) (if any)

[0155] - Use cases where MI-Option 1 is used or applicable

[0156] From a RAN1 perspective, for a UE-side model developed (e.g., trained, updated) on the UE side, the following procedure is an example (AI-Example 1) for further study (including feasibility / necessity) of MI-Option 1.

[0157] - A: For data collection, NW signals the data collection related configuration and its / their associated ID.

[0158] An association ID for each sub-use case associated with the NW-side additional conditions.

[0159] - B: UE collects data corresponding to the associated ID.

[0160] - C: AI / ML models are developed (e.g., trained, updated) on the UE side based on collected data corresponding to the associated ID.

[0161] - D: The UE reports its AI / ML model information corresponding to the associated ID to the NW. A model ID is determined / assigned for each AI / ML model.

[0162] Relationship between Model ID and Association ID

[0163] How the model ID is determined / assigned, for example, NW assigns the model ID, UE assigns / reports the model ID, or the association ID is considered as the model ID, and D's "model ID is determined / assigned for each AI / ML model" is not required, and the model ID is determined according to predefined rules.

[0164] D is to facilitate AI / ML model inference.

[0165] Additional interactions of steps A / B / C and association IDs between UE and NW can be considered as other solutions for consistency resolution without model identification.

[0166] With respect to the association ID, the UE assumes that the NW-side additional conditions with the same association ID are consistent at least within the cell. Further research is needed to determine whether and how the UE's assumption can be applied across multiple cells (including feasibility studies).

[0167] To ensure consistency of NW-side additional conditions throughout the learning and inference of the UE-side models for BM-Case 1 and BM-Case 2, either association ID-based or performance monitoring-based methods can be defined.

[0168]

[0169] Currently, 3GPP is considering associated ID-based and monitoring-based approaches to ensure the consistency of network-side additional conditions applied between UE-sided model inference and training. Associated IDs are defined as being consistent within at least one cell and can be applied across multiple cells. The term "associated ID" in this disclosure is used as an example and is not intended to be limiting.

[0170] According to an example procedure for associated ID based on the discussion of Model ID, a terminal entering a cell receives the associated ID provided by the cell and related data collection configuration information from the base station, and collects the data to perform model training / update. If such associated ID is information applicable only within a specific cell, it means that the terminal must perform model training every time it enters a new cell. This means that model-based beam management techniques are performed through model training for the new cell every time the terminal performs a handover (HO). This requires mobile terminals to perform training every time they move from cell to cell, which can be a significant burden for mobile terminals that require AI / ML model-based communication.

[0171] If an associated ID is applicable to more than one cell, a UE can perform model-based beam management quickly by applying a model trained with the same associated ID to the new cell without model training, based on the information about the same associated ID, even when entering a new cell. However, even if model inference is possible based on the data configuration for the same associated ID, i.e., the same NW-side additional condition, it may be difficult to ensure consistency between training and inference when inferring a model trained in a different cell in a new cell. Additional model updates may be required due to different channel environments for each cell, and therefore, an efficient method is needed.

[0172] Below, a method of performing communication using artificial intelligence and machine learning will be described with reference to relevant drawings.

[0173] FIG. 8 is a diagram illustrating a procedure (800) in which a terminal performs communication using artificial intelligence and machine learning according to one embodiment.

[0174] Referring to FIG. 8, the terminal receives a handover command message from the first base station, including at least one associated ID supported by the second base station for the AI / ML model (S810), and can perform a handover to the second base station based on the handover command message (S820).

[0175] For example, it is assumed that a handover procedure is performed when a terminal moves from a first base station, which is a serving base station, to a second base station. In this case, the handover to the second base station itself can be performed according to a known handover procedure. That is, the first base station is the serving base station to which the terminal is currently connected in communication, and the second base station corresponds to the target base station to which the handover will be performed as the terminal moves. Hereinafter, the first base station may also be referred to as a base station or a serving base station. Additionally, the second base station may also be referred to as a target base station.

[0176] The terminal measures the signal strength of the serving cell and neighboring cells based on pre-determined configuration information through upper layer signaling, etc., and transmits a measurement report message to the first base station when a predetermined event condition is met.

[0177] In addition, in order for the terminal to use the AI / ML model used for communication with the first base station for communication with the second base station after the handover, the terminal may transmit information about the trained AI / ML model stored in the terminal to the first base station. The information about the AI / ML model may include at least one associated ID for the AI / ML model. In one example, the associated ID may be identification information about settings such as data collection used to train the AI / ML model. In this case, the term is an example and is not limited thereto. When the same associated ID is set for the AI / ML model, it may mean that the same settings and network-side additional conditions (NW-side additional conditions) are applied to one or more base stations for training the AI / ML model stored in the terminal.

[0178] An association ID for the AI / ML model that the terminal uses for communication with the second base station after a handover needs to be determined. In order for the terminal to use the trained AI / ML model used for communication with the first base station for communication with the second base station as well, the association ID for the AI / ML model used for communication with the first base station must also be supported by the second base station. If the association ID is not supported by the second base station, the terminal must retrain the AI / ML model according to data collection settings, etc. identified by the new association ID. Alternatively, if training the AI / ML model is difficult, the terminal can communicate with the second base station according to a fallback mode that does not use the AI / ML model.

[0179] For example, the association ID for the AI / ML model to be used by the terminal for communication with the second base station after the handover may be configured to be determined by any one of the terminal, the first base station, and the second base station. That is, whether the association ID for the AI / ML model stored in the terminal matches at least one association ID supported by the second base station may be confirmed by any one of the terminal, the first base station, and the second base station.

[0180] For example, assume that a terminal is configured to determine an association ID to be used after a handover. In this case, the terminal may receive neighbor cell information from the first base station or neighboring cells, including at least one association ID supported by each of the adjacent cells.

[0181] The terminal can check whether at least one association ID of each of the surrounding cells matches at least one association ID for an AI / ML model stored in the terminal. If there is a matching association ID, the terminal can report to the first base station the at least one association ID corresponding to the reported cell in the aforementioned measurement report message. Alternatively, the terminal can transmit to the first base station, separately from the measurement report message, information about the AI / ML model stored in the terminal, including the matching association ID for each cell. If there is no matching association ID, at least one new association ID for a functionality / feature corresponding to the association ID of the terminal can be determined for each cell and transmitted to the first base station.

[0182] The first base station determines the need for a handover based on the received measurement report. If a handover is determined to be necessary, the first base station transmits a handover request message to the second base station. In this case, the handover request message may include the aforementioned matching association ID or a new association ID. That is, the first base station may transmit a handover request message to the second base station that includes at least one association ID for the AI / ML model stored in the terminal.

[0183] When the second base station completes handover preparations, such as resource allocation, based on the handover request message, the second base station may transmit a handover request ACK message to the first base station, including at least one association ID supported by the second base station for the AI / ML model. In this case, if there is a matching association ID as described above, the at least one association ID supported by the second base station may be the matching association ID as described above. If there is no matching association ID as described above, the at least one association ID supported by the second base station may be the new association ID as described above.

[0184] If the aforementioned matching association ID exists, the handover request confirmation message may include the aforementioned matching association ID and configuration information for model monitoring and / or model inference based on the aforementioned matching association ID. If the aforementioned matching association ID does not exist, the handover request confirmation message may include the aforementioned new association ID and configuration information for model training based on the aforementioned new association ID.

