Method and device for performing handover procedure in wireless communication system

Abstract

Disclosed is a method for performing a handover. The method for performing a handover, which is performed by a base station, comprises the steps of: receiving a measurement control report from a terminal; determining a handover on the basis of the measurement control report from the terminal; and transmitting a handover request to a target base station, wherein the handover request includes AI / ML-related information about the terminal.

Description

Method and device for performing a handover procedure in a wireless communication system

[0001] The present disclosure relates to a handover procedure in a wireless communication system, and more specifically, to a method and apparatus for performing a handover procedure in a wireless communication system that supports AI / ML.

[0002] The communication system may include a core network, base stations (e.g., macro base stations, small base stations, relays, etc.), terminals, etc. Communication between the base station and the terminal may be performed based on various radio access technologies (RATs) (e.g., 4G communication technology, 5G communication technology, WiBro (wireless broadband) technology, WLAN (wireless local area network) technology, WPAN (wireless personal area network) technology, etc.).

[0003] Handover is one of the mobility management technologies in wireless communication systems. Handover is a technology that allows a terminal to disconnect from its existing source base station and establish a connection with a new target base station when it moves from one cell or base station to another. Through handover, terminals in wireless communication systems can receive stable service.

[0004] 3GPP selected AI / ML as a Study Item for 5G Release 18 and conducted research and discussions. AI / ML technology utilizes artificial intelligence and machine learning to streamline network operations and optimize service quality. 3GPP is prioritizing the introduction of AI / ML in scenarios such as 5G beam management, positioning, channel state prediction, and channel state compression, expecting improvements in system throughput, resource utilization, and coverage.

[0005] According to at least one embodiment, a method and apparatus are disclosed in which at least one communication node in a communication system supporting AI / ML can perform a handover.

[0006] In one aspect, a handover method performed by a base station is disclosed. The disclosed method comprises the steps of: receiving a measurement control report of a terminal; determining a handover based on the measurement control report of the terminal; and transmitting a handover request to a target base station, wherein the handover request includes AI / ML-related information of the terminal.

[0007] According to at least one embodiment, AI / ML-related information can be shared between a terminal and a target base station while minimizing the increase in signaling overhead during a handover procedure. According to at least one embodiment, the source base station can increase the benefits from AI / ML utilization by determining the target base station by considering at least one of the terminal's AI / ML-related information, the target base station's AI / ML-related information, and the source base station's AI / ML-related information. According to at least one embodiment, the effect of AI / ML utilization can be maximized by adjusting the handover margin by considering the AI / ML-related information during the handover process, either by the source base station or the terminal. According to at least one embodiment, in a conventional handover procedure, the terminal must go through four additional steps, such as reconfiguring the RRC to use the AI / ML model after handover to the target base station; however, by exchanging AI / ML-related information in advance during the handover process, the RRC configuration for utilizing AI / ML functions can be processed in advance. According to at least one embodiment, after handover to a target base station that supports AI / ML functions, the terminal can monitor the performance of the base station and the AI / ML model by reusing the previously used AI / ML model. If the performance of a recycled AI / ML model is better than that of a legacy mobile communication system that does not use AI / ML models, the time required for training AI / ML models can be reduced by continuously reusing the model. Furthermore, through the recycling of existing AI / ML models, the terminal can utilize AI / ML even before building a new AI / ML model.

[0008] FIG. 1 is a drawing showing a wireless communication system according to an exemplary embodiment.

[0009] Figure 2 is a block diagram illustrating the configuration of a communication node that constitutes a communication system in an exemplary manner.

[0010] Figure 3 is a diagram illustrating an exemplary 5G network architecture.

[0011] Figure 4 is a block diagram showing the full cycle of AI / ML functions in a mobile communication system.

[0012] FIG. 5 is a flowchart illustrating the process of notifying the network of the capabilities of a terminal (120-1) in a mobile communication system.

[0013] FIG. 6 is a flowchart illustrating the process of performing a handover procedure according to an exemplary embodiment.

[0014] FIG. 7 is a conceptual diagram exemplarily illustrating the method by which a source base station selects a target base station in step S104 of FIG. 6.

[0015] FIG. 8 is a conceptual diagram exemplarily illustrating the method by which a source base station selects a target base station in step S104 of FIG. 6.

[0016] FIG. 9 is a flowchart illustrating the process of performing a conditional handover (CHO) according to an exemplary embodiment.

