Method and device for measurement prediction using cell geographic information in mobile communication system

Abstract

The present disclosure relates to a method for cell measurement prediction in a wireless communication system. The method is performed by a user equipment (UE) and comprises the steps of: receiving, from a base station, radio resource management (RRM) configuration information including cell geographic information; performing RRM prediction by using the cell geographic information and cell measurement information as inputs to an artificial intelligence / machine learning (AI / ML) model; detecting triggering of a measurement report event on the basis of an RRM measurement result and an RRM prediction result; and transmitting, to the base station, a measurement report message including the RRM measurement result and the RRM prediction result.

Description

Measurement prediction method and device utilizing cell geographic information in a mobile communication system

[0001] The present disclosure relates to a wireless communication system (or, mobile communication system). More specifically, the present disclosure relates to a method and apparatus that support an operation for predicting measurement results by utilizing cell geographical information in a wireless communication system (or, mobile communication system).

[0002] 5G mobile communication technology defines a wide frequency band to enable fast transmission speeds and new services, and can be implemented not only in frequency bands below 6 GHz ('Sub 6 GHz'), such as 3.5 gigahertz (3.5 GHz), but also in ultra-high frequency bands called millimeter waves (mmWave), such as 28 GHz and 39 GHz ('Above 6 GHz'). In addition, for 6G mobile communication technology, which is referred to as a system beyond 5G, implementation in the terahertz band (e.g., the 3 terahertz (3 THz) band at 95 GHz) is being considered to achieve transmission speeds 50 times faster and ultra-low latency reduced to one-tenth compared to 5G mobile communication technology.

[0003] In the early stages of 5G mobile communication technology, aiming to satisfy service support and performance requirements for enhanced Mobile BroadBand (eMBB), Ultra-Reliable Low-Latency Communications (URLLC), and massive Machine-Type Communications (mMTC), technologies such as beamforming and Massive MIMO to mitigate path loss and increase transmission distance in ultra-high frequency bands, support for various numerologies (such as the operation of multiple subcarrier spacings) and dynamic operation of slot formats for the efficient utilization of ultra-high frequency resources, initial access techniques to support multi-beam transmission and broadband, definition and operation of Band-Width Parts (BWP), Low Density Parity Check (LDPC) codes for high-volume data transmission, new channel coding methods such as Polar Codes for the reliable transmission of control information, and L2 pre-processing (L2 Standardization has been carried out for pre-processing, network slicing which provides a dedicated network specialized for specific services, and other methods.

[0004] Currently, discussions are underway to improve and enhance the performance of the initial 5G mobile communication technology, taking into account the services that the 5G mobile communication technology was intended to support. Additionally, standardization of the physical layer is in progress for technologies such as V2X (Vehicle-to-Everything), which helps autonomous vehicles make driving decisions and enhance user convenience based on their own location and status information transmitted by the vehicle; NR-U (New Radio Unlicensed), which aims for system operation in unlicensed bands to comply with various regulatory requirements; NR terminal low power consumption technology (UE Power Saving); Non-Terrestrial Network (NTN), which is direct terminal-satellite communication for securing coverage in areas where communication with the terrestrial network is impossible; and positioning.

[0005] In addition, standardization is underway in the field of wireless interface architecture / protocols for technologies such as the Industrial Internet of Things (IIoT) for supporting new services through linkage and convergence with other industries, Integrated Access and Backhaul (IAB) which provides nodes for expanding network service areas by integrating wireless backhaul links and access links, Mobility Enhancement including Conditional Handover and Dual Active Protocol Stack (DAPS) Handover, and 2-step Random Access (2-step RACH for NR) which simplifies random access procedures. Standardization is also underway in the field of system architecture / services for 5G baseline architectures (e.g., Service based Architecture, Service based Interface) for incorporating Network Functions Virtualization (NFV) and Software-Defined Networking (SDN) technologies, and Mobile Edge Computing (MEC), which provides services based on the location of the terminal.

[0006] When such 5G mobile communication systems are commercialized, connected devices, which are increasing explosively, will be connected to communication networks. Accordingly, it is expected that there will be a need to enhance the functionality and performance of 5G mobile communication systems and to integrate the operation of connected devices. To this end, new research is planned to be conducted on 5G performance improvement and complexity reduction, support for AI services, support for metaverse services, and drone communication using eXtended Reality (XR), Artificial Intelligence (AI), and Machine Learning (ML) to efficiently support Augmented Reality (AR), Virtual Reality (VR), and Mixed Reality (MR).

[0007] Furthermore, the advancement of these 5G mobile communication systems encompasses multi-antenna transmission technologies such as new waveforms to guarantee coverage in the terahertz band of 6G mobile communication technology, Full Dimensional MIMO (FD-MIMO), array antennas, and large-scale antennas; metamaterial-based lenses and antennas to improve terahertz band signal coverage; high-dimensional spatial multiplexing technology using OAM (Orbital Angular Momentum); and Reconfigurable Intelligent Surface (RIS) technology; as well as Full Duplex technology for enhancing frequency efficiency and system networks in 6G mobile communication technology; AI-based communication technologies that realize system optimization by utilizing satellites and AI from the design stage and internalizing end-to-end AI support functions; and the realization of services of complexity exceeding the limits of terminal computing capabilities by utilizing ultra-high-performance communication and computing resources. It could serve as a foundation for the development of next-generation distributed computing technologies.

[0008] Meanwhile, with the advancement of communication systems, there is a growing demand for various methods and devices to predict measurement results in order to enhance the cell measurement process.

[0009] Based on the discussion described above, the present disclosure provides an apparatus and method capable of effectively providing services in a next-generation wireless communication system. Furthermore, the present disclosure proposes various measures to improve the prediction accuracy of cell measurement results.

[0010] A method performed by user equipment (UE) according to one embodiment of the present disclosure comprises: receiving radio resource management (RRM) configuration information including cell geographical information from a base station; performing an RRM prediction by using the cell geographical information and cell measurement information as inputs to an Artificial Intelligence / Machine Learning (AI / ML) model; detecting the triggering of a measurement report event based on an RRM measurement result and an RRM prediction result; and transmitting a measurement report message including the RRM measurement result and the RRM prediction result to the base station.

[0011] User equipment (UE) according to one embodiment of the present disclosure comprises: at least one transceiver; at least one processor communicatively coupled to the at least one transceiver; and at least one memory communicatively coupled to the at least one processor for storing instructions, wherein the instructions are executed individually or in any combination by the at least one processor, so that the UE: receives radio resource management (RRM) configuration information including cell geographical information from a base station, performs an RRM prediction by using the cell geographical information and cell measurement information as inputs to an Artificial Intelligence / Machine Learning (AI / ML) model, detects the triggering of a measurement report event based on the RRM measurement result and the RRM prediction result, and transmits a measurement report message including the RRM measurement result and the RRM prediction result to the base station.

[0012] A method performed by a base station according to one embodiment of the present disclosure comprises: transmitting radio resource management (RRM) configuration information including cell geographical information to user equipment (UE); and receiving a measurement report message including the RRM measurement result and the RRM prediction result from the UE in accordance with a measurement report event based on an RRM prediction result and an RRM measurement result of an AI / ML (Artificial Intelligence / Machine Learning) model based on the cell geographical information and cell measurement information.

[0013] A base station according to one embodiment of the present disclosure comprises: at least one transceiver; at least one processor communicatively coupled to the at least one transceiver; and at least one memory communicatively coupled to the at least one processor for storing instructions, wherein the instructions are executed individually or in any combination by the at least one processor, so that the base station transmits radio resource management (RRM) configuration information including cell geographical information to user equipment (UE), and receives a measurement report message including the RRM measurement result and the RRM prediction result from the UE in accordance with an RRM prediction result of an AI / ML (Artificial Intelligence / Machine Learning) model based on the cell geographical information and cell measurement information and a measurement report event based on the RRM measurement result.

[0014] According to the various embodiments proposed in this disclosure, services can be effectively provided in a next-generation wireless communication system. In addition, according to other embodiments proposed in this disclosure, the accuracy and efficiency of measurement result prediction can be improved by predicting cell measurement results based on cell geographical information.

[0015] FIG. 1a is a drawing illustrating the structure of an NR system according to one embodiment of the present disclosure.

[0016] FIG. 1b is a diagram showing a wireless protocol structure in an LTE and NR system according to one embodiment of the present disclosure.

[0017] FIG. 1c is a diagram illustrating a use case utilizing an AI (artificial intelligence) / ML (machine learning) model for predicting cell measurement results in a next-generation mobile communication system according to one embodiment of the present disclosure.

[0018] FIG. 1d is a drawing illustrating a use case showing cell geographic information according to cell deployment and terminal location in a next-generation mobile communication system according to one embodiment of the present disclosure.

[0019] FIG. 1e is a flowchart illustrating message exchange with a base station for a terminal to perform a cell geographic information-based radio resource management (RRM) prediction according to one embodiment of the present disclosure.

[0020] FIG. 1f is a flowchart illustrating message exchange between a terminal and a base station in a situation where dynamic cell geographic information provided by a base station changes according to the current state of the terminal when the terminal performs cell geographic information-based RRM Prediction according to one embodiment of the present disclosure.

[0021] FIG. 1g is a flowchart illustrating an example of a terminal exchanging information on the preference of each AI / ML model between a base station and a terminal and determining an AI / ML model to perform cell geographic information-based RRM Prediction according to one embodiment of the present disclosure.

[0022] FIG. 1h is a diagram showing an example of an AI model used to perform RRM Prediction according to one embodiment of the present disclosure.

[0023] FIG. 1i is a flowchart illustrating a method according to one embodiment of the present disclosure in which a terminal performs RRM measurement and selectively performs at least one of a non-AI / ML model, a Basic AI / ML model, or an Advanced AI / ML model depending on the configuration status with the network and the availability of AI / ML input information.

[0024] FIG. 2 is a drawing illustrating a terminal device according to one embodiment of the present disclosure.

[0025] FIG. 3 is a drawing illustrating a base station device according to one embodiment of the present disclosure.

[0026] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. Furthermore, in describing the present disclosure, if it is determined that a detailed description of related known functions or configurations might unnecessarily obscure the essence of the present disclosure, such detailed description will be omitted. Additionally, the terms described below are defined considering their functions in the present disclosure, and these may vary depending on the intentions or conventions of the user or operator. Therefore, their definitions should be based on the content throughout this specification.

[0027] The advantages and features of the present disclosure and the methods for achieving them will become clear by referring to the embodiments described below in detail together with the accompanying drawings. However, the present disclosure is not limited to the embodiments disclosed below but may be implemented in various different forms. The embodiments provided are merely to ensure that the disclosure of the present disclosure is complete and to fully inform those skilled in the art of the scope of the disclosure, and the present disclosure is defined only by the scope of the claims. Throughout the specification, the same reference numerals refer to the same components.