[0185] As another example, assume that a first base station is configured to determine an association ID to be used after a handover. In this case, the first base station may receive neighboring cell information from neighboring cells, each of which includes at least one association ID supported by the neighboring cells.

[0186] As described above, the terminal may transmit information about the trained AI / ML model stored in the terminal to the first base station. The information about the AI / ML model may include at least one associated ID for the AI / ML model.

[0187] The first base station determines the need for a handover based on the received measurement report. If a handover is determined to be necessary, the first base station can verify that at least one associated ID of each neighboring cell matches at least one associated ID for the AI / ML model stored in the terminal. Furthermore, the first base station transmits a handover request message to the second base station.

[0188] If a matching association ID is found, the first base station may transmit a handover request message to the second base station, which includes an indicator indicating a request for configuration information for model monitoring and / or model inference for the matching association ID. If a matching association ID is not found, the first base station may transmit a handover request message to the second base station, which includes a new association ID for a functionality / feature corresponding to the association ID of the terminal and an indicator indicating a request for configuration information for model training corresponding to the new association ID.

[0189] When the second base station completes handover preparations, such as resource allocation, based on the handover request message, the second base station can transmit a handover request confirmation message including at least one associated ID supported by the second base station for the AI / ML model to the first base station.

[0190] If the aforementioned matching association ID exists, the handover request confirmation message may include the aforementioned matching association ID and configuration information for model monitoring and / or model inference based on the aforementioned matching association ID. If the aforementioned matching association ID does not exist, the handover request confirmation message may include the aforementioned new association ID and configuration information for model training based on the aforementioned new association ID.

[0191] As another example, assume that the second base station is configured to determine an associated ID to be used after a handover. In this case, as described above, the terminal may transmit information about a trained AI / ML model stored in the terminal to the first base station. The information about the AI / ML model may include at least one associated ID for the AI / ML model.

[0192] If the first base station determines that a handover is necessary based on the received measurement report, it transmits a handover request message to the second base station. In this case, the handover request message may include at least one associated ID for the terminal's AI / ML model.

[0193] The second base station can verify whether at least one association ID for the AI / ML model of the terminal included in the handover request message matches at least one association ID supported by the second base station. When handover preparations, such as resource allocation, are completed based on the handover request message, the second base station can transmit a handover request confirmation message including at least one association ID for the AI / ML model supported by the second base station to the first base station.

[0194] If the aforementioned matching association ID exists, the handover request confirmation message may include the aforementioned matching association ID and configuration information for model monitoring and / or model inference based on the aforementioned matching association ID. If the aforementioned matching association ID does not exist, the handover request confirmation message may include the aforementioned new association ID and configuration information for model training based on the aforementioned new association ID.

[0195] Based on the handover request confirmation message, the first base station may transmit a handover command message to the terminal, which includes at least one associated ID supported by the second base station. The handover command message may include at least one of data collection-related configuration information for the at least one associated ID, configuration information for training, monitoring, and / or inference of a stored AI / ML model, and network-side additional condition information.

[0196] If there is a matching association ID as described above, the handover instruction message may include the matching association ID as described above and configuration information for model monitoring and / or model inference based on the matching association ID. If there is no matching association ID as described above, the handover instruction message may include the new association ID as described above and configuration information for model training based on the matching association ID.

[0197] When a terminal receives a handover instruction message from the first base station, it disconnects from the first base station and attempts to connect to the second base station. If the terminal successfully connects to the second base station, the second base station requests a path change from the core network and then releases the terminal-related resources of the first base station, completing the handover procedure.

[0198] Referring again to FIG. 8, if the association ID for the AI / ML model stored in the terminal matches at least one association ID, the terminal may perform performance monitoring for the stored AI / ML model (S830) and transmit a message including the monitoring result to the second base station (S840).

[0199] The terminal can verify, based on the handover instruction message, whether the association ID for the AI / ML model stored in the terminal matches at least one association ID.

[0200] If at least one associated ID matches the associated ID for the AI / ML model stored in the terminal, the terminal may perform performance verification through model monitoring for the stored AI / ML model based on model monitoring configuration information and report the performance results to the second base station. Based on the report, the second base station may instruct the terminal to perform a model update through additional training of the model corresponding to the associated ID, or to perform communication through model inference. Accordingly, the terminal may perform an AI / ML model update or inference and use the AI / ML model for communication with the base station.

[0201] Alternatively, if the associated ID for the AI / ML model stored in the terminal does not match at least one associated ID, the terminal may perform data collection corresponding to the new associated ID described above based on the model training configuration information. The terminal may perform training of the stored AI / ML model based on the collected data and report the training results to the second base station. Thereafter, the terminal may communicate with the second base station using the newly trained AI / ML model.

[0202] According to this, a method and device for performing communication using artificial intelligence and machine learning in a next-generation wireless access network can be provided. Furthermore, overall system performance can be improved by minimizing the overhead of performing model inference through data collection and model training each time a terminal moves between cells.

[0203] FIG. 9 is a diagram illustrating a procedure (900) for a base station to perform communication using artificial intelligence and machine learning according to one embodiment. The descriptions previously described in FIG. 8 may be omitted to avoid redundant explanations. In this case, the omitted content may be substantially equally applied to the base station, as long as it does not conflict with the technical spirit of the invention.

[0204] Referring to FIG. 9, the base station can transmit a handover request message including at least one associated ID for an AI / ML model stored in the terminal to the target base station (S910).

[0205] As described above, it is assumed that a handover procedure is performed as the terminal moves from the base station, i.e., the serving base station, to the target base station. In this case, the handover to the target base station itself can be performed according to a known handover procedure.

[0206] The terminal measures the signal strength of the serving cell and neighboring cells based on pre-determined configuration information, such as through upper-layer signaling. A measurement report message transmitted upon fulfillment of a specified event condition is received from the terminal.

[0207] In addition, in order to use the AI / ML model used by the terminal for communication with the base station for communication with the target base station after a handover, the base station may receive information about the trained AI / ML model stored in the terminal from the terminal. The information about the AI / ML model may include at least one associated ID for the AI / ML model. In one example, the associated ID may be identification information about settings such as data collection used to train the AI / ML model. In this case, the term is an example and is not limited thereto. When the same associated ID is set for the AI / ML model, it may mean that the same settings and network-side additional conditions (NW-side additional conditions) are applied to one or more base stations for training the AI / ML model stored in the terminal.

[0208] The associated ID for the AI / ML model that the terminal uses for communication with the target base station after a handover needs to be determined. In order for the terminal to use the trained AI / ML model used for communication with the serving base station for communication with the target base station as well, the associated ID for the AI / ML model used for communication with the serving base station must also be supported by the target base station. If the target base station does not support the associated ID, the terminal must retrain the AI / ML model according to data collection settings, etc. identified by the new associated ID. Alternatively, if training the AI / ML model is difficult, the terminal can communicate with the target base station in a fallback mode that does not use the AI / ML model.