[0017] Figure 10 is a diagram showing a variation of the handover procedure shown in Figure 6.

[0018] Figure 11 is a diagram showing a variation of the handover procedure shown in Figure 6.

[0019] FIG. 12 is a conceptual diagram exemplarily illustrating a source base station adjusting the handover margin according to an exemplary embodiment.

[0020] FIG. 13 is a conceptual diagram exemplarily illustrating a source base station adjusting the handover margin according to an exemplary embodiment.

[0021] FIG. 14 is a conceptual diagram exemplarily illustrating a source base station adjusting the handover margin according to an exemplary embodiment.

[0022] Figure 15 is a flowchart showing the handover procedure according to the comparative example.

[0023] FIG. 16 is a flowchart illustrating the process of performing step S620 of FIG. 15.

[0024] Figure 17 is a flowchart illustrating a monitoring procedure for reusing an existing trained AI / ML model after the handover procedure is completed.

[0025] Figure 18 is a flowchart illustrating a monitoring procedure for reusing an existing trained AI / ML model after the handover procedure is completed.

[0026] Figure 19 is a flowchart illustrating a monitoring procedure for reusing an existing trained AI / ML model after the handover procedure is completed.

[0027] Figure 20 is a flowchart illustrating a monitoring procedure for reusing an existing trained AI / ML model after the handover procedure is completed.

[0028] In one aspect, a handover method performed by a base station is disclosed. The disclosed method comprises the steps of: receiving a measurement control report of a terminal; determining a handover based on the measurement control report of the terminal; and transmitting a handover request to a target base station, wherein the handover request includes AI / ML-related information of the terminal.

[0029] The above handover request includes information elements used for handover preparation, and the information elements used for handover preparation may include AI / ML-related information of the terminal.

[0030] The above handover request includes an information element used for handover preparation, the information element used for handover preparation includes auxiliary information of the terminal, the auxiliary information of the terminal includes AI / ML auxiliary information of the terminal to assist in the implementation of AI / ML functions of the terminal, and the AI / ML auxiliary information of the terminal may include at least one of the current state of the terminal related to the implementation of AI / ML functions, power saving according to the request of the terminal, and an option preferred by the terminal in relation to the implementation of AI / ML functions.

[0031] The AI / ML-related information of the terminal may include at least one of the following: whether the terminal supports AI / ML functions, the type of AI / ML function supported by the terminal, the ID of the AI / ML model supported by the terminal, the Associated ID of the AI / ML being used by the terminal, the ID of the AI / ML model being used by the terminal, the validity period of the AI / ML model used by the terminal (120-1), whether the AI / ML model being used by the terminal is a cell-specific model or a generalized model, the data collection resolution supported by the terminal, information regarding the configuration or weight of the AI / ML model being used by the terminal, the device that the terminal can allocate for the AI / ML model, the number of devices that the terminal can allocate for the AI / ML model, and the computing performance category of the terminal.

[0032] The above method further includes the step of obtaining capability information of the terminal from the terminal, and the capability information of the terminal may include AI / ML-related information of the terminal.

[0033] The above method further includes the step of acquiring AI / ML-related information of at least one neighboring base station, and in the step of determining the handover, the target base station can be determined based on the AI / ML-related information of the terminal and the AI / ML-related information of the at least one neighboring base station.

[0034] If the above terminal supports AI / ML, the base station may determine a higher priority for a neighboring base station that supports AI / ML among at least one neighboring base station.

[0035] In the step of determining a handover based on the measurement control report of the terminal, if the handover margin is satisfied, a handover is determined, and the handover margin may be determined based on at least one of the AI / ML related information of the terminal, the AI / ML related information of the base station, and the AI / ML related information of the target base station.

[0036] If the terminal and the base station support AI / ML and the target base station does not support AI / ML, the handover margin may be determined to be higher than the handover margin of the legacy system.

[0037] If the terminal and the target base station support AI / ML and the base station does not support AI / ML, the handover margin may be determined to be smaller than the handover margin of the legacy system.

[0038] The above method further includes the step of receiving a handover request response from the target base station, wherein the handover request response includes AI / ML related information of the target base station, and the AI / ML related information of the target base station may include information on at least one of the following: an ID of an AI / ML model supported by the target base station, an AI / ML association ID supported by the target base station, a type of AI / ML function supported by the target base station, information on whether the AI / ML model supported by the target base station is a cell-specific model or a generalized model, a data collection resolution for learning the AI / ML model of the target base station, a configuration or weight of a learned AI / ML model supported by the target base station, a device (CPU, NPU, GPU, etc.) that the target base station can allocate for the AI / ML model, and the number of devices that the target base station (110-2) can allocate for the AI / ML model.