[0028] At this point, it will be understood that each block of the process flow diagrams and combinations of the flow diagrams can be executed by computer program instructions. Since these computer program instructions can be loaded into the processor of a general-purpose computer, a special-purpose computer, or other programmable data processing equipment, the instructions executed through the processor of the computer or other programmable data processing equipment create means to perform the functions described in the flow diagram block(s). Since these computer program instructions can also be stored in computer-available or computer-readable memory that can be directed toward the computer or other programmable data processing equipment to implement the function in a specific way, the instructions stored in computer-available or computer-readable memory can also produce a manufactured item containing instruction means to perform the function described in the flow diagram block(s). Since computer program instructions can be loaded onto a computer or other programmable data processing equipment, instructions that perform a series of operation steps on the computer or other programmable data processing equipment to create a process executed by the computer can also provide steps for executing the functions described in the flowchart block(s).

[0029] Additionally, each block may represent a module, segment, or part of code containing one or more executable instructions for executing a specific logical function(s). It should also be noted that in some alternative execution examples, the functions mentioned in the blocks may occur out of order. For example, two blocks described in succession may actually be executed substantially simultaneously, or the blocks may sometimes be executed in reverse order according to their corresponding functions.

[0030] In this embodiment, the term "part" refers to a software or hardware component such as an FPGA (Field Programmable Gate Array) or an ASIC (Application Specific Integrated Circuit), and the "part" performs certain roles. However, the meaning of "part" is not limited to software or hardware. The "part" may be configured to reside in an addressable storage medium or configured to run one or more processors. Thus, as an example, the "part" includes components such as software components, object-oriented software components, class components, and task components, as well as processes, functions, attributes, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuits, data, databases, data structures, tables, arrays, and variables. The functions provided within the components and "parts" may be combined into a smaller number of components and "parts" or further separated into additional components and "parts." In addition, the components and 'parts' may be implemented to utilize one or more CPUs within the device or secure multimedia card. Also, in the embodiments, 'parts' may include one or more processors.

[0031] In describing the present disclosure below, specific descriptions of related known functions or configurations will be omitted if it is determined that such detailed descriptions would unnecessarily obscure the essence of the present disclosure. Embodiments of the present disclosure will be described below with reference to the attached drawings.

[0032] Terms used in the following description to identify connection nodes, terms referring to network entities, terms referring to messages, terms referring to interfaces between network entities, terms referring to various identification information, etc., are examples provided for the convenience of explanation. Accordingly, the present disclosure is not limited to the terms described below, and other terms referring to objects having equivalent technical meanings may be used.

[0033] In the following description, the terms "physical channel" and "signal" may be used interchangeably with "data" or "control signal." For example, PDSCH (physical downlink shared channel) is a term referring to a physical channel through which data is transmitted, but PDSCH may also be used to refer to data. That is, in this disclosure, the expression "transmits a physical channel" may be interpreted as equivalent to the expression "transmits data or a signal through a physical channel."

[0034] In the present disclosure, upper signaling refers to a signal transmission method transmitted from a base station to a terminal using a physical layer downlink data channel, or from a terminal to a base station using a physical layer uplink data channel. Upper signaling may be understood as radio resource control (RRC) signaling or a media access control (MAC) control element (CE).

[0035] For convenience of explanation, the present disclosure uses terms and names defined in the 3GPP NR (3rd Generation Partnership Project NR (New Radio)) or 3GPP LTE (3rd Generation Partnership Project Long Term Evolution) standards. However, the present disclosure is not limited by the above terms and names and may be applied equally to systems conforming to other standards. In the present disclosure, gNB may be used interchangeably with eNB for convenience of explanation. That is, a base station described as an eNB may represent a gNB. Additionally, the term terminal may refer to mobile phones, MTC devices, NB-IoT devices, sensors, as well as other wireless communication devices.

[0036] Hereinafter, a base station (BS) is an entity that performs resource allocation for terminals and may be at least one of a gNodeB (gNB), eNodeB (eNB), NodeB, a wireless access unit, a transmission and reception point (TRP), a base station controller, or a node on a network. A terminal may include a User Equipment (UE), a Mobile Station (MS), a cellular phone, a smartphone, a computer, or a multimedia system capable of performing communication functions. Of course, it is not limited to the above examples.

[0037] Additionally, in this disclosure, gNB may be used interchangeably with eNB for convenience of explanation. That is, a base station described as an eNB may represent a gNB. Furthermore, the term terminal may refer to mobile phones, MTC devices, NB (narrow band)-IoT (internet of things) devices, sensors, as well as other wireless communication devices.

[0038] Wireless communication systems are evolving from providing early voice-oriented services to broadband wireless communication systems that provide high-speed, high-quality packet data services, such as communication standards like 3GPP’s HSPA (High Speed ​​Packet Access), LTE (Long Term Evolution or E-UTRA (Evolved Universal Terrestrial Radio Access)), LTE-Advanced (LTE-A), LTE-Pro, 3GPP2’s HRPD (High Rate Packet Data), UMB (Ultra Mobile Broadband), and IEEE’s 802.16e.

[0039] As a representative example of a broadband wireless communication system, the LTE system employs the Orthogonal Frequency Division Multiplexing (OFDM) method for the downlink (DL) and the Single Carrier Frequency Division Multiple Access (SC-FDMA) method for the uplink (UL). The uplink refers to a wireless link through which a terminal (or UE) transmits data or control signals to a base station (or eNB, gNB), and the downlink refers to a wireless link through which a base station transmits data or control signals to a terminal. The above multiple access method distinguishes the data or control information of each user by allocating and operating time-frequency resources to be transmitted for each user so that they do not overlap, that is, so that orthogonality is established.

[0040] As a future communication system following LTE, 5G communication systems must be able to freely reflect the diverse requirements of users and service providers, and therefore, services that satisfy various requirements simultaneously must be supported. Services being considered for 5G communication systems include enhanced mobile broadband communication (eMBB), massive machine-based communication (mMTC), and ultra-reliable low-latency communication (URLLC).

[0041] According to one embodiment, eMBB may aim to provide a data transmission speed that is higher than the data transmission speed supported by existing LTE, LTE-A, or LTE-Pro. For example, in a 5G communication system, eMBB must be able to provide a peak data rate of 20 Gbps in the downlink and a peak data rate of 10 Gbps in the uplink from the perspective of a single base station. In addition, the 5G communication system may need to provide a user-perceived data rate while simultaneously providing the peak data rate. To satisfy these requirements, the 5G communication system may require improvements in various transmission and reception technologies, including enhanced Multiple Input Multiple Output (MIMO) transmission technology. Furthermore, while current LTE transmits signals using a maximum transmission bandwidth of 20 MHz in the 2 GHz band, the 5G communication system can satisfy the data transmission speed required by the 5G communication system by using a frequency bandwidth wider than 20 MHz in frequency bands of 3 to 6 GHz or above 6 GHz.

[0042] Simultaneously, mMTC is being considered to support application services such as the Internet of Things (IoT) in 5G communication systems. To efficiently provide IoT services, mMTC may require support for a large number of terminal connections within a cell, improved terminal coverage, enhanced battery life, and reduced terminal costs. Since IoT devices are attached to various sensors and equipment to provide communication functions, a cell must be capable of supporting a large number of terminals (e.g., 1,000,000 terminals / km²). Furthermore, due to the nature of the service, terminals supporting mMTC are likely to be located in dead zones not covered by cells, such as building basements; therefore, wider coverage may be required compared to other services provided by 5G communication systems. Terminals supporting mMTC must consist of low-cost devices, and since it is difficult to frequently replace terminal batteries, a very long battery life of 10 to 15 years may be required.

[0043] Finally, URLLC is a mission-critical cellular-based wireless communication service that can be used for services such as remote control of robots or machinery, industrial automation, unmanned aerial vehicles, remote health care, and emergency alerts. Therefore, the communication provided by URLLC may need to offer very low latency and very high reliability. For example, services supporting URLLC must satisfy an air interface latency of less than 0.5 milliseconds and may simultaneously require a packet error rate of 10^-5 or less. Consequently, for services supporting URLLC, 5G systems must provide a Transmission Time Interval (TTI) smaller than other services, and design considerations may be required to allocate wide resources in the frequency band to ensure the reliability of the communication link.

[0044] The three services considered in the aforementioned 5G communication system, namely eMBB, URLLC, and mMTC, can be multiplexed and transmitted within a single system. In this case, different transmission and reception techniques and parameters may be used between the services to satisfy the different requirements of each service. However, the aforementioned mMTC, URLLC, and eMBB are merely examples of different service types, and the service types to which the present disclosure applies are not limited to the examples mentioned above.

[0045] In addition, although embodiments of the present disclosure are described below with reference to LTE, LTE-A, LTE Pro, 5G (or NR), or 6G systems, the embodiments of the present disclosure may be applied to other communication systems having similar technical backgrounds or channel types. Furthermore, the embodiments of the present disclosure may be applied to other communication systems with some modifications made at the discretion of a person skilled in the art, without significantly departing from the scope of the present disclosure.

[0046] FIG. 1a is a drawing illustrating the structure of an NR system according to one embodiment of the present disclosure.

[0047] Referring to FIG. 1a, a wireless communication system may be composed of multiple base stations (e.g., gNB (105), ng-eNB (110), ng-eNB (115), gNB (120)), an Access and Mobility Management Function (AMF) (125), and a User Plane Function (UPF) (130). A user terminal (User Equipment, hereinafter UE or terminal) (135) can connect to an external network through the base stations (e.g., gNB (105), ng-eNB (110), ng-eNB (115), gNB (120)) and the UPF (130).

[0048] In FIG. 1a, base stations (e.g., gNB (105), ng-eNB (110), ng-eNB (115), gNB (120)) can provide wireless access to terminals connected to the network as access nodes of a cellular network. That is, the base stations (e.g., gNB (105), ng-eNB (110), ng-eNB (115), gNB (120)) can support a connection between the terminals and the core network (CN, Core network; specifically, the CN of NR is referred to as 5GC) by collecting state information such as the buffer state, available transmission power state, and channel state of the terminals to service the traffic of the users and performing scheduling. Meanwhile, in communication, the User Plane (UP), which is related to the transmission of actual user data, and the Control Plane (CP), which is related to connection management, can be configured separately. In this drawing, gNB (105) and gNB (120) use the UP and CP technologies defined in NR technology, and ng-eNB (110) and ng-eNB (115), although connected to 5GC, can use the UP and CP technologies defined in LTE technology.

[0049] The above AMF (125) is a device responsible for various control functions as well as mobility management functions for the terminal and is connected to multiple base stations, and the UPF (130) may refer to a type of gateway device that provides data transmission. Although not shown in FIG. 1a, the NR wireless communication system may include a Session Management Function (SMF). The SMF can manage packet data network connections, such as protocol data unit (PDU) sessions provided to the terminal.

[0050] FIG. 1b is a diagram showing a wireless protocol structure in an LTE and NR system according to one embodiment of the present disclosure.