[0209] For example, the association ID for the AI / ML model to be used by the terminal for communication with the target base station after handover may be configured to be determined by any one of the terminal, the serving base station, and the target base station. That is, whether the association ID for the AI / ML model stored in the terminal matches at least one association ID supported by the target base station may be confirmed by any one of the terminal, the serving base station, and the target base station.

[0210] For example, assume that a terminal is configured to determine an association ID to be used after a handover. In this case, the terminal may receive neighbor cell information from the serving base station or neighboring cells, including at least one association ID supported by each of the adjacent cells.

[0211] The terminal can check whether at least one association ID of each surrounding cell matches at least one association ID for an AI / ML model stored in the terminal. If there is a matching association ID, the serving base station can receive from the terminal the measurement report message described above, including at least one association ID corresponding to the reported cell. Alternatively, the serving base station can receive from the terminal, separately from the measurement report message, information about the AI / ML model stored in the terminal, including the matching association ID for each cell. If there is no matching association ID, the terminal can determine at least one new association ID for each cell corresponding to the terminal's association ID and transmit the determined association ID to the serving base station.

[0212] The serving base station determines the need for a handover based on the received measurement report. If a handover is determined necessary, the serving base station transmits a handover request message to the target base station. In this case, the handover request message may include the aforementioned matching association ID or a new association ID. In other words, the serving base station may transmit a handover request message to the target base station that includes at least one association ID for the AI / ML model stored in the terminal.

[0213] As another example, assume that a serving base station is configured to determine an association ID to be used after a handover. In this case, the serving base station may receive neighboring cell information from neighboring cells, including at least one association ID supported by each of the neighboring cells.

[0214] As described above, the serving base station may receive information about a trained AI / ML model stored in the terminal from the terminal. The information about the AI / ML model may include at least one associated ID for the AI / ML model.

[0215]

[0216] *The serving base station determines the need for a handover based on the received measurement report. If a handover is determined necessary, the serving base station can verify that at least one association ID of each neighboring cell matches at least one association ID for the AI / ML model stored in the terminal. Additionally, the serving base station transmits a handover request message to the target base station.

[0217] If a matching association ID is found, the serving base station may transmit a handover request message to the target base station, which includes an indicator indicating a request for configuration information for model monitoring and / or model inference for the matching association ID. If a matching association ID is not found, the serving base station may transmit a handover request message to the target base station, which includes a new association ID for a functionality / feature corresponding to the association ID of the terminal and an indicator indicating a request for configuration information for model training corresponding to the new association ID.

[0218] As another example, assume that the target base station is configured to determine an association ID to be used after a handover. In this case, as described above, the serving base station may receive information about a trained AI / ML model stored in the terminal from the terminal. The information about the AI / ML model may include at least one association ID for the AI / ML model.

[0219] If the serving base station determines that a handover is necessary based on the received measurement report, it transmits a handover request message to the target base station. In this case, the handover request message may include at least one associated ID for the terminal's AI / ML model.

[0220] Referring again to FIG. 9, the base station may receive a handover request ACK message from the target base station, which includes at least one association ID supported by the target base station for the AI / ML model (S920).

[0221] For example, if the target base station is configured to determine an association ID to be used after a handover, as described above, the target base station may complete preparations for the handover, such as resource allocation, based on the handover request message. The serving base station may receive a handover request ACK message from the target base station, including at least one association ID supported by the target base station for the AI / ML model. In this case, if there is a matching association ID as described above, the at least one association ID supported by the target base station may be the matching association ID as described above. If there is no matching association ID as described above, the at least one association ID supported by the target base station may be the new association ID as described above.

[0222] If the aforementioned matching association ID exists, the handover request confirmation message may include the aforementioned matching association ID and configuration information for model monitoring and / or model inference based on the aforementioned matching association ID. If the aforementioned matching association ID does not exist, the handover request confirmation message may include the aforementioned new association ID and configuration information for model training based on the aforementioned new association ID.

[0223] In another example, as described above, if the target base station is configured to determine an association ID to be used after a handover from the serving base station, the target base station may complete handover preparations, such as resource allocation, based on the handover request message. The serving base station may receive a handover request confirmation message from the target base station, including at least one association ID supported by the target base station for the AI / ML model.

[0224] If the aforementioned matching association ID exists, the handover request confirmation message may include the aforementioned matching association ID and configuration information for model monitoring and / or model inference based on the aforementioned matching association ID. If the aforementioned matching association ID does not exist, the handover request confirmation message may include the aforementioned new association ID and configuration information for model training based on the aforementioned new association ID.

[0225] In another example, when the target base station is configured to determine an association ID to be used after a handover as described above, the target base station can check whether at least one association ID for the AI / ML model of the terminal included in the handover request message matches at least one association ID supported by the target base station. When the handover preparation, such as resource allocation based on the handover request message at the target base station, is completed, the serving base station can receive a handover request confirmation message from the target base station, including at least one association ID for the AI / ML model supported by the target base station.

[0226] If the aforementioned matching association ID exists, the handover request confirmation message may include the aforementioned matching association ID and configuration information for model monitoring and / or model inference based on the aforementioned matching association ID. If the aforementioned matching association ID does not exist, the handover request confirmation message may include the aforementioned new association ID and configuration information for model training based on the aforementioned new association ID.

[0227] Referring again to FIG. 9, the base station can transmit a handover command message to the terminal, which includes at least one associated ID supported by the target base station (S930).

[0228] Based on the handover request confirmation message, the serving base station may transmit a handover command message to the terminal, the handover command message including at least one associated ID supported by the target base station. The handover command message may include at least one of data collection-related configuration information for the at least one associated ID, configuration information for training, monitoring, and / or inference of a stored AI / ML model, and network-side additional condition information.

[0229] If there is a matching association ID as described above, the handover instruction message may include the matching association ID as described above and configuration information for model monitoring and / or model inference based on the matching association ID. If there is no matching association ID as described above, the handover instruction message may include the new association ID as described above and configuration information for model training based on the matching association ID.

[0230] When a terminal receives a handover instruction message from the serving base station, it disconnects from the serving base station and attempts to connect to the target base station. If the terminal successfully connects to the target base station, the target base station requests a path change from the core network and then releases terminal-related resources from the serving base station, completing the handover process.

[0231] If at least one associated ID matches an associated ID for an AI / ML model stored in the terminal, the terminal may perform performance monitoring on the stored AI / ML model and transmit a message including the monitoring results to the target base station. Based on the handover instruction message, the terminal may verify whether at least one associated ID matches an associated ID for the AI / ML model stored in the terminal.

[0232] If at least one associated ID matches the associated ID for the AI / ML model stored in the terminal, the terminal may perform performance verification through model monitoring for the stored AI / ML model based on model monitoring configuration information and report the performance results to the target base station. Based on the report, the target base station may instruct the terminal to perform a model update through additional training of the model corresponding to the associated ID, or to perform communication through model inference. Accordingly, the terminal may perform an AI / ML model update or inference and use the AI / ML model for communication with the target base station.

[0233] Alternatively, if the associated ID for the AI / ML model stored in the terminal does not match at least one associated ID, the terminal may perform data collection corresponding to the new associated ID described above based on the model training configuration information. The terminal may perform training of the stored AI / ML model based on the collected data and report the training results to the target base station. Thereafter, the terminal may communicate with the target base station using the newly trained AI / ML model.