[0039] The above method includes the step of transmitting a message that triggers a handover to the terminal, and the message that triggers the handover may include AI / ML-related information of the target base station.

[0040] The message triggering the above handover may be an RRCReconfiguration message.

[0041] A base station comprising a communication unit and a processor is disclosed. The processor performs the steps of receiving a measurement control report from a terminal; determining a handover based on the measurement control report from the terminal; and transmitting a handover request to a target base station, wherein the handover request may include AI / ML-related information of the terminal.

[0042] A handover method performed by a target base station is disclosed. The disclosed method includes the steps of: receiving a handover request from a source base station; and transmitting a handover request response to the source base station, wherein the handover request may include AI / ML-related information of a terminal performing the handover.

[0043] The above handover request response may include AI / ML-related information of the target base station.

[0044] The above method may further include the step of receiving a report of inference values ​​of the reused AI / ML model from the terminal when the terminal reuses the AI / ML model used at the source base station after the handover is completed; and the step of monitoring the performance of the reused AI / ML model.

[0045] The above method may further include the step of obtaining performance monitoring results of the reused AI / ML model from the terminal when the terminal reuses the AI / ML model used at the source base station after the handover is completed.

[0046] The above method may further include the steps of: after a handover is completed, reusing the AI / ML model that the terminal used at the source base station, but if the AI / ML association ID used by the terminal does not match the association ID supported by the target base station, transmitting reference signals corresponding to a plurality of association IDs to the terminal; obtaining a monitoring result based on the measurement result of the reference signals corresponding to the plurality of association IDs from the terminal; and generating a management instruction for the reused AI / ML model based on the monitoring result and transmitting the management instruction to the terminal.

[0047] A handover method performed by a terminal is disclosed. The disclosed method comprises the steps of: transmitting a measurement control report to a source base station; receiving a message from the source base station that triggers a handover; and performing random access on a target base station identified by the message that triggers the handover, wherein the message that triggers the handover may include AI / ML-related information of the target base station.

[0048] The above method further includes the step of providing capability information of the terminal to the source base station, and the capability information of the terminal may include AI / ML-related information of the terminal.

[0049] The above measurement control report may include AI / ML-related information of the terminal.

[0050] The above method may further include the step of transmitting a report of the inference value of the reused AI / ML model to the target base station when the terminal reuses the AI / ML model used at the source base station after the handover is completed; and the step of receiving a management instruction for the reused AI / ML model from the target base station.

[0051] The above method may further include the step of reporting the performance monitoring results of the reused AI / ML model; and the step of receiving management instructions for the reused AI / ML model from the target base station.

[0052] The above method may further include the step of, after the handover is completed, reusing the AI / ML model that the terminal used at the source base station, but when the AI / ML association ID used by the terminal does not match the association ID supported by the target base station, receiving reference signals corresponding to a plurality of association IDs supported by the target base station from the target base station; and the step of providing a monitoring result based on the measurement result of the reference signals corresponding to the plurality of association IDs to the target base station.

[0053] A terminal comprising a communication unit and a processor is disclosed. The processor performs the steps of receiving a measurement control report of the terminal; determining a handover based on the measurement control report of the terminal; and transmitting a handover request to a target base station, wherein the handover request includes AI / ML-related information of the terminal.

[0054] The present invention is capable of various modifications and may have various embodiments, and specific embodiments are illustrated in the drawings and described in detail. However, this is not intended to limit the invention to specific embodiments, and it should be understood that the invention includes all modifications, equivalents, and substitutions that fall within the spirit and scope of the invention.

[0055] Terms such as "first," "second," etc., may be used to describe various components, but said components should not be limited by said terms. These terms are used solely for the purpose of distinguishing one component from another. For example, without departing from the scope of the present invention, the first component may be named the second component, and similarly, the second component may be named the first component. The term "and / or" includes a combination of a plurality of related described items or any of a plurality of related described items.

[0056] When it is stated that one component is "connected" or "connected" to another component, it should be understood that while it may be directly connected or connected to that other component, there may also be other components in between. On the other hand, when it is stated that one component is "directly connected" or "directly connected" to another component, it should be understood that there are no other components in between.