[0051] Referring to FIG. 1b, the wireless protocol of the LTE system may consist of a Packet Data Convergence Protocol (PDCP) (105)(140), Radio Link Control (RLC) (110)(135), and Medium Access Control (MAC) (115)(130) at the terminal and eNB, respectively. The Packet Data Convergence Protocol (PDCP) (105)(140) is responsible for operations such as IP header compression / recovery, and the Radio Link Control (hereinafter referred to as RLC) (110)(135) reconstructs the PDCP Protocol Data Unit (PDU), i.e., the RLC Service Data Unit (SDU), into an appropriate size. The MAC (115)(130) is connected to multiple RLC layer devices configured in a terminal and performs the operation of multiplexing RLC PDUs into MAC PDUs and demultiplexing RLC PDUs from MAC PDUs. The physical (PHY) layer (120) (125) performs the operation of channel coding and modulating upper layer data, creating OFDM (orthogonal frequency division multiplexing) symbols to transmit to the wireless channel, or demodulating OFDM symbols received through the wireless channel and channel decoding to transmit to the upper layer. In addition, the physical layer also uses HARQ (Hybrid Automatic Repeat reQuest) for additional error correction, and the receiving end transmits 1 bit indicating whether the packet transmitted by the transmitting end has been received. This is called HARQ ACK (acknowledgement) / NACK (negative ACK) information.In the case of LTE, downlink HARQ ACK / NACK information for uplink data transmission is transmitted through the physical channel of PHICH (Physical Hybrid-ARQ Indicator Channel), while in the case of NR, it is possible to determine whether retransmission is required or if a new transmission can be performed through the scheduling information of the terminal in the PDCCH (Physical Downlink Control Channel), which is the channel where downlink / uplink resource allocation is transmitted. This is because asynchronous HARQ is applied in NR. Uplink HARQ ACK / NACK information for downlink data transmission can be transmitted through the physical channels of PUCCH (Physical Uplink Control Channel) or PUSCH (Physical Uplink Shared Channel). The above PUCCH is generally transmitted in the uplink of the Pcell (Primary Cell) described later, but if supported by the terminal, it may be additionally transmitted through the SCell (Secondary Cell) described later, and such a SCell is referred to as PUCCH SCell.

[0052] Although not shown in this drawing, an RRC (Radio Resource Control) layer exists above the PDCP layer of the terminal and the base station, respectively, and the RRC layer can exchange connection and measurement-related setting control messages for wireless resource control.

[0053] Meanwhile, the above PHY layer can be composed of one or more frequencies / carriers, and the technology of setting and using multiple frequencies simultaneously is called carrier aggregation (hereinafter referred to as CA). CA technology allows for a significant increase in transmission capacity by using one or more secondary carriers in addition to the primary carrier, thereby increasing the transmission capacity by the number of secondary carriers, whereas previously only one carrier was used for communication between a terminal (or User Equipment, UE) and a base station (E-UTRAN NodeB, eNB). Meanwhile, in LTE, a cell within a base station that uses the primary carrier is called a primary cell or PCell (Primary Cell), and a cell within a base station that uses a secondary carrier is called a secondary cell or SCell (Secondary Cell).

[0054] FIG. 1c is a diagram illustrating a use case in which an AI / ML model is utilized to predict cell measurement results in a next-generation mobile communication system according to an embodiment of the present disclosure. Referring to FIG. 1c, an AI / ML model (100) can be utilized to predict cell measurement results. For reference, the cell measurement results may refer to RSRP (reference signal received power), RSRQ (reference signal received quality), and SINR (signal to interference plus noise ratio) values ​​measured by a terminal for each cell. Additionally, for each cell, if there are multiple beams transmitted by the cell, the cell measurement results may include RSRP / RSRQ / SINR values ​​measured by the terminal for each beam. Furthermore, each of the RSRP / RSRQ / SINR values ​​may refer to one of the following values.

[0055] - RSRP and / or RSRQ and / or SINR measured at Layer 1

[0056] - RSRP and / or RSRQ and / or SINR measured / acquired at Layer 3

[0057] - Values ​​obtained by filtering RSRP and / or RSRQ and / or SINR measured at Layer 1 (e.g., (weighted) average values ​​using measurements over a specified period)

[0058] - Values ​​obtained by filtering RSRP and / or RSRQ and / or SINR measured / acquired at Layer 3 (e.g., (weighted) average values ​​using measurements over a specified period)

[0059] Here, "cell" refers to all or measurable cells existing at the same frequency as the cell currently connected to the terminal, or all or measurable cells operating at a different frequency from the cell connected to the terminal.

[0060] Meanwhile, by utilizing an AI / ML model, it is possible to predict unmeasured information based on measured information. For example, the AI / ML model can be designed so that information measured during a specific past period is used as the input (110) to the AI / ML model, and cell measurement information to be measured in the future becomes the output (120) of the AI / ML model. At this time, how much cell measurement information from the past is accumulated to predict cell measurement information for a future period may be flexible depending on the design of the AI / ML model and the implementation of the terminal, and it can be assumed that the terminal and the base station have agreed on the options for this in advance.

[0061] Additionally, terminals and base stations (or networks) may use AI / ML models for frequency domain prediction. For example, to measure cells existing on a frequency different from the cell currently connected to the terminal, the terminal may need to stop transmitting and receiving at the currently connected cell and proceed with measurements at the other frequency. If AI / ML models are utilized for frequency domain prediction, it may be possible to use information measured during a specific past time as input to the AI / ML model to predict cell measurement values ​​for cells existing on different frequencies that were not actually measured during the same specific past time, or to predict cell measurement values ​​for cells existing on different frequencies for a certain period in the future.

[0062] The terminal and the base station (or the network) can predict multiple cell measurement results by utilizing multiple cell measurement results through the aforementioned AI / ML model. Of course, the terminal and the base station (or the network) can also predict a single cell measurement result by utilizing a single cell measurement result. In this case, including measurement results for a specific cell means not only results measured or predicted for the same cell, but also results measured or predicted for other cells including the cell in question.

[0063] Meanwhile, in the aforementioned AI / ML model usage case, when only cell measurement results such as RSRP / RSRQ / SINR are used, the prediction accuracy of the AI / ML model may be relatively low. A typical example is when there are obstacles that can significantly affect signal sensitivity, such as a building (150). In such cases, rather than using only cell measurement results as input to the AI / ML model, it may be helpful to increase the prediction accuracy of the AI / ML model by including information such as the location information of the terminal (140) and base station (130), or the angle offset (131) through the cell's antenna direction vector and the current location vector of the cell and the terminal. For example, if only information of cell measurement signals such as RSRP is used as input to the AI / ML, it may be determined that the terminal (141) has moved away from the antenna location of the cell when the RSRP value of a specific area within the cell decreases; however, in reality, if the terminal moves and moves behind a building (150), the RSRP value may decrease significantly due to the presence of a building between the base station antenna and the terminal. In such cases, when the direction of the base station (135) antenna, the location of the base station antenna, etc., are used as inputs in the AI / ML model, situations such as when the RSRP is low even though the distance is short can be identified more accurately. Through these methods, the accuracy of the measurement results and / or measurement prediction results using the AI / ML model can be improved.

[0064] When the AI / ML model (100) described above is used for time domain prediction, the predicted future cell measurement results can be used for the following exemplary purposes.

[0065] The first objective is to improve the handover performance of the terminal. Based on predicted future cell measurement results, the base station can predict the optimal target cell for a rapidly moving terminal and hand over the terminal to the optimal cell at the appropriate time. More specifically, the base station can receive cell measurement results by periodically receiving measurement reports from the terminal and hand over the terminal to the optimal cell based on these results. However, because the actual channel environment changes according to the cell measurement result reporting cycle (measurement report transmission cycle), a delay may occur between the time the optimal cell changes and the time the base station actually receives the measurement report from the terminal and identifies this change. Furthermore, when the terminal is moving rapidly, this delay in the change of cell measurement results can cause the terminal to fail to hand over. For example, due to the delay between the base station receiving the measurement report from the terminal, deciding to hand over the terminal, requesting a handover from the adjacent base station corresponding to the target cell, and receiving approval, the terminal may not be handed over in a timely manner, and the terminal may fall into a Radio Link Failure (RLF) state. To improve these problems, the base station can prevent handover failures by predicting the optimal cell change of the terminal in advance based on predicted cell measurement results in the time domain and handing over the terminal at an appropriate time.

[0066] The second objective is to reduce the measurement overhead of the terminal. The terminal can reduce the cell measurement overhead while maintaining handover performance by performing cell measurements, which were previously performed every SSB (synchronization signal and physical broadcast channel (PBCH) block) cycle, only for some cycles, and by replacing the cell measurement values ​​at omitted or skipped points with predicted cell measurement results. More specifically, the terminal can measure cell signal strength every SSB cycle and perform RRM-related operations based on those measurements. The RRM operations may include detecting and reporting RRM measurement events, and detecting RLF (Radio Link Failure). For the RRM operations, instead of using the measured result value every T_per every SSB cycle, the terminal may skip the SSB measurement once in a while and use the measured result value every 2*T_per. As described above, if RRM operations are performed using only the result values ​​measured every 2*T_per, the cell measurement load of the terminal is reduced by 50%, but the accuracy of the RRM operation may decrease. For example, measurement events and RLF may not be detected in time. Furthermore, this can eventually lead to a degradation of the terminal's handover performance. Therefore, to reduce the cell measurement load while preventing handover performance degradation, the terminal can predict the cell measurement result at the point where the measurement was skipped and use that result in the RRM operation. By doing so, the terminal can reduce the cell measurement load without degrading the terminal's handover performance.

[0067] When the AI / ML model (100) described above is used for frequency domain prediction, the prediction results can be used to prevent a drop in the terminal's transmission rate. If frequency domain prediction is not possible, the base station may set a measurement gap for the terminal, instructing it to stop the transmission and reception process in the currently connected cell during the measurement gap and to measure cells at other frequencies. When such a measurement gap is activated, the measurement gap period and measurement gap duration are set, and since transmission and reception are impossible for a duration at each period, a drop in the terminal's transmission rate of up to (duration / period)*100% may occur. To solve this problem, the base station may utilize the terminal's AI / ML model to instruct or set the terminal to replace the cell measurement results for other frequencies with the cell measurement prediction results of the AI / ML model without setting a measurement gap. Alternatively, if the base station utilizes the AI / ML model, the base station may instruct the terminal to report the measurement results of the currently connected cell, and based on that information, replace the cell measurement results for other frequencies with the cell measurement prediction results of the AI / ML model. In the above example, a decrease in the data transmission rate of the terminal caused by the measurement gap setting and operation can be prevented.

[0068] FIG. 1d is a drawing illustrating a use case showing cell geographic information according to cell placement and terminal location in a next-generation mobile communication system according to one embodiment of the present disclosure.