[0234] According to this, a method and device for performing communication using artificial intelligence and machine learning in a next-generation wireless access network can be provided. Furthermore, overall system performance can be improved by minimizing the overhead of performing model inference through data collection and model training each time a terminal moves between cells.

[0235]

[0236] Below, with reference to the relevant drawings, each embodiment related to a method of performing communication using artificial intelligence and machine learning will be described in detail.

[0237] The associated ID associated with data collection used to train the AI / ML model of a terminal proposed in this disclosure assumes that the same configuration and network-side additional conditions (NW-side additional conditions) apply to one or more cells. This assumes that the same configuration and NW-side additional conditions apply within the same Public Land Mobile Network (PLMN), or that the same information is shared in advance between each base station through predefined configuration information within the specification, or through reception of predefined configuration information from a specific server.

[0238] In more detail, the terminal proposed in the present disclosure can collect data corresponding to any associated ID, and perform model inference after verifying performance through model performance monitoring for an AI / ML model trained using the collected data when entering a cell that provides data for the same associated ID. In addition, the terminal proposed in the present disclosure can perform beam management based on the new model after performing new model training when moving to a cell that does not have a data collection-related setting for the same associated ID, or can operate in fallback mode (i.e., NR beam management mode that does not use the AI / ML model). If the associated ID(s) corresponding to the model stored / trained in the terminal are not supported by the target base station but support data corresponding to a new associated ID for the same functionality / feature, the terminal can perform model training by collecting data for the new associated ID, and the model trained in this way can be defined as a new model. If the target base station does not support the associated ID for the trained model and does not support new associated ID-based data corresponding to the same functionality / feature for the trained model, the terminal may operate in a fallback mode that does not use the AI / ML model.

[0239] For the above-described operation, the UE must compare / verify whether the associated ID(s) for the model(s) stored in the UE are also supported / valid in the target cell before (or during, or immediately after) cell movement. In addition, the UE must provide the base station with configuration information in advance for model-related procedures that the UE must perform after entering the cell (e.g., related model monitoring, new model training, model inference, etc.) based on whether the associated ID(s) for the models stored / trained in the UE are also supported in the target gNB.

[0240] The technology of the present disclosure proposes that a terminal capable of performing AI / ML model inference performs a process of comparing / verifying associated ID(s) for stored / trained model(s) with associated ID(s) supported by a target base station before moving to a new cell. This can be performed using one of the following methods.

[0241] The following examples 1, 2, and 3 are methods for receiving model training / monitoring / inference-related configuration information for an associated ID supported by a target base station in advance before a handover (HO), and determining and starting a model-related procedure to be performed with the target base station after the HO based on the configuration of the target base station.

[0242]

[0243] Example 1. When the target base station determines the associated ID(s) of the terminal

[0244] FIG. 10 and FIG. 11 are diagrams for explaining an operation when a target base station determines an associated ID according to one embodiment.

[0245] The terminal reports to the base station the associated ID(s) corresponding to the UE-sided model(s) stored / trained in the terminal. This may be included as UE context information, or, upon completion of training, may be information reported to the base station as model information. In other words, the serving base station assumes that it has previously received associated ID information for the terminal's stored / trained model from the terminal.

[0246] When the serving base station determines the HO of the terminal, it transmits a handover request (HO request) message including the associated ID(s) information of the terminal to the target base station(s). If the target base station supports any of the associated ID(s) of the terminal, it transmits a handover request acknowledgment (HO request ACK) message including the associated ID(s) and the corresponding configuration information for model monitoring and / or model inference to the serving base station. This means that a handover instruction (HO command) (i.e., RRCReconfiguration) message including the associated ID(s) corresponding to the model(s) stored / trained in the terminal is transmitted from the serving base station to the terminal.

[0247] If there is no associated ID(s) supported by the target base station among the associated ID(s) of the terminal, a HO request ACK message is transmitted to the serving base station, including new associated ID(s) for the functionality / feature corresponding to the associated ID(s) of the terminal and the corresponding configuration information for model training. This means that there is no model(s) stored / trained in the terminal that can be applied to the target base station, and that communication using a new model must be performed through new model training.

[0248] Example 2. When the serving base station determines the associated ID(s) of the terminal

[0249] FIG. 12 and FIG. 13 are diagrams for explaining an operation when a serving base station determines an association ID according to one embodiment.

[0250] The terminal reports to the base station the associated ID(s) corresponding to the UE-sided model(s) stored / trained in the terminal. This may be included as part of the UE context information, or, if training is complete, may be information reported to the base station as part of the model information. In other words, the serving base station assumes that it has previously received the associated ID information for the terminal's stored / trained model from the terminal.

[0251] The serving base station receives information about neighboring cells in advance. The neighboring cell information may include associated ID(s) supported by the neighboring cells. When HO of the terminal is determined, the associated ID(s) of the terminal are compared / verified with the associated ID(s) of the target base station. If identical associated ID(s) exist, the HO request message including an indicator requesting configuration information for model monitoring / model inference for the associated ID(s) is transmitted to the target base station. Through this, the target base station confirms that the associated ID(s) it supports exist among the associated ID(s) of the terminal and transmits an HO request ACK message including the associated ID(s) and the corresponding configuration information for model monitoring and / or model inference to the serving base station. This means that an HO command message including the associated ID(s) corresponding to the model(s) stored / trained in the terminal is transmitted from the target base station to the terminal via the serving base station.

[0252] If there is no associated ID(s) supported by the target base station among the associated ID(s) of the terminal, an HO request message including a new associated ID(s) for the functionality / feature corresponding to the associated ID(s) of the terminal and an indicator requesting configuration information for corresponding model training is transmitted to the target base station. The target base station receiving this confirms that there is no associated ID supported by it among the associated ID(s) of the terminal and that model training is required through a new associated ID, and transmits an HO request ACK message including model training configuration information for the associated ID requested from the serving base station to the serving base station. This means that there is no model(s) stored / trained in the terminal that can be applied to the target base station, and that communication using a new model must be performed through new model training.

[0253] Example 3. When the terminal determines the associated ID(s) of the terminal.

[0254]

[0255] *Figures 14 and 15 are diagrams for explaining the operation when a terminal determines a related ID according to one embodiment.

[0256] The terminal acquires information about neighboring cells, including associated ID(s) of the neighboring cells. If the associated ID(s) of the neighboring cells match the associated ID(s) of the model stored / trained in the terminal and the cell reports a measurement result, the terminal reports a measurement report message including the associated ID(s) corresponding to the reported cell (expressed as available associated ID in Fig. 14) to the base station. If the base station determines HO based on the measurement reporting of the terminal, it transmits an HO request message including an indicator requesting the target base station to include configuration information for model monitoring corresponding to the available associated ID(s) of the terminal in the HO command. The terminal, which receives the HO command message including the associated ID(s) of the model stored / trained in the terminal from the target base station, performs model monitoring / inference based on the related configuration information after the HO.

[0257] If monitoring is performed, the terminal can report the model performance results to the base station, and based on the model monitoring results, the base station can instruct the terminal to perform model update by transmitting data corresponding to the associated ID, or to perform communication through model inference.