[0057] The terms used in this application are used merely to describe specific embodiments and are not intended to limit the invention. The singular expression includes the plural expression unless the context clearly indicates otherwise. In this application, terms such as "comprising" or "having" are intended to specify the presence of the features, numbers, steps, actions, components, parts, or combinations thereof described in the specification, and should be understood as not precluding the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0058] Unless otherwise defined, all terms used herein, including technical or scientific terms, have the same meaning as generally understood by those skilled in the art to which the present invention pertains. Terms such as those defined in commonly used dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an ideal or overly formal sense unless explicitly defined in this application.

[0059] Hereinafter, preferred embodiments of the present invention will be described in more detail with reference to the attached drawings. In order to facilitate an overall understanding of the present invention, the same reference numerals are used for identical components in the drawings, and redundant descriptions of identical components are omitted.

[0060] Table 1 shows the abbreviations used in the present disclosure.

[0061] 약어전체 이름AI / MLArtificial Intelligence / Machine Learning3GPP3rd Generation Partnership ProjectACKAcknowledgementAFApplication FunctionAIArtificial IntelligenceAMFAccess and Mobility Management FunctionAUSFAuthentication Server FunctionBWPBandwidth PartC-RNTICell RNTICSIChannel State InformationCSI-RSChannel State Information Reference SignalCLICross link InterferenceCEControl ElementDCIDownlink Control InformationIEInformation elementMACMedium Access ControlMLMachine LearningNSSFNetwork Slicing Selection FunctionNEFNetwork Exposure FunctionNRFNF Repository FunctionPCFPolicy Control FunctionPDCCHPhysical Downlink Control ChannelPDSCHPhysical Downlink Shared ChannelPDUProtocol Data UnitPRACHPhysical Random Access ChannelPT-RSPhase Tracking Reference SignalPUCCHPhysical Uplink Control ChannelPUSCHPhysical Uplink Shared ChannelRARandom AccessRACHRandom Access ChannelRANRadio Access NetworkRBResource BlockRRCRadio Resource ControlRSRPReference Signal Received PowerRSRQReference Signal ReceivedQualityRSSIReceived Signal Strength IndicatorSBFDSub-Band Full DuplexSISelf InterferenceSIBSystem information blockUEUser EquipmentDLDownlinkULUplinkRARRandom Access ResponseRRMRadio Resource ManagementRRCRadio Resource ControlRSRPReference Signal Received PowerQoSQuality of ServiceDRBData Radio BearerDAPSDual Active Protocol StackTDDTime Division DuplexingFDDFrequency Division DuplexingSSBSS BlockCQIChannel Quality IndicatorPMIPrecoding Matrix IndicatorLILayer IndicatorRIRank IndicatorQCLQuasi coLocation

[0062] FIG. 1 is a drawing showing a wireless communication system (100) according to an exemplary embodiment.

[0063] Referring to FIG. 1, a wireless communication system (100) may be composed of a plurality of communication nodes (110-1, 110-2, 110-3, 120-1, 120-2, 120-3, 120-4). Here, a communication node refers to a node capable of transmitting and receiving signals in the wireless communication system (100), and each of the plurality of communication nodes (110-1, 110-2, 110-3, 120-1, 120-2, 120-3, 120-4) may support at least one communication protocol. For example, each of the multiple communication nodes (110-1, 110-2, 110-3, 120-1, 120-2, 120-3, 120-4) can support cellular communication (e.g., LTE (long term evolution), LTE-A (advanced), 5G NR, 5G-Advanced, etc. as defined in the 3GPP (3rd generation partnership project) standard).

[0064] For example, each of the multiple communication nodes (110-1, 110-2, 110-3, 120-1, 120-2, 120-3, 120-4) can support a CDMA (code division multiple access) based communication protocol, a WCDMA (wideband CDMA) based communication protocol, a TDMA (time division multiple access) based communication protocol, a FDMA (frequency division multiple access) based communication protocol, an OFDM (orthogonal frequency division multiplexing) based communication protocol, an OFDMA (orthogonal frequency division multiple access) based communication protocol, a SC (single carrier)-FDMA based communication protocol, a NOMA (non-orthogonal multiple access) based communication protocol, a SDMA (space division multiple access) based communication protocol, a SBFD (sub-band full duplex), AI / ML, etc.