[0069] FIG. 1d describes an example of a method for selecting B cells when B cell measurement information, which is smaller than A, is to be used as input to an AI / ML model in a situation where cell measurement information for A cells is given to a terminal and / or base station in the embodiment mentioned in FIG. 1c. As in the example, it can be assumed that the current terminal (100) is connected to cell 6 (120) operated by the base station (110). At this time, for instance, it can be assumed that the terminal has collected cell measurement information for cells 2, 4, 5, 6, 7, and 8, and is predicting future measurement information for the actual cell 6. At this time, by means of the methods described below, it can be assumed that the terminal has decided to use only the information of C cells (C <= 6) as input to the base station and the AI / ML model. In such a situation, when selecting C cells with the best AI / ML performance, the terminal may consider the following methods.

[0070] * Method 1: The terminal may assign priorities using its own method based on the signal characteristics of each cell measured over a predetermined period of time. For example, the terminal may evaluate the average RSRP of each cell signal and select C cells with the highest values ​​among them.

[0071] * Method 2: The base station may prioritize providing information about other cells operated by the base station of the cell currently connected to the terminal (hereinafter referred to as co-located cells) and assign a high priority to those cells. For example, when C = 3, the base station may inform the terminal that cells 4 and 5 are operated by the base station located in the same position as the current cell 6. Based on this information, the terminal can utilize the information of co-located cells 4, 5, and 6 to operate AI / ML when using only three cell information. For signaling to support Method 2, the base station may utilize broadcast messages such as System Information Blocks (SIBs), and may use information that denotes co-located cells as a set or group, such as {1,2,3}, {4,5,6}, {7,8,9}, etc., or may use information indicating which base station operates each cell, such as {cells 1,2,3 - Base Station 1}, {cells 4,5,6 - Base Station 2}. In addition, all possible expressions that can inform the base station and the terminal that multiple cells are co-located may be considered. If the base station is in a situation where it can change co-located cell information through dynamic cell on-off operations, such information may be provided not only via broadcast through the SIB messages described above, but may also be provided to the terminal using unicast messages such as RRCReconfiguration. Meanwhile, in the above example, cell 6 is a co-located cell with cells 4 and 5, but the co-located information for cell 6 may also include information about surrounding co-located cells.This is because, for example, even though the terminal (100) is connected to cell 6, it may be necessary to predict cell measurement information for cell 2 in order to determine whether the handover to cell 2 was successful. In this case, cell measurement information previously measured for cells 1, 2, and 3 can be used as input to the AI / ML model for predicting measurement information for cell 2.

[0072] * Method 3: The base station can provide the terminal with information about adjacent cells based on the cell the terminal is currently connected to, thereby assigning a high priority to the adjacent cells. For example, the base station can inform the terminal that in a cell arrangement like that shown in FIG. 1d, the cells adjacent to cell 6 are cells 2, 4, 5, 7, and 8, and that the cells adjacent to cell 8 are cells 6, 7, and 9. As described in Method 2 above, this information is shared from cell 6, but a list of adjacent cells surrounding that cell can be provided independently for each of the surrounding adjacent cells. Furthermore, it goes without saying that broadcast methods via SIB messages and unicast methods via RRCReconfiguration messages can all be considered.

[0073] * Method 4: Similar to Method 3, the base station provides information about adjacent cells, but may assign different priorities to each adjacent cell. For example, in a cell layout like that shown in FIG. 1d, the cells adjacent to cell 6 are cells 2, 4, 5, 7, and 8; however, the base station may assign the highest priority to cells 4 and 5, a medium priority to cell 2, and the lowest priority to cells 7 and 8. Such a method can be utilized when large obstacles, such as buildings, exist within a cell, or when the signal strength coming from a specific cell cannot be properly measured, thereby reducing the importance of information regarding that cell. In this case, the information can be provided via a broadcast-based message utilizing SIB. Alternatively, there may be cases where the priority of the information in Method 4 varies depending on the terminal's current location. In this case, the base station may instruct the terminal to provide the relevant information through an RRCReconfiguration message, etc. To generate the message, a preliminary operation may be required in which the base station requests the terminal's location information. Alternatively, information may be provided so that a specific area is divided into multiple zones based on location or measured signal strength, and different priorities are applied to each adjacent cell within each zone.

[0074] * Method 5: The terminal and / or base station may directly provide information, such as the base station's location and the antenna direction of each cell, as input to the AI / ML model. Methods 1 through 4 mentioned above were primarily applicable when location information of the terminal or base station, or base station antenna information, was not utilized as input to the AI / ML model. If the base station's location and the antenna direction of each cell are directly used as input to the AI / ML model, such information can be shared with the terminal by the base station. In a typical network situation, information such as the base station's location or the antenna direction of each cell may not change frequently. In this case, such information can be provided via broadcast-based messages utilizing SIB. However, considering environments where the base station may move, or cases where the base station may variably operate the antenna direction of each cell to optimize coverage, such information may be dynamic. In this case, the base station can instruct the terminal to provide such information through messages such as RRCReconfiguration. Furthermore, it may be difficult for the base station to acquire information regarding the location of adjacent base stations or the antenna direction of each cell in real time. In this case, methods may be considered in which a base station identifies information regarding the location or antenna direction of an adjacent base station through message exchange between base stations or with a core network, and shares this information with the terminal.

[0075] For convenience of explanation, exemplary information that can be provided through the methods 1 to 5 described above may be referred to as 'cell geographic information'. Even for methods not described above, it is possible for a terminal and a base station to exchange information through appropriate signaling and use it as input for an AI / ML model, for any additional methods that combine methods 1 to 5 or derive logic similar to methods 1 to 5.

[0076] FIG. 1e is a flowchart illustrating message exchange with a base station for a terminal to perform cell geographic information-based RRM Prediction according to one embodiment of the present disclosure.

[0077] Referring to FIG. 1e, it can be assumed that the terminal (100) has previously obtained static cell geographic information from the base station (110) via a broadcast-based message (120). Additionally, the terminal (100) can determine whether the cell geographic information can be used as an AI / ML model through signaling with the base station (110), and if the cell geographic information is used as an input to the AI / ML model, it can determine whether the corresponding AI / ML has been trained in the current cell. Furthermore, the terminal (100) can report whether the AI / ML has been trained to the base station (110) in a subsequent Applicability Report stage, and for this purpose, the terminal's RRM prediction-related capability can be reported from the terminal (100) to the base station (110) during the process of exchanging UE capability information between the terminal and the base station. This RRM prediction-related capability may include various information related to AI / ML-based prediction, such as whether RRM prediction is possible and the type of input to the AI / ML model. Based on the information, the base station may provide the terminal with RRM prediction settings when configuring RRM measurement settings, and may provide RRC prediction settings by including the terminal's AI / ML model that the terminal needs to use and dynamic cell geographic information. When the terminal uses the base station's dynamic cell geographic information as input to the AI / ML model, it can determine at this stage whether the AI / ML model has been trained.If it is determined that the terminal's AI / ML model has not been trained, the terminal may report this to the base station via an applicability report, and the base station may transmit data for the terminal to train the AI / ML model, or transmit RRCReconfiguration information including scheduling information for said data. Alternatively, the base station may transmit information about a trained AI / ML model that it already possesses.

[0078] Alternatively, the base station may disable the AI / ML-based RRM prediction. If the AI / ML model has completed training, the base station may command the terminal to perform an RRM prediction operation based on the AI / ML model, and based on this, the terminal may be able to perform RRM measurement and RRM prediction. Additionally, the base station may configure the terminal to trigger a measurement report based on RRC measurement and RRC prediction. When a measurement report is triggered, the base station may check the measurement report message received from the terminal and make a decision regarding the handover operation of the terminal.

[0079] A detailed description of each step for the above-described operation is as follows.

[0080] In step 120, the base station may broadcast the base station's non-variable or static cell geographic information through a message such as a SIB. Even if the cell geographic information contained within the SIB changes, the SIB is transmitted repeatedly and periodically, so the terminal can recognize that the cell geographic information has changed through the received SIB. If necessary, the terminal may transmit to the base station a field including a check to verify whether it has received and recognized the latest SIB information in the UE Capability exchange (step 130) or Applicability report (step 150). The information representing the cell geographic information within the SIB transmitted by the base station may include one or more combinations of the following information, and such information may be used for the methods described in FIG. 1d.

[0081] - Indicator indicating whether the message can be used as input to an AI / ML model: This indicator may indicate whether the terminal receives the message and modifies the input to the AI / ML model based on the information, or whether the message content itself or some fields can be used as input to the AI / ML model. This indicator may be set to values ​​such as ENUMERATED / BOOLEAN{True}, ENUMERATED{supported}, or ENUMERATED / BOOLEAN{1} to mean that it can be used as input to the AI / ML model, or set to values ​​such as ENUMERATED / BOOLEAN{False}, ENUMERATED{not supported}, or ENUMERATED / BOOLEAN{0} to mean that it cannot be used as input to the AI / ML model.

[0082] - Indicator indicating whether each message field can be used as input for an AI / ML model: When a terminal receives a message, various fields may exist within the message, and the base station may determine in advance that some of these fields are not suitable for use by the terminal as an AI / ML model. In such cases, whether each field included in the message can be used as input for an AI / ML model may be expressed, and an indicator to indicate such usage as input for an AI / ML model may be included for each field. This indicator may be set to a value such as ENUMERATED / BOOLEAN{True}, ENUMERATED{supported}, or ENUMERATED / BOOLEAN{1} to indicate that it can be used as input for an AI / ML model, or set to a value such as ENUMERATED / BOOLEAN{False}, ENUMERATED{not supported}, or ENUMERATED / BOOLEAN{0} to indicate that it cannot be used as input for an AI / ML model.

[0083] - Information on co-located cells: As described in Method 2 of FIG. 1d, the base station may provide the terminal with information on a co-located cell set including the currently serving cell, or information on a co-located cell set operating at an adjacent base station. Here, a co-located cell set may refer to cells operating at the same frequency that steer their antennas in different directions (or have different beam radiation directions), or cells installed in the same location that operate at different frequencies. It is also possible for the base station to inform the terminal of the co-located cell set operating at the same frequency and the co-located cell set operating at a different frequency as separate information. Alternatively, it is possible for the base station to inform the terminal of the co-located cell set operating at each location and at each frequency. In such cases, when a base station operates multiple frequencies at the same location, multiple co-located cell sets may be defined, and a situation may arise where multiple co-located cell sets operating at nearby base stations are defined simultaneously. For example, each co-located cell set may contain an index representing information on three cells, and it can be assumed that RSRP values ​​from six cells are to be input into an AI / ML model. In this case, the terminal can generally select a co-located cell set with the same location and frequency as the current serving cell as the highest priority input to the AI / ML model. Subsequently, when selecting the next three cells, the terminal may not be able to determine a criterion for determining priority between a co-located cell set with a different frequency at the same location and a co-located cell set with the same frequency at an adjacent location.To this end, the base station may transmit a message to the terminal containing priority information between each co-located cell set based on the current serving cell. If multiple co-located cell sets have the same priority value, the terminal may select a specific co-located cell set first, either arbitrarily or through separate criteria. Alternatively, if the terminal has previously obtained priority information regarding co-located cell sets for a specific area (for example, if the terminal independently determines the importance of each input by acquiring data on the relevant cell and neighboring cells and directly training an AI / ML model), the terminal may disregard the priority of the co-located cell set designated by the base station and determine the priority itself.