[0258] Through one of the three embodiments described above, a terminal that receives an HO command message including an associated ID for its trained model performs model monitoring and / or model inference for the model corresponding to the received associated ID(s) after performing the HO. That is, a terminal that receives an HO command message including an associated ID for its trained model can recognize that the AI / ML model corresponding to the associated ID can be applied as is even after the HO to the target base station. Alternatively, a terminal that receives an HO command message including a new associated ID, other than the associated ID for its trained model, can recognize that after the HO to the target base station, it must train the AI / ML model by collecting data corresponding to the new associated ID and report information about the new model to the base station. If a HO command that does not include any associated ID is received, the terminal can recognize that it operates in fallback mode after the HO to the target base station. Here, if the HO Command message includes at least one associated ID, it may additionally include configuration information for model monitoring / inference or model training for the associated ID, which means that the terminal can perform the instructed procedure based on the configuration information after the HO.

[0259] That is, in the present disclosure, based on the reception of an HO command message including associated ID(s), the terminal performs model training or model monitoring / inference by collecting data for the received associated ID(s) based on the received configuration information after moving to the target base station.

[0260] If the associated ID(s) corresponding to the associated ID(s) for the models stored / trained on the terminal are received in the HO command and the terminal performs model monitoring after the HO, the terminal can report the model monitoring performance results to the base station. Based on this report, the base station can instruct the terminal to perform a model update through additional training of the model corresponding to the associated ID, or to perform communication through model inference.

[0261] The following is an example of the operation of a terminal and a base station according to embodiments of the present disclosure.

[0262]

[0263] [Terminal operation]

[0264] - Receive a HO command message containing at least one associated ID from the base station.

[0265] May contain settings information related to data collection for Associated ID.

[0266] The configuration information contains at least one of the following information related to model training / monitoring / inference.

[0267] It may include NW-side additional condition information for the associated ID.

[0268] - Perform handover to the target base station.

[0269] - If the above associated ID matches the associated ID for your trained model,

[0270] Perform model monitoring for the model corresponding to the above associated ID.

[0271] Report monitoring performance results to the base station.

[0272] Model updates or inferences can be performed based on performance results.

[0273] - If not,

[0274] Perform data collection corresponding to the above associated ID.

[0275] Model training is performed based on the collected data.

[0276] Report the model training results to the base station.

[0277]

[0278] [Target Base Station Operation]

[0279] - Transmits a HO command message containing at least one associated ID to the terminal.

[0280] May contain settings information related to data collection for Associated ID.

[0281] The configuration information contains at least one of the following information related to model training / monitoring / inference.

[0282] It may include NW-side additional condition information for the associated ID.

[0283] - Confirm that the terminal is connected to the base station (e.g., via RACH).

[0284] - If the above associated ID matches the associated ID for the trained model of the terminal,

[0285] Transmits model monitoring related data for the model corresponding to the above associated ID.

[0286] Receive monitoring performance results from the terminal.

[0287] You can instruct it to update the model or perform inference based on the performance results.

[0288] - If not,

[0289] Start transmitting data for model training corresponding to the above associated ID.

[0290] Receive model training results (e.g., model information including model ID, associated ID, etc.) from the terminal.

[0291]

[0292] The following describes a method for determining a model procedure based on an associated ID according to a terminal request after a HO to a target base station during HO.

[0293] Example 4. When receiving the associated ID(s) of the target base station through the HO Command or cell information

[0294] FIG. 16 and FIG. 17 are diagrams for explaining an operation of receiving an association ID of a target base station according to one embodiment.

[0295] The terminal receives the associated ID(s) supported by the target base station via an HO command message or cell information. The terminal checks whether there is a model trained with data corresponding to the associated ID(s) supported by the target base station among its trained / stored models. If there is a model, the terminal requests model monitoring / inference for the model corresponding to the associated ID(s) from the base station after HO. When the terminal receives model monitoring configuration information corresponding to the associated ID(s) of the model trained / stored in the terminal, the terminal performs model monitoring and reports the performance results to the base station. Based on the report, the base station may instruct the terminal to perform a model update through additional training of the model corresponding to the associated ID, or to perform communication through model inference.

[0296] The terminal checks whether a model trained with data corresponding to the associated ID(s) supported by the target base station exists among its trained / stored models. If not, it requests model training for a model corresponding to the new associated ID(s) after an HO. Upon receiving this, the base station can instruct model training by sending a message containing configuration information for model training to the terminal.

[0297] As described above, the present disclosure proposes a method for comparing / verifying the associated ID of a terminal with the associated ID of a target base station, as a method for applying a trained model based on data collection corresponding to an arbitrary associated ID of a terminal and supporting this model in the cell to which the terminal has moved, as is, after HO. This has the effect of improving overall system performance by minimizing the overhead of performing model inference through data collection and model training each time the terminal moves cells.

[0298]

[0299] Hereinafter, the configuration of a terminal and a base station capable of performing some or all of the embodiments described with reference to FIGS. 1 to 17 will be described with reference to the drawings. The above description may be omitted to avoid redundant description, and in this case, the omitted content may be substantially equally applied to the following description, as long as it does not contradict the technical spirit of the invention.

[0300] Fig. 18 is a drawing showing the configuration of a terminal (1800) according to another embodiment.

[0301] Referring to FIG. 18, a terminal (1800) according to another embodiment includes a transmitter (1820), a receiver (1830), and a control unit (1810) that controls the transmitter and receiver.

[0302] The control unit (1810) controls the overall operation of the terminal (1800) according to the method of performing communication using artificial intelligence and machine learning necessary to perform the aforementioned embodiments.

[0303] The transmitter (1820) and receiver (1830) are used to transmit and receive signals, messages, and data necessary to perform the aforementioned embodiments with the base station.

[0304] The control unit (1810) may receive a handover command message from the first base station, including at least one associated ID supported by the second base station for the AI / ML model. The control unit (1810) may perform a handover to the second base station based on the handover command message.

[0305] The control unit (1810) measures the signal strength of the serving cell and the adjacent cell based on pre-determined configuration information through upper layer signaling, etc., and transmits a measurement report message to the first base station when a predetermined event condition is met.

[0306] Additionally, in order to enable the terminal to use the AI / ML model used for communication with the first base station for communication with the second base station after handover, the control unit (1810) may transmit information about the trained AI / ML model stored in the terminal to the first base station. The information about the AI / ML model may include at least one associated ID for the AI / ML model.

[0307] For example, the association ID for the AI / ML model to be used by the terminal for communication with the second base station after the handover may be configured to be determined by any one of the terminal, the first base station, and the second base station. That is, whether the association ID for the AI / ML model stored in the terminal matches at least one association ID supported by the second base station may be confirmed by any one of the terminal, the first base station, and the second base station.

[0308] For example, assume that a terminal is configured to determine an association ID to be used after a handover. In this case, the control unit (1810) may receive neighbor cell information from the first base station or neighboring cells, including at least one association ID supported by each of the adjacent cells.

[0309] The control unit (1810) can check whether at least one association ID of each surrounding cell matches at least one association ID for an AI / ML model stored in the terminal. If there is a matching association ID, the control unit (1810) can report to the first base station the at least one association ID corresponding to the reported cell in the aforementioned measurement report message. Alternatively, the control unit (1810) can transmit to the first base station, separately from the measurement report message, information about the AI / ML model stored in the terminal, including the matching association ID for each cell. If there is no matching association ID, at least one new association ID for a functionality / feature corresponding to the association ID of the terminal can be determined for each cell and transmitted to the first base station.