[0065] A plurality of communication nodes (110-1, 110-2, 110-3, 120-1, 120-2, 120-3, 120-4) may include a plurality of base stations (110-1, 110-2, 110-3) and a plurality of terminals (120-1, 120-2, 120-3, 120-4). Each of the base stations (110-1, 110-2, 110-3) may form a cell. The cell may include a small cell, a macro cell, a pico cell, a femto cell, etc., but the embodiments are not limited thereto. For example, the first terminal (120-1) and the second terminal (120-2) may be in the coverage of the first base station (110-1), the third terminal (120-3) may be in the coverage of the second base station (110-2), and the fourth terminal (120-4) may be in the coverage of the third base station (110-3).

[0066] Multiple base stations (110-1, 110-2, 110-3) may be referred to as gNodeB (gNB), NodeB, evolved NodeB, BTS (base transceiver station), radio base station, radio transceiver, access point, access node, roadside unit (RSU), RRH (radio remote head), TP (transmission point), TRP (transmission and reception point), relay node, etc. Multiple base stations (110-1, 110-2, 110-3) may form a Radio Access Network (RAN). The Radio Access Network may be connected to a core network.

[0067] Each of the multiple terminals (120-1, 120-2, 120-3, 120-4) may be referred to as a terminal, access terminal, mobile terminal, station, subscriber station, mobile station, portable subscriber station, node, device, etc.

[0068] The communication protocol support ranges of each of the multiple base stations (110-1, 110-2, 110-3) may differ from one another. For example, some of the multiple base stations (110-1, 110-2, 110-3) may support AI / ML functions while others do not. Similarly, the communication protocol support ranges of each of the terminals (120-1, 120-2, 120-3, 120-4) may also differ from one another. For example, some of the multiple terminals (120-1, 120-2, 120-3, 120-4) may support AI / ML functions while others do not.

[0069] FIG. 2 is a block diagram illustrating the configuration of a communication node (200) that constitutes a communication system. At least some of the communication nodes (110-1, 110-2, 110-3, 120-1, 120-2, 120-3) shown in FIG. 1 may correspond to the communication node (200) shown in FIG. 2.

[0070] Referring to FIG. 2, the communication node (200) may include at least one processor (210), a memory (220), and a transceiver (230) that is connected to a network to perform communication. Additionally, the communication node (200) may further include an input interface device (240), an output interface device (250), a storage device (260), etc. Each component included in the communication node (200) may be connected by a bus (270) to communicate with one another.

[0071] The processor (210) can execute a program command stored in at least one of the memory (220) and the storage device (260). The processor (210) may mean a central processing unit (CPU), a graphics processing unit (GPU), or a dedicated processor on which methods according to embodiments of the present invention are performed. Each of the memory (220) and the storage device (260) may be composed of at least one of a volatile storage medium and a non-volatile storage medium. For example, the memory (220) may be composed of at least one of read-only memory (ROM) and random access memory (RAM).

[0072] Figure 3 is a diagram illustrating an exemplary 5G network architecture.

[0073] Referring to FIG. 3, the 5G network architecture may include an NSSF providing a network slicing control solution, an NEF providing a network function opening solution, an NRF providing an NF interoperability control solution within the 5G network, a PCF providing a data packet flow policy control solution, a UDM providing a user information and policy management solution, an AMF providing a network access and mobility management solution, an SMF providing a terminal / network session management solution, a UPF providing a solution for user packet routing and terminal connectivity between base stations, a radio access network (RAN), and a user terminal (UE).

[0074] N1 reference point is defined to transmit signaling between the UE and the AMF, N2 is defined as the reference point to connect the RAN access node and the AMF, and N3 is defined as the reference point to connect the RAN access node and the UPF.

[0075] FIG. 4 is a block diagram illustrating the full cycle of an AI / ML function in a mobile communication system. The AI / ML function blocks shown in FIG. 6 can be implemented by at least one of the communication nodes of the mobile communication system. For example, the AI / ML function blocks shown in FIG. 4 can be implemented by at least one of a base station and a terminal.

[0076] When the first terminal (120-1) is connected to the service area of ​​the first base station (110-1) that supports AI / ML functions, the first terminal (120-1) or the first base station (110-1) can activate at least some of the function blocks (410, 412, 414, 416, 418) of FIG. 4.