[0084] - Information on Adjacent Cells: The base station can provide the terminal with information regarding which cells are adjacent to the current serving cell. In this case, the base station may indicate the priority of each adjacent cell; a separate field value to indicate this priority may be included within the message, or a rule may be pre-defined stating that priority is higher in the order in which the adjacent cells are listed within the message (i.e., the order of entries within the message). The base station may be able to provide information on adjacent cells not only for the current serving cell but also for each surrounding cell. This is to accommodate situations where the terminal needs to predict the measurement information of adjacent cells in addition to that of the serving cell when predicting cell measurement information.

[0085] - Antenna location information of the relevant cell and adjacent cells: The base station can provide antenna location information for the currently serving cell and adjacent cells. This is for situations where the antenna location information of each cell is directly used as input for an AI / ML model, or where the information is used for the terminal to independently prioritize when selecting measurements to use as input for the AI / ML model.

[0086] - Antenna direction information of the relevant cell and adjacent cells: The base station can provide antenna direction information for the currently serving cell and adjacent cells. This is for situations where the antenna direction information of each cell is directly used as input for an AI / ML model, or where the terminal uses this information to determine priorities when selecting measurements to use as input for the AI / ML model.

[0087] Step 130 is a process in which the terminal and the base station exchange information regarding the presence of RRM prediction capability, the type of AI / ML input, and whether the corresponding model has been trained. The base station may request information related to the terminal's RRM prediction capability by sending a message such as UECapabilityEnquiry, and the terminal may report its current RRM prediction capability by sending a message such as UECapabilityInformation. At this time, the terminal may include one or more of the following information in UECapabilityInformation and transmit it to the base station.

[0088] - Indicator indicating whether RRM Prediction operation is possible: The terminal can inform the base station whether the terminal is currently capable of RRM prediction. The indicator may be set to a value such as ENUMERATED / BOOLEAN{True}, ENUMERATED{supported}, or ENUMERATED / BOOLEAN{1} to indicate that RRC prediction is possible, or set to a value such as ENUMERATED / BOOLEAN{False}, ENUMERATED{not supported}, or ENUMERATED / BOOLEAN{0} to indicate that RRC prediction is not possible.

[0089] - Types of AI / ML Input / Output: The terminal may provide information on the input types of the AI / ML models currently available to it. If the terminal can use multiple AI / ML models, it may provide input type information separately for each model. For example, the terminal may separately transmit information regarding the types of input / output in cases where it can predict m future RSRP values ​​of a cell by utilizing k recently acquired RSRP values ​​of a single cell through the simplest AI / ML model, and cases where it can predict m future RSRP values ​​of multiple cells (B) by utilizing k recently acquired RSRP values ​​of multiple cells (A) through a more complex AI / ML model and utilizing location information of each of the A cells.

[0090] - Training Status of AI / ML Models: If the terminal can use multiple AI / ML models, it may provide information regarding the training status for each AI / ML model. For example, consider a case where the terminal can predict m future RSRP values ​​for a single cell by utilizing k recently acquired RSRP values ​​for that cell using the simplest AI / ML model, and a case where the terminal can predict m future RSRP values ​​for multiple (B) cells by utilizing location information for each of A cells through k recently acquired RSRP values ​​for multiple (A) cells using a more complex AI / ML model. If the first AI / ML model is trained, but the second AI / ML model is not trained because the terminal has never attempted to train it on the current cell layout, the terminal may report the training status for each AI / ML model differently. This information may not be properly determined at this stage if the terminal's AI / ML utilizes dynamic cell geographic information. In such cases, the information may be omitted or transmitted as information indicating a status such as 'undetermined'. According to one embodiment, information regarding whether training is performed for each AI / ML model may be transmitted to a base station through a process in which the terminal reports UE assistance information to the base station, which is a process separate from the capability exchange process described above. Specific details are described later in FIG. 1g.

[0091] Step 140 describes the step in which a base station recognizes the terminal's RRM prediction-related capability and, when setting up RRM measurement-related settings, provides RRM prediction-related settings together. In this step, the base station may transmit at least one of the following information to the terminal.

[0092] - Indicator indicating whether RRM measurement can be replaced by RRM prediction: This indicator indicates whether the base station permits the terminal's RRM prediction. The indicator can be set to values ​​such as ENUMERATED / BOOLEAN{True}, ENUMERATED{supported}, or ENUMERATED / BOOLEAN{1} to signify that RRM prediction is allowed, or set to values ​​such as ENUMERATED / BOOLEAN{False}, ENUMERATED{not supported}, or ENUMERATED / BOOLEAN{0} to signify that RRM prediction is not allowed. Alternatively, if the indicator does not exist, it is possible to implicitly express its content by indicating whether or not RRM prediction-related settings exist within the base station's RRCReconfiguration field.

[0093] - Information on which RRM measurements can be replaced with RRM predictions when RRM measurements can be replaced with RRM predictions: For RRM predictions in the time domain and RRM predictions in the frequency domain described in Fig. 1c above, RRM prediction-related settings may be provided respectively. When performing RRM predictions in the time domain, the indicator may indicate whether to predict future information based on past information, or whether to perform RRM measurements intermittently in the time domain and utilize / apply the prediction results for measurements that were not performed. In addition, similarly for RRM in the frequency domain, the indicator may indicate whether the value can be replaced with a prediction without performing an actual RRM measurement for the frequency currently to be specified.

[0094] - Triggering conditions for measurement reports based on RRM measurement and RRM prediction results: The indicator may specify conditions for triggering a measurement report based on the aforementioned RRM prediction results for each domain. For example, in the case of RRM prediction in the time domain, the indicator may specify whether to report an event in advance if a specific event is predicted earlier than the actual measurement basis, or to specify that predictions should be disabled from then on if a specific condition is satisfied by performing fewer measurements, or to specify whether to apply a mode that activates all measurements to increase event accuracy if a specific condition is satisfied.

[0095] - Types of AI / ML models and inputs that the terminal must use when making RRM predictions: In step 130 of FIG. 1e, if the terminal reports that multiple AI / ML models are available, the base station may provide information to specify which AI / ML model the terminal should use for operation. In addition, as inputs to the AI / ML model, for example, in the case of information such as RSRP information accumulated along the time axis, the indicator may indicate information regarding the inputs and outputs of the AI / ML model, such as how many accumulated pieces of information to use as inputs to the model, or how many RSRP values ​​measured at what intervals should be inputs. If the size of the input to the terminal's AI / ML model reported by the terminal in step 130 is variable and determined to be within a predetermined range (for example, the terminal can use the RSRPs of a total of 6 cells as inputs), the base station may specify one value within that range to the terminal, or instruct the terminal to operate within a range smaller than the range suggested by the terminal.

[0096] - Dynamic cell geographic information: This may include any information having the same or similar effects as the methods described in methods 1 through 5 of FIG. 1d. Representative examples may include cases where the location of the current serving cell and adjacent cells changes, or where the antenna angle (or direction) of the current serving cell and adjacent cells changes. In such cases, the base station may provide the relevant information, and even if the information is transmitted via broadcast in step 120, it may be shared again through the message.

[0097] In steps 150 / 170, it may indicate whether the base station's RRCReconfiguration message configured in step 140 can be applied. In particular, if the setting cannot be applied immediately, it may mean that the terminal's AI / ML model has not completed training. For example, in a situation where the location or antenna angle of the serving cell and neighboring cells changes, the terminal can check the dynamic cell geographic information within the RRCReconfiguration message in step 140 to determine the current location of the serving cell and neighboring cells, and identify that the performance of the terminal's AI / ML model is unstable in that deployment pattern. In this case, as in step 150, the terminal can notify the base station that the AI / ML model has not been trained by transmitting an indicator such as isModelTrained = 0.

[0098] In step 160, if the terminal informs the base station that prediction is impossible for the AI / ML model configured via RRCReconfiguration, the base station may provide information for training the model. The base station may provide training data so that the terminal can train the AI / ML model directly. However, since it takes time for the terminal to train directly in this case, the base station may instruct the terminal to operate without RRM prediction for the time being by regenerating and transmitting the RRCReconfiguration message; since the operation in this case is the same as the existing terminal's operation, a detailed explanation is omitted. Alternatively, the base station may transmit information about the trained AI / ML model to the terminal; in this case, the terminal can store such information internally and run the AI / ML.

[0099] In step 170, the terminal can notify that the AI / ML model configured by the base station's RRCReconfiguration has completed training and can be run.

[0100] In step 180, the base station may instruct the terminal to perform the AI / ML model-based RRM prediction and RRM measurement operations set in step 140 by transmitting an RRCReconfiguration message. Some of the information described in step 140 may be transmitted to the terminal redundantly in step 180. If there has been a change to the information but the terminal cannot apply it, the terminal may provide the base station with information regarding the reason why the settings cannot be applied, etc., through an applicability report. If the terminal can apply all message contents in step 180, the terminal may report to the base station to perform the RRM prediction and RRM measurement operations by applying the settings by transmitting the RRCReconfigurationComplete message in step 190 to the base station.

[0101] In step 200, the terminal performs RRM measurement and prediction functions and can continuously monitor whether a specific event situation set by the base station is triggered. If a specific event situation is triggered in step 200, the terminal can activate the measurement report process as in step 210 and report a message to the base station containing related information, such as at least one of the RRM measurement result and the RRM prediction result.

[0102] In step 220, the base station can determine whether to perform handover operations, etc., based on the measurement report of the reported terminal.

[0103] Meanwhile, according to one embodiment, the order of the message exchange procedure described in FIG. 1e can be changed at any time depending on the situation. For example, in FIG. 1e, the terminal notifies the base station that the model it intends to use has not been trained through step 150 after step 140, but it is obvious that the terminal can notify the base station of whether the model available to the terminal is trained even before step 140. Other message exchange processes can also be performed with the order described above changed at any time.

[0104] FIG. 1f is a flowchart illustrating message exchange between a terminal (100) and a base station (110) in a situation where dynamic cell geographic information provided by a base station changes according to the current state of the terminal when the terminal performs cell geographic information-based RRM Prediction according to one embodiment of the present disclosure.

[0105] Steps 120 and 130 of FIG. 1f may be applied in the same or similar manner as the descriptions in steps 120 and 130 of FIG. 1e. However, if, in providing the AI / ML input types available to the terminal in step 130, the cell geographic information that the base station must provide may vary depending on the terminal's current state (e.g., the terminal's location), then content different from that described in FIG. 1e may be applied.