[0310] The first base station determines the need for a handover based on the received measurement report. If a handover is determined to be necessary, the first base station transmits a handover request message to the second base station. In this case, the handover request message may include the aforementioned matching association ID or a new association ID. That is, the first base station may transmit a handover request message to the second base station that includes at least one association ID for the AI / ML model stored in the terminal.

[0311] When the second base station completes handover preparations, such as resource allocation, based on the handover request message, the second base station may transmit a handover request ACK message to the first base station, including at least one association ID supported by the second base station for the AI / ML model. In this case, if there is a matching association ID as described above, the at least one association ID supported by the second base station may be the matching association ID as described above. If there is no matching association ID as described above, the at least one association ID supported by the second base station may be the new association ID as described above.

[0312] If the aforementioned matching association ID exists, the handover request confirmation message may include the aforementioned matching association ID and configuration information for model monitoring and / or model inference based on the aforementioned matching association ID. If the aforementioned matching association ID does not exist, the handover request confirmation message may include the aforementioned new association ID and configuration information for model training based on the aforementioned new association ID.

[0313] As another example, assume that a first base station is configured to determine an association ID to be used after a handover. In this case, the first base station may receive neighboring cell information from neighboring cells, each of which includes at least one association ID supported by the neighboring cells.

[0314] As described above, the control unit (1810) may transmit information about the trained AI / ML model stored in the terminal to the first base station. The information about the AI / ML model may include at least one associated ID for the AI / ML model.

[0315] The first base station determines the need for a handover based on the received measurement report. If a handover is determined to be necessary, the first base station can verify that at least one associated ID of each neighboring cell matches at least one associated ID for the AI / ML model stored in the terminal. Furthermore, the first base station transmits a handover request message to the second base station.

[0316] If a matching association ID is found, the first base station may transmit a handover request message to the second base station, which includes an indicator indicating a request for configuration information for model monitoring and / or model inference for the matching association ID. If a matching association ID is not found, the first base station may transmit a handover request message to the second base station, which includes a new association ID for a functionality / feature corresponding to the association ID of the terminal and an indicator indicating a request for configuration information for model training corresponding to the new association ID.

[0317] When the second base station completes handover preparations, such as resource allocation, based on the handover request message, the second base station can transmit a handover request confirmation message including at least one associated ID supported by the second base station for the AI / ML model to the first base station.

[0318] If the aforementioned matching association ID exists, the handover request confirmation message may include the aforementioned matching association ID and configuration information for model monitoring and / or model inference based on the aforementioned matching association ID. If the aforementioned matching association ID does not exist, the handover request confirmation message may include the aforementioned new association ID and configuration information for model training based on the aforementioned new association ID.

[0319] As another example, assume that the second base station is configured to determine an associated ID to be used after a handover. In this case, as described above, the control unit (1810) may transmit information about a trained AI / ML model stored in the terminal to the first base station. The information about the AI / ML model may include at least one associated ID for the AI / ML model.

[0320] If the first base station determines that a handover is necessary based on the received measurement report, it transmits a handover request message to the second base station. In this case, the handover request message may include at least one associated ID for the terminal's AI / ML model.

[0321] The second base station can verify whether at least one association ID for the AI / ML model of the terminal included in the handover request message matches at least one association ID supported by the second base station. When handover preparations, such as resource allocation, are completed based on the handover request message, the second base station can transmit a handover request confirmation message including at least one association ID for the AI / ML model supported by the second base station to the first base station.

[0322] If the aforementioned matching association ID exists, the handover request confirmation message may include the aforementioned matching association ID and configuration information for model monitoring and / or model inference based on the aforementioned matching association ID. If the aforementioned matching association ID does not exist, the handover request confirmation message may include the aforementioned new association ID and configuration information for model training based on the aforementioned new association ID.

[0323] Based on the handover request confirmation message, the first base station may transmit a handover command message to the terminal, which includes at least one associated ID supported by the second base station. The handover command message may include at least one of data collection-related configuration information for the at least one associated ID, configuration information for training, monitoring, and / or inference of a stored AI / ML model, and network-side additional condition information.

[0324] If there is a matching association ID as described above, the handover instruction message may include the matching association ID as described above and configuration information for model monitoring and / or model inference based on the matching association ID. If there is no matching association ID as described above, the handover instruction message may include the new association ID as described above and configuration information for model training based on the matching association ID.

[0325] When the control unit (1810) receives a handover instruction message from the first base station, it releases the wireless connection with the first base station and attempts to establish a wireless connection to the second base station. If the terminal successfully connects to the second base station, the second base station requests a path change to the core network and then releases the terminal-related resources of the first base station, thereby completing the handover procedure.

[0326] The control unit (1810) may perform performance monitoring on the stored AI / ML model when the association ID for the AI / ML model stored in the terminal matches at least one association ID, and transmit a message including the monitoring result to the second base station.

[0327] The control unit (1810) can check whether the association ID for the AI / ML model stored in the terminal matches at least one association ID based on the handover instruction message.

[0328] If at least one associated ID matches the associated ID for the AI / ML model stored in the terminal, the control unit (1810) may perform performance verification through model monitoring on the stored AI / ML model based on the model monitoring configuration information and report the performance results to the second base station. Based on the report, the second base station may instruct the terminal to perform a model update through additional training of the model corresponding to the associated ID, or to perform communication through model inference. Accordingly, the control unit (1810) may perform an update or inference of the AI / ML model and use the corresponding AI / ML model for communication with the base station.

[0329] Alternatively, if the associated ID for the AI / ML model stored in the terminal does not match at least one associated ID, the control unit (1810) may perform data collection corresponding to the new associated ID described above based on the model training configuration information. The control unit (1810) may perform training of the stored AI / ML model based on the collected data and report the training performance results to the second base station. Thereafter, the control unit (1810) may perform communication with the second base station using the newly trained AI / ML model.

[0330] According to this, a method and device for performing communication using artificial intelligence and machine learning in a next-generation wireless access network can be provided. Furthermore, overall system performance can be improved by minimizing the overhead of performing model inference through data collection and model training each time a terminal moves between cells.

[0331] Fig. 19 is a drawing showing the configuration of a base station (1900) according to another embodiment.

[0332] Referring to FIG. 19, a base station (1900) according to another embodiment includes a transmitter (1920), a receiver (1930), and a control unit (1910) that controls the transmitter and receiver.

[0333] The control unit (1910) controls the overall operation of the base station (1900) and the operation of the repeater according to the method of performing communication using artificial intelligence and machine learning necessary to perform the aforementioned embodiments.

[0334] The transmitter (1920) and receiver (1930) are used to transmit and receive signals, messages, and data necessary for performing the above-described embodiments to and from the terminal.

[0335] The control unit (1910) can transmit a handover request message including at least one associated ID for an AI / ML model stored in the terminal to the target base station.

[0336] As described above, it is assumed that a handover procedure is performed as the terminal moves from the base station, i.e., the serving base station, to the target base station. In this case, the handover to the target base station itself can be performed according to a known handover procedure.