[0077] The terminal (120-1) can collect data related to the wireless channel by activating the data collection block (410). The terminal (120-1) can collect data related to wireless channel information from reference signals (CSI-RS, SSB, PT-RS, PRS, downlink reference signals, etc.). When the base station (110-1) activates the data collection block (610), it can collect data related to wireless channel information from reference signals (SRS, PRS, PT-RS) and channel status reports, etc.

[0078] Data collected from the data collection block (410) can be used as data for training / monitoring / inference. The model training block (412) can train an AI / ML model using the data collected from the data collection block (410). The model storage block (418) can store the trained AI / ML model. The stored model can be used by the inference block (416). The inference block (416) can infer information necessary for the operation of a wireless communication system based on the data collected from the data collection block (410).

[0079] The inference block (416) can report the inference results to the AI / ML management block (414). The AI / ML management block (414) can perform overall functions related to training the AI / ML model, selecting the AI / ML model, enabling / disabling the AI / ML model, and controlling other function blocks. The AI / ML management block (414) can monitor the performance of the AI / ML model to maintain the performance of the AI / ML and provide feedback on the training results to the model training block (412). If the performance of the monitored AI / ML model is low and retraining is required, the AI / ML management block (414) can request retraining from the model training block (412). The AI / ML management block (414) can determine the enabling / disabling of the AI / ML model and the selection / change of the model. The AI / ML management block (414) performs management instructions such as a fallback to an existing communication method that does not use AI / ML functions.

[0080] At least one of the terminal (120-1) and base stations (110-1, 110-2, 110-3) can provide performance feedback for an AI / ML model. Below, the terminal (120-1) providing performance feedback is described as an example.

[0081] The terminal (120-1) can generate performance feedback of the AI / ML model by comparing the inference result of the AI / ML model with the measurement result measured by the terminal (120-1). For example, the terminal (120-1) can measure at least one beam using a predetermined set of monitoring resources. Based on the reception performance of the beams, the terminal (120-1) can select K beams (where K is an arbitrary natural number) with the best reception performance.

[0082] The terminal (120-1) can infer the reception performance of a beam using an AI / ML model. For example, the second AI / ML model can select M beams for the terminal (120-1) based on the inference result for at least one beam.

[0083] The terminal (120-1) can generate performance feedback of the first AI / ML model by comparing K beams derived based on the measurement results of the beams with M beams inferred by the AI / ML model. For example, if there is an intersection of K beam sets and M beam sets, the terminal (120-1) can determine that the performance of the AI / ML model is excellent. If there is no intersection of K beam sets and M beam sets, the terminal (120-1) can determine that the performance of the AI / ML model is poor. The terminal (120-1) can generate performance feedback of the AI / ML model based on at least one of the presence or absence of an intersection of K beam sets and M beam sets and the size of the intersection.

[0084] In the description above, the terminal (120-1) is described as an example of generating performance feedback of AI / ML on a model basis, but the embodiments are not limited thereto. For example, the terminal (120-1) may generate performance feedback of related AI / ML model(s) on a functionality basis. The terminal (120-1) may also generate performance feedback of AI / ML model(s) corresponding to a related ID.

[0085] The terminal (120-1) may optionally generate performance feedback of the AI / ML model. For example, the terminal (120-1) may omit generating performance feedback of the AI / ML model in consideration of the reliability of the performance feedback of the AI / ML model.

[0086] For example, if the interval between the time when the terminal (120-1) measures a predetermined parameter and the time when the AI / ML model infers the predetermined parameter is greater than a threshold value, the terminal (120-1) may determine that the reliability of the performance feedback based on the measurement result and the inference result is low. In this case, the terminal (120-1) may not generate performance feedback.

[0087] As another example, the first base station (110-1) may not use performance feedback with low reliability. For example, since the first base station (110-1) performs resource scheduling for the terminal (120-1), it can know the measurement resources used by the terminal (120-1). If the time interval between the time when the AI / ML model of the terminal (120-1) infers a predetermined parameter and the time when the terminal (120-1) measures a predetermined parameter is greater than or equal to a threshold value, the first base station (110-1) may not evaluate the AI / ML model using performance feedback obtained from the terminal (120-1). In this case, the first base station (110-1) may ignore the performance feedback without using it.

[0088] AI / ML models can be classified in various ways. For example, a model ID can be assigned to each AI / ML model. As another example, AI / ML models can be classified by function, and a functionality ID can be assigned to each function. Associated IDs may also be used to distinguish AI / ML models.