[0106] Step 140 describes a step in which the base station recognizes the terminal's RRM prediction-related capability and, when setting up RRM measurement-related settings, provides RRM prediction-related settings together. In this step, the base station may share at least one of the information described in Step 140 of FIG. 1e or the information described below.

[0107] - As described in step 140 of FIG. 1e, if the base station can replace an RRM measurement with an RRM prediction, it can provide information on which RRM measurement can be replaced with an RRM prediction.

[0108] - As described in step 140 of Fig. 1e, the base station can provide information on the measurement report triggering conditions based on the results of RRM measurement and RRM prediction.

[0109] - As described in step 140 of FIG. 1e, the base station can provide information about the AI / ML model and the type of input that the terminal should use when making an RRM prediction.

[0110] - Terminal state information that the terminal must provide to provide dynamic geographic information of the base station: For example, the base station may provide dynamic information based on the terminal's location. For instance, as in Method 4 described in Fig. 1d, when the base station performs RRM prediction for a specific cell for the terminal, when providing information on important cells, the current terminal's serving cell may be designated as the most important cell, and the adjacent cell closest to the current terminal's location may be designated as the next most important cell. Here, the term "closest adjacent cell" can be understood as having the shortest distance from the antenna location, or methods may be considered such as recognizing the center of the virtual space covered by the cell (a virtual cell space consisting of a circle or a hexagon) as the representative location of the cell, and defining the adjacent cell closest to the terminal's distance from that center as the "closest adjacent" cell.

[0111] - Triggering conditions for measurement reports based on changes in terminal status information: The base station may be configured to continuously monitor the terminal's status, and if that status changes beyond a specific condition, to execute a measurement report or trigger a similar type of event report. For example, conditions such as X amount of time having elapsed from the terminal's last reported location or Y amount of movement can be set as event triggering conditions. Alternatively, reporting the relevant information at periodic intervals may also be considered.

[0112] In step 150, the terminal can provide the 'information that the terminal must provide to provide the base station's dynamic geographic information requested by the base station in step 140' through a procedure such as an Applicability report. For example, if the requested location was the terminal's location, the terminal can transmit the terminal's location information to the base station through the Applicability report.

[0113] In step 160, the base station can provide the terminal with dynamic cell geographic information reflecting the current state information of the terminal as follows.

[0114] - Dynamic cell geographic information: This may include any information having the same or similar effects as the methods described in methods 1 to 5 of FIG. 1d. Representative examples may include cases where the location of the current serving cell and adjacent cells changes, or where the antenna angle of the current serving cell and adjacent cells changes. In such cases, the base station may provide the relevant information, and even if the information is transmitted via broadcast in step 120, it may be shared again through the message.

[0115] The terminal can receive information at step 160 from the base station, report that the AI / ML model corresponding to steps 150 and 160 of FIG. 1e has not been trained, and perform a procedure to receive information for training the AI / ML model from the base station. Since this process can be applied identically or similarly to the content previously described in FIG. 1e, a detailed explanation is omitted.

[0116] In step 170, the terminal can notify the base station that the AI / ML model configured by the base station's RRCReconfiguration has completed training and is ready to run, and can also report updated terminal status information (e.g., UE location) through the message.

[0117] In step 180, the base station may instruct the terminal to perform the AI / ML model-based RRM prediction and RRM measurement operations and terminal status monitoring (e.g., UE location monitoring) configured in step 140 via the RRCReconfiguration message. Some of the information described in step 140 may be transmitted to the terminal redundantly in step 180. If there has been a change to the information but the terminal cannot apply it, the terminal may provide the base station with information regarding the reason why the configuration cannot be applied, etc., via an applicability report. If the terminal can apply all message contents in step 180, the terminal may report to the base station to apply the configuration and perform the RRM prediction and RRM measurement operations and terminal status monitoring by transmitting the RRCReconfigurationComplete message in step 190 to the base station.

[0118] In step 200, the terminal performs RRM measurement / prediction / terminal status information monitoring operations and can continuously monitor whether a specific event situation set by the base station is triggered.

[0119] If an event situation for updating terminal status information is triggered in step 200, the terminal can activate the measurement report process as in step 210 and report a message to the base station containing at least one of the relevant information, such as the RRM measurement result and the RRM prediction result.

[0120] The base station can update dynamic cell geographic information in accordance with the status information of the new terminal through step 220, and the content of the message may be the same as step 160.

[0121] If a specific event situation is triggered by information obtained from the terminal's RRM measurement or prediction in step 230, the terminal may activate the measurement report process as in step 240 and report a message to the base station containing related information, such as at least one of the RRM measurement result and the RRM prediction result.

[0122] In step 250, the base station can determine whether to perform handover operations, etc., based on the measurement report of the reported terminal.

[0123] FIG. 1g is a flowchart illustrating an example of a terminal exchanging information on the preference of each AI / ML model between a base station and a terminal and determining an AI / ML model to perform cell geographic information-based RRM Prediction according to one embodiment of the present disclosure.

[0124] FIG. 1g is an example in which, when a terminal (100) reports to a base station (110) about the RRM prediction capability and the types of AI / ML models available for use, if the terminal reports that multiple types of AI / ML models are available for use, the base station determines which type of AI / ML model to select by requesting additional information from the terminal.

[0125] Step 130 may be the same as Step 130 of FIG. 1e, and in this example, it may be assumed that the terminal has reported multiple types of available AI / ML models.

[0126] The base station and the terminal can request and collect information about each AI / ML model the terminal has through UEInformationRequest, UEInformationResponse (or UAI(UE assistance information)) messages or other messages with a similar purpose through steps 140 and 150, and the message transmitted by the terminal to the base station may include at least one of the following information.

[0127] - Expected accuracy for each available AI / ML type of the terminal: The terminal may provide the base station with information regarding the expected performance when running an AI / ML model, using information available from the current serving cell for each model. This information may be a numerical value representing the average error between the measured value and the expected value.

[0128] - Terminal preference for each available AI / ML type: The terminal can provide its preference for each AI / ML type to the base station. For example, if the terminal's battery level is high, it is charging, or its internal temperature is determined to be low, the terminal may report to the base station that it has a high preference for an AI / ML model with high complexity but excellent predictive performance. Conversely, in situations such as when the terminal's current battery level is low or its internal temperature is high, the terminal may report to the base station that it prefers a model with lower predictive performance but low complexity that does not place a burden on operation. The reporting method may involve listing the models considered good by the terminal in ascending or descending order, or assigning scores to each preference.

[0129] - Training status based on the current serving cell for each available AI / ML type of the terminal: The terminal can provide the base station with information such as whether the AI / ML model can operate immediately based on the information available in the current serving cell, whether additional training is required, or whether such information is currently unknown.

[0130] Based on the information obtained through steps 140 and 150, the base station may command the terminal to perform an RRM prediction operation through an RRCReconfiguration message in step 160, and may instruct the terminal to use a specific AI / ML model. Depending on whether the RRCReconfiguration content can be applied or not, the terminal may transmit an Applicability Report or an RRCReconfigurationComplete message to the base station, and such operations may be applied in the same or similar manner as those described in FIG. 1e and FIG. 1f.

[0131] FIG. 1h is a diagram showing an example of an AI model used to perform RRM Prediction according to one embodiment of the present disclosure.

[0132] Referring to FIG. 1h, the AI ​​model used by the terminal to perform the RRM prediction described above may include an LSTM model composed of Long Short-Term Memory (LSTM) Blocks. The LSTM model is a general-purpose model with excellent processing power for time series inputs (100), and may be configured to perform time series prediction by utilizing and reconstructing the output of the LSTM model. Additionally, the LSTM model of the terminal may include a structure that stacks LSTM Blocks (110, 120, 130) by repeatedly connecting them (e.g., k, where k is a positive integer), and this may be called a Stacked LSTM structure. The terminal and the base station may use an Autoencoder structure as an AI model that utilizes the Stacked LSTM structure as an Encoder and uses another similar structure of Stacked LSTM (composed of j LSTM Block layers, where j is a positive integer) (150, 160, 170) as a Decoder. In this case, k, the number of LSTM blocks on the Encoder side, and j, the number of LSTM blocks on the Decoder side, may be the same or different. That is, the number of LSTM block layers included in the Stacked LSTM structure of the Encoder and the number of LSTM block layers included in the Stacked LSTM structure of the Decoder may be the same or different.

[0133] As previously explained in FIG. 1c, when an AI model is developed for the purpose of reducing the measurement overhead of a terminal, there may be unmeasured information, i.e., missing values, in the input of the AI ​​model. In the case of a structure including a general LSTM model, it may malfunction in response to AI inputs containing such missing values, and to prevent this, a structure including an Autoencoder may be considered. In the case of an AI model using the aforementioned Autoencoder structure, during the process in which the output of the encoder is compressed into a latent representation (140), only important information from the input can be selected, and by undergoing the process of restoring the latent representation (140) in the decoder, a model robust against missing values ​​or noise can be constructed. Additionally, a fully connected (FC) (180) layer, which is a general neural network structure, can be added to the Stacked LSTM output portion of the Decoder to finally configure the model to express the desired output (190).

[0134] Although the structure of the AI ​​model described above was explained based on a terminal, it can be utilized by any object responsible for AI functions in a base station or core network as well as by a terminal, and the inputs and outputs of the AI ​​model may be identical or similar to those described above. If necessary, the inputs and outputs of the AI ​​model may be information provided by a terminal or other external objects.

[0135] Although the above description explains an example of an AI model that constructs an Autoencoder using an LSTM model, the form of an autoencoder can be constructed using other models other than the LSTM model, such as RNN (recurrent neural network), CNN (convolutional neural network), DNN (deep neural network), and Transformer. In addition, instead of using an LSTM model or other models to construct an autoencoder, methods such as creating virtual measurements by processing missing values ​​in a different way (e.g., by replacing missing values ​​with previously measured values) and designing the AI ​​model can also be considered.

[0136] Referring to FIG. 1i, a method for selecting a terminal model and performing RRM measurement according to one embodiment of the present disclosure can be described as follows.

[0137] Before performing RRM measurement, the terminal may start a model selection step to determine whether to perform the measurement using a non-AI / ML based model or an AI / ML based model (1i-10).

[0138] In one embodiment, the model selection step can be understood as a starting step for performing RRM measurement or prediction.

[0139] When the model selection phase begins, the terminal can report to the network the types of AI / ML input information that the terminal can support, the types of AI / ML models, and whether each AI / ML model is trained (1i-20).

[0140] This reporting process can be performed for each of the Basic AI / ML model and the Advanced AI / ML model, and if the Advanced AI / ML model is defined by a combination of multiple input information or multiple versions, the reporting process can be performed for each combination of input information or for each version.

[0141] Afterward, the terminal determines whether there is a valid configuration for performing AI / ML-based operations based on configuration information provided from the network (1i-30).

[0142] The aforementioned valid configuration may include AI / ML operation-related configurations agreed upon between the terminal and the network, for example, through the exchange process of RRCReconfiguration messages and the subsequent exchange process of RRCReconfigurationComplete messages.