[0337] The terminal measures the signal strength of the serving cell and neighboring cells based on pre-determined configuration information, such as through upper-layer signaling. A measurement report message transmitted upon fulfillment of a specified event condition is received from the terminal.

[0338] Additionally, in order to use the AI / ML model used by the terminal for communication with the base station for communication with the target base station after handover, the control unit (1910) may receive information about the trained AI / ML model stored in the terminal from the terminal. The information about the AI / ML model may include at least one associated ID for the AI / ML model.

[0339] For example, the association ID for the AI / ML model to be used by the terminal for communication with the target base station after handover may be configured to be determined by any one of the terminal, the serving base station, and the target base station. That is, whether the association ID for the AI / ML model stored in the terminal matches at least one association ID supported by the target base station may be confirmed by any one of the terminal, the serving base station, and the target base station.

[0340] For example, assume that a terminal is configured to determine an association ID to be used after a handover. In this case, the terminal may receive neighbor cell information from the serving base station or neighboring cells, including at least one association ID supported by each of the adjacent cells.

[0341] The terminal can check whether at least one association ID of each surrounding cell matches at least one association ID for an AI / ML model stored in the terminal. If there is a matching association ID, the control unit (1910) can receive from the terminal the measurement report message described above, including at least one association ID corresponding to the cell being reported. Alternatively, the control unit (1910) can receive from the terminal, separately from the measurement report message, information about the AI / ML model stored in the terminal, including the matching association ID for each cell. If there is no matching association ID, the terminal can determine at least one new association ID for a functionality / feature corresponding to the association ID of the terminal for each cell and transmit it to the serving base station.

[0342] The control unit (1910) determines the necessity of a handover based on the received measurement report. If a handover is determined to be necessary, the control unit (1910) transmits a handover request message to the target base station. In this case, the handover request message may include the aforementioned matching association ID or a new association ID. That is, the control unit (1910) may transmit a handover request message to the target base station that includes at least one association ID for the AI / ML model stored in the terminal.

[0343] As another example, assume that the serving base station is configured to determine an association ID to be used after a handover. In this case, the control unit (1910) may receive neighboring cell information from neighboring cells, including at least one association ID supported by each of the neighboring cells.

[0344] As described above, the control unit (1910) may receive information about a trained AI / ML model stored in the terminal from the terminal. The information about the AI / ML model may include at least one associated ID for the AI / ML model.

[0345] The control unit (1910) determines the need for a handover based on the received measurement report. If a handover is determined to be necessary, the control unit (1910) can verify whether at least one associated ID of each neighboring cell matches at least one associated ID for the AI / ML model stored in the terminal. Furthermore, the control unit (1910) transmits a handover request message to the target base station.

[0346] If a matching association ID is found, the control unit (1910) may transmit a handover request message to the target base station, which includes an instruction indicating a request for configuration information for model monitoring and / or model inference for the matching association ID. If a matching association ID is not found, the control unit (1910) may transmit a handover request message to the target base station, which includes an instruction indicating a request for a new association ID for a functionality / feature corresponding to the association ID of the terminal and configuration information for model training corresponding to the new association ID.

[0347] As another example, assume that the target base station is configured to determine an associated ID to be used after a handover. In this case, as described above, the control unit (1910) may receive information about a trained AI / ML model stored in the terminal from the terminal. The information about the AI / ML model may include at least one associated ID for the AI / ML model.

[0348] If the control unit (1910) determines that a handover is necessary based on the received measurement report, it transmits a handover request message to the target base station. In this case, the handover request message may include at least one associated ID for the AI / ML model of the terminal.

[0349] The control unit (1910) may receive a handover request ACK message from the target base station, which includes at least one association ID supported by the target base station for the AI / ML model.

[0350] For example, if the terminal is configured to determine an association ID to be used after a handover, as described above, the target base station may complete preparations for the handover, such as resource allocation, based on the handover request message. The control unit (1910) may receive, from the target base station, a handover request ACK message including at least one association ID supported by the target base station for the AI / ML model. In this case, if there is a matching association ID as described above, the at least one association ID supported by the target base station may be the matching association ID as described above. If there is no matching association ID as described above, the at least one association ID supported by the target base station may be the new association ID as described above.

[0351] If the aforementioned matching association ID exists, the handover request confirmation message may include the aforementioned matching association ID and configuration information for model monitoring and / or model inference based on the aforementioned matching association ID. If the aforementioned matching association ID does not exist, the handover request confirmation message may include the aforementioned new association ID and configuration information for model training based on the aforementioned new association ID.

[0352] In another example, as described above, if the target base station is configured to determine an association ID to be used after a handover from the serving base station, the target base station may complete handover preparations, such as resource allocation, based on the handover request message. The control unit (1910) may receive a handover request confirmation message from the target base station, including at least one association ID supported by the target base station for the AI / ML model.

[0353] If the aforementioned matching association ID exists, the handover request confirmation message may include the aforementioned matching association ID and configuration information for model monitoring and / or model inference based on the aforementioned matching association ID. If the aforementioned matching association ID does not exist, the handover request confirmation message may include the aforementioned new association ID and configuration information for model training based on the aforementioned new association ID.

[0354] In another example, when the target base station is configured to determine an association ID to be used after a handover as described above, the target base station can check whether at least one association ID for the AI / ML model of the terminal included in the handover request message matches at least one association ID supported by the target base station. When the handover preparation, such as resource allocation based on the handover request message at the target base station, is completed, the control unit (1910) can receive a handover request confirmation message from the target base station, including at least one association ID supported by the target base station for the AI / ML model.

[0355] If the aforementioned matching association ID exists, the handover request confirmation message may include the aforementioned matching association ID and configuration information for model monitoring and / or model inference based on the aforementioned matching association ID. If the aforementioned matching association ID does not exist, the handover request confirmation message may include the aforementioned new association ID and configuration information for model training based on the aforementioned new association ID.

[0356] The control unit (1910) can transmit a handover command message including at least one associated ID supported by the target base station to the terminal.

[0357] Based on the handover request confirmation message, the control unit (1910) may transmit a handover command message to the terminal, the handover command message including at least one associated ID supported by the target base station. The handover command message may include at least one of data collection-related configuration information for at least one associated ID, configuration information for training, monitoring, and / or inference of a stored AI / ML model, and network-side additional condition information.

[0358] If there is a matching association ID as described above, the handover instruction message may include the matching association ID as described above and configuration information for model monitoring and / or model inference based on the matching association ID. If there is no matching association ID as described above, the handover instruction message may include the new association ID as described above and configuration information for model training based on the matching association ID.

[0359] When a terminal receives a handover instruction message from the serving base station, it disconnects from the serving base station and attempts to connect to the target base station. If the terminal successfully connects to the target base station, the target base station requests a path change from the core network and then releases terminal-related resources from the serving base station, completing the handover process.

[0360] If at least one associated ID matches an associated ID for an AI / ML model stored in the terminal, the terminal may perform performance monitoring on the stored AI / ML model and transmit a message including the monitoring results to the target base station. Based on the handover instruction message, the terminal may verify whether at least one associated ID matches an associated ID for the AI / ML model stored in the terminal.