[0089] An Associated ID may represent a configuration for data collection to train an AI / ML model. An Associated ID can be used to ensure the identity between the training environment and the inference environment of the AI / ML model. A terminal (120-1) can verify the configuration for data collection from the Associated ID and collect data according to that configuration. The terminal (120-1) can train or develop an AI / ML model using the collected data based on the Associated ID.

[0090] AI / ML models possess the characteristics of overfitting and generalization. Overfitting refers to a situation where a model performs excellently only on a specific dataset but degrades on others. Conversely, generalization refers to the characteristic where performance can be guaranteed across various datasets. However, there are limitations to the generalization of AI / ML models. In other words, the performance of an AI / ML model can be guaranteed for data collected under a somewhat limited data collection configuration. Therefore, wireless communication networks can define a data collection configuration that guarantees the performance of AI / ML models and assign an Associated ID to each data collection configuration.

[0091] A wireless communication network can define a set (range) of generalizable datasets where the performance of AI / ML models is guaranteed, and determine a data collection configuration to correspond to this. The wireless communication network can define various data collection configurations and assign an Associated ID to each of them. The wireless communication network can train or develop AI / ML models according to the data collection configuration corresponding to each Associated ID.

[0092] The method for assigning Associated IDs may vary depending on the use case of the AI / ML model. For example, for AI / ML models used for beam management, data collection configurations may be determined based on the same downlink spatial domain transmission filter, reference signals in Quasi Co-location (QCL) status, or antenna panels used for transmission and reception. A different Associated ID may be assigned to each of the data collection configurations. In the example of CSI compression and prediction, data collection configurations may be determined based on SNR, CQI, PMI, RI representing channel environment status, and antenna panels used for transmission and reception. Associated IDs are uniquely defined within the network and can be classified into global Associated IDs representing the same environment and local Associated IDs valid only for certain cell groups / cells / sites / regions or equipment vendors. In the following, the Associated ID may correspond to at least one of the global Associated ID and the local Associated ID.

[0093] Each AI / ML model can be assigned a Model ID. An Associated ID can correspond to a single Model ID. As another example, an Associated ID can correspond to multiple Model IDs. The Associated ID and Model ID can have a many-to-many correspondence.

[0094] For example, if the downlink spatial domain transmission filters are configured as a number (N), the Associated IDs for beam management can also be configured as a number (N), and the AI / ML models used for the corresponding AI / ML beam management can be configured as N corresponding. Similarly, if the base station (110-1) configures a number (N) of reference signal / resource sets that are not in a QCL state, the AI / ML models used for the corresponding AI / ML beam management and the associated IDs can be configured as N corresponding. On the other hand, if the AI / ML models have excellent generalization features and their performance is guaranteed across all datasets, they can be configured and operated with a small number (less than N) of AI / ML models.

[0095] The operation of AI / ML models can be broadly classified into single-sided operation (one-side AI / ML model) and joint operation (two-side AI / ML model). In the case of a one-side AI / ML model, the AI / ML model is operated on only one side, either the base station or the terminal, whereas in the case of a two-side AI / ML model, the AI / ML model can be operated simultaneously on both the base station and the terminal.

[0096]

[0097] FIG. 5 is a flowchart illustrating the process of notifying the network of the capabilities of a terminal (120-1) in a mobile communication system.

[0098] Referring to FIG. 5, at step S50, the AMF can transmit a message (UE Capability Match Request) requesting capability information of the terminal. The first base station (110-1) can receive the message transmitted by the AMF.

[0099] In step S52, the first base station (110-1) may transmit a message (UE Capability Enquiry) requesting capability information to the first terminal (120-1) in response to a request from the AMF. In step S54, the first terminal (120-1) may transmit a message (UE Capability Information) containing information about its own capability. The first base station (110-1) may receive the message (UE Capability Information) from the first terminal (120-1).

[0100] The message (UE Capability Information) transmitted by the first terminal (120-1) may include information related to the capabilities of the first terminal (120-1). The information related to the capabilities of the first terminal (120-1) may include AI / ML related information of the first terminal (120-1).