[0143] If there is no valid configuration for AI / ML-based operation, the terminal can perform RRM measurement using an existing non-AI / ML-based model without using an AI / ML model (1i-40).

[0144] If there is a valid configuration for AI / ML-based operation, the terminal determines whether there is input information from the network for performing an Advanced AI / ML model (1i-50).

[0145] Here, the existence of input information for performing an Advanced AI / ML model can be determined, for example, based on whether the input information is received from a network and stored in a terminal.

[0146] If there is no input information for an Advanced AI / ML model, the terminal can perform a Basic AI / ML model that does not rely on network-provided input information (1i-70).

[0147] According to one embodiment, the input of the Basic AI / ML model may include RRM-related parameters that the terminal can measure independently, such as RSRP, RSRQ, SINR, CQI, or historical information thereof.

[0148] In addition, the execution of the above Basic AI / ML model may be performed based on prior agreement between the network and the terminal, rather than solely on the terminal's own judgment.

[0149] For example, the terminal can consult with the network in advance regarding the performance of RRM measurement or prediction using a Basic AI / ML model through the exchange process of RRCReconfiguration messages and the subsequent exchange process of RRCReconfigurationComplete messages.

[0150] While performing a Basic AI / ML model, the terminal can monitor changes in the existence or availability of input information for an Advanced AI / ML model (1i-80), and if a predefined condition is satisfied, it can selectively transmit a signal to the network requesting the provision of said input information (1i-90).

[0151] If input information for an Advanced AI / ML model exists, the terminal can additionally determine whether the input information is valid.

[0152] Here, the validity determination of the input information may be performed based on validity period information provided by the network, predefined validity conditions, or judgment criteria considering the internal state of the terminal.

[0153] That is, in the present disclosure, the determination of the validity of input information for an Advanced AI / ML model can be understood as being performed on the premise that said input information exists in the terminal.

[0154] If input information for an advanced AI / ML model is valid, the terminal can perform an advanced AI / ML model using the input information (1i-60).

[0155] Meanwhile, in one embodiment, when the input information is no longer valid, the terminal may stop executing the Advanced AI / ML model and switch to the Basic AI / ML model to operate.

[0156] Additionally, according to an embodiment of the present disclosure, in the process of transmitting a measurement report to a network, the terminal may also report information identifying which model, among a non-AI / ML based model, a Basic AI / ML model, or an Advanced AI / ML model, generated the measurement result.

[0157] Alternatively, the terminal may report indicators related to expected error, reliability, or accuracy when using the AI / ML model, along with the above measurement results.

[0158] Furthermore, the ApplicabilityReport described in another embodiment of the present disclosure may be reported for each AI / ML model, and the terminal may provide applicability information for each of the Basic AI / ML model and the Advanced AI / ML model to the network.

[0159] Accordingly, the terminal according to the present disclosure can stably perform RRM measurement and prediction operations even in situations of discontinuity of network-provided information or signaling delay by flexibly selecting and switching models based on network configuration status and the existence and validity of input information for an Advanced AI / ML model.

[0160] FIG. 2 is a drawing illustrating a terminal device according to one embodiment of the present disclosure.

[0161] Referring to FIG. 2, the terminal may include an RF (Radio Frequency) processing unit (2-10), a baseband processing unit (2-20), a storage unit (2-30), and a control unit (2-40). The configuration of the terminal is not limited to the exemplary configuration shown in FIG. 2 and may include fewer or more configurations than the configuration shown in FIG. 2.

[0162] The RF processing unit (2-10) can perform functions for transmitting and receiving signals through a wireless channel, such as signal band conversion and amplification. For example, the RF processing unit (2-10) can up-convert a baseband signal provided by the baseband processing unit (2-20) into an RF band signal and then transmit it through an antenna, and can down-convert an RF band signal received through an antenna into a baseband signal. For example, the RF processing unit (2-10) may include a transmission filter, a reception filter, an amplifier, a mixer, an oscillator, a DAC (digital to analog converter), an ADC (analog to digital converter), etc., but is not limited to these examples. Although only one antenna is shown in FIG. 2, the terminal may be equipped with multiple antennas. In addition, the RF processing unit (2-10) may include multiple RF chains. Furthermore, the RF processing unit (2-10) may perform beamforming. For beamforming, the RF processing unit (2-10) can adjust the phase and magnitude of each of the signals transmitted and received through multiple antennas or antenna elements. Additionally, the RF processing unit (2-10) can perform MIMO and can receive multiple layers when performing MIMO operation.

[0163] The baseband processing unit (2-20) can perform conversion functions between baseband signals and bit sequences according to the physical layer specifications of the system. For example, when transmitting data, the baseband processing unit (2-20) can generate complex symbols by encoding and modulating the transmitted bit sequence. Additionally, when receiving data, the baseband processing unit (2-20) can restore the received bit sequence by demodulating and decoding the baseband signal provided by the RF processing unit (2-10). For example, in the case of following the OFDM (orthogonal frequency division multiplexing) method, when transmitting data, the baseband processing unit (2-20) can generate complex symbols by encoding and modulating the transmitted bit sequence, map the generated complex symbols to subcarriers, and then construct OFDM symbols through IFFT (inverse fast Fourier transform) operation and CP (cyclic prefix) insertion. Additionally, upon receiving data, the baseband processing unit (2-20) can divide the baseband signal provided by the RF processing unit (2-10) into OFDM symbol units, restore the signals mapped to subcarriers through a fast Fourier transform (FFT) operation, and then restore the received bit sequence through demodulation and decoding.

[0164] The baseband processing unit (2-20) and the RF processing unit (2-10) can transmit and receive signals as described above. Accordingly, the baseband processing unit (2-20) and the RF processing unit (2-10) may be referred to as a transmitting unit, a receiving unit, a transmitting and receiving unit, or a communication unit. Furthermore, at least one of the baseband processing unit (2-20) and the RF processing unit (2-10) may include a plurality of communication modules to support a plurality of different wireless access technologies. Additionally, at least one of the baseband processing unit (2-20) and the RF processing unit (2-10) may include different communication modules to process signals of different frequency bands. For example, different wireless access technologies may include wireless LAN (e.g., IEEE 802.11), cellular network (e.g., LTE), etc. In addition, different frequency bands may include super high frequency (SHF) bands (e.g., 2.NRHz, NRHz) and millimeter wave (e.g., 60GHz) bands. The terminal can transmit and receive signals with the gNB using the baseband processing unit (2-20) and the RF processing unit (2-10), and the signals may include control information and data.

[0165] The storage unit (2-30) can store data such as a basic program, an application program, and setting information for the operation of the terminal. For example, the storage unit (2-30) can store data information such as a basic program, an application program, and setting information for the operation of the terminal. In addition, the storage unit (2-30) can provide the stored data upon a request from the control unit (2-40).

[0166] The storage unit (2-30) may be composed of a storage medium or a combination of storage media such as ROM, RAM, hard disk, CD-ROM, and DVD. Additionally, the storage unit (2-30) may be composed of multiple memories. According to one embodiment of the present disclosure, the storage unit (2-30) may store a program for performing a measurement prediction method according to the present disclosure.

[0167] The control unit (2-40) can control the overall operations of the terminal. For example, the control unit (2-40) can transmit and receive signals through the baseband processing unit (2-20) and the RF processing unit (2-10).

[0168] Additionally, the control unit (2-40) can write and read data to and from the storage unit (2-30). To this end, the control unit (2-40) may include at least one processor. For example, the control unit (2-40) may include a communication processor (CP) that performs control for communication and an application processor (AP) that controls upper layers such as applications. Additionally, according to one embodiment of the present disclosure, the control unit (2-40) may include a multi-connection processing unit (2-42) configured to process a process operating in a multi-connection mode. Additionally, at least one component within the terminal may be implemented as a single chip.

[0169] FIG. 3 is a drawing illustrating a base station device according to one embodiment of the present disclosure.

[0170] The base station of Fig. 3 may be included in the aforementioned network.

[0171] As illustrated in FIG. 3, the base station may include an RF processing unit (3-10), a baseband processing unit (3-20), a backhaul communication unit (3-30), a storage unit (3-40), and a control unit (3-50). The configuration of the base station is not limited to the exemplary configuration illustrated in FIG. 3, and the base station may include fewer or more configurations than the configuration illustrated in FIG. 3. The RF processing unit (3-10) may perform functions for transmitting and receiving signals through a wireless channel, such as signal band conversion and amplification. For example, the RF processing unit (3-10) may up-convert a baseband signal provided by the baseband processing unit (3-20) into an RF band signal and then transmit it through an antenna, and may down-convert an RF band signal received through an antenna into a baseband signal. For example, the RF processing unit (3-10) may include a transmission filter, a reception filter, an amplifier, a mixer, an oscillator, a DAC, an ADC, etc. In FIG. 3, only one antenna is shown, but the RF processing unit (3-10) may be equipped with multiple antennas. Additionally, the RF processing unit (3-10) may include multiple RF chains. Furthermore, the RF processing unit (3-10) may perform beamforming. For beamforming, the RF processing unit (3-10) may adjust the phase and magnitude of each of the signals transmitted and received through multiple antennas or antenna elements. The RF processing unit (3-10) may perform down-MIMO operation by transmitting one or more layers.

[0172] The baseband processing unit (3-20) can perform conversion functions between baseband signals and bit sequences according to physical layer specifications. For example, when transmitting data, the baseband processing unit (3-20) can generate complex symbols by encoding and modulating the transmitted bit sequence. Additionally, when receiving data, the baseband processing unit (3-20) can restore the received bit sequence by demodulating and decoding the baseband signal provided by the RF processing unit (3-10). For example, in the case of an OFDM method, when transmitting data, the baseband processing unit (3-20) can generate complex symbols by encoding and modulating the transmitted bit sequence, map the generated complex symbols to subcarriers, and then construct OFDM symbols through IFFT operations and CP insertion. Additionally, upon receiving data, the baseband processing unit (3-20) can divide the baseband signal provided by the RF processing unit (3-10) into OFDM symbol units, restore the signals mapped to subcarriers through FFT operations, and then restore the received bit sequence through demodulation and decoding. The baseband processing unit (3-20) and the RF processing unit (3-10) can transmit and receive signals as described above. Accordingly, the baseband processing unit (3-20) and the RF processing unit (3-10) may be referred to as a transmitting unit, a receiving unit, a transmitting and receiving unit, a communication unit, or a wireless communication unit. A base station can transmit and receive signals with a terminal using the baseband processing unit (3-20) and the RF processing unit (3-10), and the signal may include control information and data.

[0173] The backhaul communication unit (3-30) can provide an interface for communicating with other nodes within the network. For example, the backhaul communication unit (3-30) can convert a bit sequence transmitted from the base station (or main base station) to another node, e.g., an auxiliary base station, a core network, etc., into a physical signal, and convert a physical signal received from another node into a bit sequence.