[0361] If at least one associated ID matches the associated ID for the AI / ML model stored in the terminal, the terminal may perform performance verification through model monitoring for the stored AI / ML model based on model monitoring configuration information and report the performance results to the target base station. Based on the report, the target base station may instruct the terminal to perform a model update through additional training of the model corresponding to the associated ID, or to perform communication through model inference. Accordingly, the terminal may perform an AI / ML model update or inference and use the AI / ML model for communication with the target base station.

[0362] Alternatively, if the associated ID for the AI / ML model stored in the terminal does not match at least one associated ID, the terminal may perform data collection corresponding to the new associated ID described above based on the model training configuration information. The terminal may perform training of the stored AI / ML model based on the collected data and report the training results to the target base station. Thereafter, the terminal may communicate with the target base station using the newly trained AI / ML model.

[0363] According to this, a method and device for performing communication using artificial intelligence and machine learning in a next-generation wireless access network can be provided. Furthermore, overall system performance can be improved by minimizing the overhead of performing model inference through data collection and model training each time a terminal moves between cells.

[0364] The above-described embodiments may be supported by standard documents disclosed in at least one of the wireless access systems, IEEE 802, 3GPP, and 3GPP2. That is, steps, components, and parts not described in the present embodiments to clearly illustrate the technical concepts herein may be supported by the above-described standard documents. Furthermore, all terms disclosed in this specification may be explained by the above-described standard documents.

[0365] The embodiments described above may be implemented through various means. For example, the embodiments may be implemented through hardware, firmware, software, or a combination thereof.

[0366] In the case of hardware implementation, the method according to the present embodiments 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, or microprocessors.

[0367] When implemented using firmware or software, the methods according to the present embodiments may be implemented in the form of devices, procedures, or functions that perform the functions or operations described above. The software code may be stored in a memory unit and executed by a processor. The memory unit may be located within or outside the processor and may exchange data with the processor using various known means.

[0368] Additionally, terms such as "system," "processor," "controller," "component," "module," "interface," "model," or "unit" as described above may generally refer to a computer-related entity, such as hardware, a combination of hardware and software, software, or software in execution. For example, the aforementioned components may be, but are not limited to, a process driven by a processor, a processor, a controller, a control processor, an object, a thread of execution, a program, and / or a computer. For example, both an application running on a controller or a processor and the controller or the processor may be components. One or more components may be within a process and / or thread of execution, and the components may be located on a single device (e.g., a system, a computing device, etc.) or distributed across two or more devices.

[0369] The above description is merely an illustrative example of the technical idea of ​​the present disclosure, and those skilled in the art to which the present disclosure pertains will appreciate that various modifications and variations can be made without departing from the essential characteristics of the technical idea of ​​the present disclosure. In addition, the present embodiments are not intended to limit the technical idea of ​​the present disclosure but rather to explain it, and therefore the scope of the technical idea of ​​the present disclosure is not limited by these embodiments. The scope of protection of the present disclosure should be interpreted by the claims below, and all technical ideas within a scope equivalent thereto should be interpreted as being included within the scope of the rights of the present disclosure.

[0370]

[0371] CROSS-REFERENCE TO RELATED APPLICATION

[0372] This patent application claims priority under 35 USC §119(a) to Korean Patent Application No. 10-2024-0086332, filed in Korea on July 1, 2024, and Korean Patent Application No. 10-2025-0087427, filed in Korea on June 30, 2025, the entire contents of which are incorporated herein by reference. In addition, this patent application claims priority in countries other than the United States for the same reasons, the entire contents of which are incorporated herein by reference.

Claims

1. A method for a terminal (user equipment; UE) to perform communication using artificial intelligence and machine learning (AI / ML), A step of receiving, from a first base station, a handover command message including at least one associated ID supported by a second base station for an AI / ML model; A step of performing a handover to the second base station based on the handover instruction message; A step of performing performance monitoring for the stored AI / ML model when the association ID for the AI / ML model stored in the terminal matches at least one of the association IDs; and A method comprising the step of transmitting a message including a monitoring result to the second base station.

2. In paragraph 1, The above handover instruction message is, A method comprising at least one of data collection related configuration information for the at least one associated ID, configuration information for training, monitoring and / or inference of the stored AI / ML model, and network-side additional condition information.

3. In paragraph 1, A method further comprising a step of performing an update or inference of the stored AI / ML model based on the monitoring results.

4. In paragraph 1, A method further comprising the step of performing data collection corresponding to the at least one association ID when the association ID for the AI / ML model stored in the terminal does not match the at least one association ID.

5. In paragraph 4, A method for training the stored AI / ML model based on the collected data and reporting the training results to the second base station.

6. In paragraph 1, Whether the association ID for the AI / ML model stored in the terminal matches at least one of the association IDs above, A method confirmed by any one of the terminal, the first base station and the second base station.

7. In a method for a base station to perform communication using artificial intelligence and machine learning (AI / ML), A step of transmitting a handover request message including at least one associated ID for an AI / ML model stored in a terminal to a target base station; A step of receiving a handover request ACK message from the target base station, the handover request ACK message including at least one associated ID supported by the target base station for the AI / ML model; and A method comprising the step of transmitting a handover command message including at least one associated ID supported by the target base station to the terminal.

8. In paragraph 7, The above handover request confirmation message is: A method comprising: at least one association ID supported by the target base station and configuration information for monitoring and / or inferring an AI / ML model corresponding to at least one association ID supported by the target base station, when the association ID for the stored AI / ML model matches at least one association ID supported by the target base station.

9. In paragraph 7, The above handover request confirmation message is: A method comprising, when the association ID for the stored AI / ML model does not match at least one association ID supported by the target base station, at least one new association ID supported by the target base station and configuration information for training the AI / ML model corresponding to the new association ID.

10. In paragraph 7, The above handover instruction message is, A method comprising at least one of configuration information related to data collection for an associated ID for the stored AI / ML model, configuration information for training, monitoring, and / or inference of the stored AI / ML model, and network-side additional condition information.

11. In a terminal (user equipment; UE) that performs communication using artificial intelligence and machine learning (AI / ML), Transmitter; Receiver; and Including a control unit that controls the operation of the above transmitter and receiver, The above control unit, Receive a handover command message from a first base station, including at least one associated ID supported by a second base station for an AI / ML model, Performing a handover to the second base station based on the handover instruction message; If the association ID for the AI / ML model stored in the terminal matches at least one of the association IDs, performance monitoring is performed for the stored AI / ML model, A terminal that transmits a message including the monitoring result to the second base station.

12. In paragraph 11, The above handover instruction message is, A terminal including at least one of data collection related configuration information for at least one associated ID, configuration information for training, monitoring and / or inference of the stored AI / ML model, and network-side additional condition information.

13. In paragraph 11, The above control unit, A terminal that performs update or inference of the stored AI / ML model based on the above monitoring results.

14. In paragraph 11, The above control unit, A terminal that performs data collection corresponding to at least one association ID when the association ID for the AI / ML model stored in the terminal does not match the at least one association ID.

15. In paragraph 14, A terminal that performs training of the stored AI / ML model based on the collected data and reports the training results to the second base station.

16. In paragraph 11, Whether the association ID for the AI / ML model stored in the terminal matches at least one of the association IDs above, A terminal identified in any one of the above terminals, the first base station, and the second base station.

Images