[0174] The storage unit (3-40) can store data such as basic programs, application programs, and configuration information for the operation of the base station. For example, the storage unit (3-40) can store information about a bearer assigned to a connected terminal, measurement results reported from the connected terminal, etc. Additionally, the storage unit (3-40) can store information that serves as a criterion for determining whether to provide or discontinue multiple connections to the terminal. Furthermore, the storage unit (3-40) can provide the stored data upon a request from the control unit (3-50). The storage unit (3-40) may be composed of a storage medium or a combination of storage media such as ROM, RAM, hard disk, CD-ROM, and DVD. Additionally, the storage unit (3-40) may be composed of multiple memories. According to one embodiment of the present disclosure, the storage unit (3-40) may store a program for performing methods to support the measurement prediction method according to the present disclosure.

[0175] The control unit (3-50) can control the overall operations of the base station. For example, the control unit (3-50) can transmit and receive signals through the baseband processing unit (3-20) and the RF processing unit (3-10) or through the backhaul communication unit (3-30). Additionally, the control unit (3-50) can write and read data to and from the storage unit (3-40). To this end, the control unit (3-50) may include at least one processor. Furthermore, according to one embodiment of the present disclosure, the control unit (3-50) may include a multi-connection processing unit (3-52) configured to process a process operating in a multi-connection mode.

[0176] Methods according to the embodiments described in the claims or specification of the present disclosure may be implemented in the form of hardware, software, or a combination of hardware and software.

[0177] When implemented in software, a computer-readable storage medium may be provided for storing one or more programs (software modules). One or more programs stored in the computer-readable storage medium are configured for execution by one or more processors within an electronic device. One or more programs include instructions that cause the electronic device to execute methods according to the embodiments described in the claims or specification of this disclosure.

[0178] Such programs (software modules, software) may be stored in random access memory, non-volatile memory including flash memory, ROM (Read Only Memory), Electrically Erasable Programmable Read Only Memory (EEPROM), magnetic disc storage devices, Compact Disc-ROM (CD-ROM), Digital Versatile Discs (DVDs), or other forms of optical storage devices, magnetic cassettes. Alternatively, they may be stored in memory composed of some or all of these. Additionally, each constituent memory may include multiple units.

[0179] Additionally, the above program may be stored on an attachable storage device that can be accessed via a communication network such as the Internet, Intranet, Local Area Network (LAN), Wide LAN (WLAN), or Storage Area Network (SAN), or a combination thereof. Such a storage device may be connected to a device performing an embodiment of the present disclosure through an external port. Additionally, a separate storage device on a communication network may be connected to a device performing an embodiment of the present disclosure.

[0180] In the present disclosure, the terms “computer program product” or “computer readable medium” are used to collectively refer to media such as memory, a hard disk installed in a hard disk drive, and signals. These “computer program product” or “computer readable medium” are configurations provided in a method for reporting terminal capability in a wireless communication system according to the present disclosure.

[0181] A device-readable storage medium may be provided in the form of a non-transitory storage medium. Here, 'non-transitory storage medium' simply means that it is a tangible device and does not contain a signal (e.g., electromagnetic waves), and the term does not distinguish between cases where data is stored semi-permanently and cases where it is stored temporarily. For example, a 'non-transitory storage medium' may include a buffer in which data is stored temporarily.

[0182] According to one embodiment, the method according to the various embodiments disclosed herein may be provided by being included in a computer program product. The computer program product may be traded between a seller and a buyer as a product. The computer program product may be distributed in the form of a device-readable storage medium (e.g., compact disc read-only memory (CD-ROM)), or distributed online (e.g., download or upload) through an application store (e.g., Play Store™) or directly between two user devices (e.g., smartphones). In the case of online distribution, at least a portion of the computer program product (e.g., downloadable app) may be temporarily stored or temporarily created on a device-readable storage medium, such as the memory of a manufacturer's server, an application store's server, or a relay server.

[0183] In the specific embodiments of the present disclosure described above, the components included in the disclosure are expressed in a singular or plural form according to the specific embodiments presented. However, the singular or plural expression is selected to suit the situation presented for convenience of explanation, and the present disclosure is not limited to singular or plural components; even if a component is expressed in the plural form, it may be composed of a singular form, and even if a component is expressed in the singular form, it may be composed of a plural form.

[0184] Meanwhile, the embodiments of the present disclosure disclosed in this specification and drawings are merely specific examples provided to facilitate the explanation of the technical content of the present disclosure and to aid in understanding the present disclosure, and are not intended to limit the scope of the present disclosure. That is, it is obvious to those skilled in the art that other variations based on the technical concept of the present disclosure are possible.

[0185] In addition, each of the above embodiments may be combined and operated as needed. For example, parts of one embodiment of the present disclosure and another embodiment may be combined to operate a base station and a terminal.

[0186] Furthermore, the embodiments of the present disclosure are applicable to other communication systems, and other variations based on the technical concept of the embodiments may also be implemented. For example, the embodiments may be applied to LTE systems, 5G, NR systems, or 6G systems, etc. Therefore, the scope of the present disclosure should not be limited to the described embodiments but should be defined by the claims set forth below as well as equivalents thereof.

Claims

1. A method performed by a UE (user equipment) of a wireless communication system, A step of receiving radio resource management (RRM) configuration information including cell geographical information from a base station; A step of performing RRM prediction by using the above cell geographic information and cell measurement information as input to an AI / ML (Artificial Intelligence / Machine Learning) model; A step of detecting the triggering of a measurement report event based on RRM measurement results and RRM prediction results; and A method comprising the step of transmitting a measurement report message, including the RRM measurement result and the RRM prediction result, to the base station.

2. In Paragraph 1, A method comprising at least one of the cell geographic information, wherein the cell geographic information comprises co-located cell information related to the serving cell of the UE, a list of adjacent cells of the serving cell, priority information between adjacent cells of the serving cell, antenna location information of each adjacent cell of the serving cell, or antenna direction information of each adjacent cell of the serving cell.

3. In Paragraph 1, The step of receiving the RRM setting information including the cell geographic information is: A step of receiving static cell geographic information from the above base station through a System Information Block (SIB); or A method comprising at least one step of receiving dynamic cell geographic information from the above base station through an RRC (radio resource control) reconfiguration message.

4. In Paragraph 1, The above method is: The method further includes the step of transmitting RRM prediction capability information of the UE to the base station. A method comprising at least one of the above capability information, which includes information on the types of AI / ML models supported by the UE, information on the input and output types of each AI / ML model, or information on whether each AI / ML model has been trained.

5. In Paragraph 1, If the AI / ML model associated with the above RRM configuration information has not been trained, the above method is: A step of transmitting an applicability report message to the base station to indicate that the AI / ML model is in an untrained state; and A method further comprising the step of updating the AI / ML model by receiving at least one of data for training the AI / ML model or information on a trained model from the base station.

6. In Paragraph 1, The step of performing the above RRM prediction is: A step of selecting a non-AI / ML model if valid settings for a specific AI / ML model do not exist; A step of selecting a basic AI / ML model when a valid setting for the specific AI / ML model exists and the cell geographic information does not include information related to the specific AI / ML model; and A method comprising the step of selecting an advanced AI / ML model when a valid setting for the specific AI / ML model exists and the cell geographic information includes information related to the specific AI / ML model.

7. In Paragraph 6, The step of performing the above RRM prediction is: A method further comprising the step of switching to the basic AI / ML model when, while performing the RRM prediction based on the improved AI / ML model, the validity period of the cell geographic information expires or the validity condition of the cell geographic information is not satisfied.

8. Regarding UE (user equipment): At least one transceiver; At least one processor communicatively coupled to the above at least one transceiver; and It includes at least one memory that is communicationally coupled to the above at least one processor and stores instructions, and The above instructions are executed individually or in any combination by the above at least one processor, so that the UE: Receive radio resource management (RRM) configuration information including cell geographical information from a base station, and By using the above cell geographic information and cell measurement information as input to an AI / ML (Artificial Intelligence / Machine Learning) model, RRM prediction is performed, and Detecting the triggering of measurement report events based on RRM measurement results and RRM prediction results, and A UE that transmits a measurement report message including the above RRM measurement result and the above RRM prediction result to the base station.

9. In Paragraph 8, The cell geographic information comprises at least one of co-located cell information related to the serving cell of the UE, a list of adjacent cells of the serving cell, priority information between adjacent cells of the serving cell, antenna location information of each adjacent cell of the serving cell, or antenna direction information of each adjacent cell of the serving cell.

10. In Paragraph 8, The above cell geographic information is: A UE that receives static cell geographic information via a System Information Block (SIB) or dynamic cell geographic information via a Radio Resource Control (RRC) reconfiguration message.

11. In Paragraph 8, The above commands cause the UE to transmit the UE's RRM prediction capability information to the base station, and The above capability information comprises at least one of information regarding the types of AI / ML models supported by the UE, information regarding the input and output types of each AI / ML model, or information regarding whether each AI / ML model has been trained.

12. In Paragraph 8, If the AI / ML model associated with the above RRM configuration information has not been trained, the above commands are for the UE: Sending an applicability report message to the base station to indicate that the above AI / ML model is in an untrained state, and A UE that updates the AI / ML model by receiving at least one of data for training the AI / ML model or information on a trained model from the base station.

13. In Paragraph 8, The above RRM prediction is: If valid settings for a specific AI / ML model do not exist, a non-AI / ML model is selected and executed, and If a valid setting for the specific AI / ML model exists and the cell geographic information does not include information related to the specific AI / ML model, a basic AI / ML model is selected and executed. If a valid configuration for the above specific AI / ML model exists and the cell geographic information includes information related to the above specific AI / ML model, an advanced AI / ML model is selected and executed, and A UE that, while performing the RRM prediction based on the improved AI / ML model, switches the improved AI / ML model to the basic AI / ML model if the validity period of the cell geographic information expires or the validity condition of the cell geographic information is not satisfied.

14. A method performed by a base station of a wireless communication system, A step of transmitting RRM (radio resource management) configuration information including cell geographical information to UE (user equipment); and A method comprising the step of receiving a measurement report message including the RRM measurement result and the RRM prediction result from the UE according to a measurement report event based on the RRM prediction result and the RRM measurement result of an AI / ML (Artificial Intelligence / Machine Learning) model based on the cell geographic information and cell measurement information.

15. Regarding base stations: At least one transceiver; At least one processor communicatively coupled to the above at least one transceiver; and It includes at least one memory that is communicationally coupled to the above at least one processor and stores instructions, and The above instructions are executed individually or in any combination by the above at least one processor, so that the base station: RRM (radio resource management) configuration information including cell geographical information is transmitted to UE (user equipment), and A base station that receives a measurement report message including the RRM measurement result and the RRM prediction result from the UE according to a measurement report event based on the RRM prediction result and the RRM measurement result of an AI / ML (Artificial Intelligence / Machine Learning) model based on the cell geographic information and cell measurement information.

Images