[0101] In the following, AI / ML-related information of a communication node may include at least one of the following: whether the communication node supports AI / ML functions, the type of AI / ML function supported by the communication node, the ID of the AI / ML model supported by the communication node, the associated ID of the AI / ML being used by the communication node, the ID of the AI / ML model being used by the communication node, the type of AI / ML function being used by the communication node, the validity period of the AI / ML model being used by the communication node, whether the AI / ML model being used by the communication node is a cell-specific model or a generalized model, the data collection resolution (granularity or quantization level) supported by the communication node, information regarding the configuration or weights of the trained AI / ML model used by the communication node, devices (CPU, NPU, GPU, etc.) that the communication node can allocate for the AI / ML model, the number of devices that the communication node can allocate for the AI / ML model, and the computing performance category of the communication node.

[0102] The first base station (110-1) can obtain AI / ML related information of the first terminal (120-1) from the message (UE Capability Information) transmitted by the first terminal (120-1). In step S56, the first base station (110-1) can transmit information related to the capabilities of the first terminal (120-1) to the AMF through the message (UE Capability Info Indication). At least some of the AI / ML related information of the first terminal (120-1) may be included in the message (UE Capability Info Indication). The AMF can verify the AI / ML related information of the first terminal (120-1) from the message (UE Capability Info Indication). As described above, AI / ML related information of the first terminal (120-1) is included in messages (UE Capability Information, UE Capability Info Indication) containing capability information of the first terminal (120-1), so that AI / ML related information of the first terminal (120-1) can be shared with a network including an AMF.

[0103] For convenience, the first terminal (120-1) is referred to as the terminal, and the first base station (110-1) is referred to as the source base station. Additionally, the second base station (120-2) and the third base station (120-3) are referred to as the target base station or candidate target base station.

[0104] FIG. 6 is a flowchart illustrating the process of performing a handover procedure according to an exemplary embodiment. FIG. 6 assumes a case where the first base station (110-1) is determined to be the source base station and the second base station (110-2) is determined to be the target base station.

[0105] Before the handover procedure shown in FIG. 6 is initiated, AI / ML-related information of the terminal (120-1) can be transmitted to the AMF as shown in FIG. 5.

[0106] Referring to FIG. 6, at step S100, mobility control information by AMF can be provided to base stations (110-1, 110-2). The source base station (110-1) can obtain at least some of the AI / ML-related information of the target base station (110-2) in advance through the Xn interface. For example, the source base station (110-1) can obtain information through the Xn interface regarding whether the target base station (110-2) supports AI / ML functions, the ID of the AI / ML model supported by the target base station (110-2), the function or function ID of the AI / ML model supported by the target base station (110-2), the Associated ID of the AI / ML model supported by the target base station (110-2), the validity period of the AI / ML model supported by the target base station (110-2), whether the AI / ML model supported by the target base station (110-2) is a cell-specific model or a generalized model, the granularity or quantization level of the data collection resolution supported by the target base station (110-2), and information regarding the configuration or weight of the learned AI / ML model.

[0107] The source base station (110-1) can determine the target base station (110-2) to hand over the mobile communication terminal from a group of target base station candidates based on the AI / ML-related information. For example, the channel condition of the third base station (110-3) may be relatively better than that of the second base station (110-2) of FIG. 1. Nevertheless, if the third base station (110-3) does not support AI / ML services and the second base station (110-2) supports AI / ML services, and the performance improvement resulting from the terminal (120-1) using AI / ML services is significant, the source base station (110-1) can determine the second base station (110-2) as the target base station.

[0108] A handover event may occur in step S101. The handover event may occur due to the movement of the terminal (120-1), etc.

[0109] In step S102, the terminal (120-1) may transmit Measurement Control and Reports to the source base station (110-1). The Measurement Control and Reports in step S102 may include AI / ML related information of the terminal (120-1). For example, the process shown in FIG. 5 may not have been performed before the handover, or some of the AI / ML related information of the terminal (120-1) may not have been transmitted to the source base station (110-1) in step S54 of FIG. 5. Alternatively, there may be an update to the AI / ML related information of the terminal (120-1). For such cases, the Measurement Control and Reports may include AI / ML related information of the terminal (120-1). However, if the source base station (110-1) has already obtained the AI / ML related information of the terminal (120-1), at least some of the AI / ML related information of the terminal (120-1) may be omitted from the Measurement Control and Reports.

[0110] AI / ML related information of the terminal (120-1) includes whether the terminal (120-1) supports AI / ML functions, the type of AI / ML function supported by the terminal (120-1), the ID of the AI / ML model supported by the terminal (120-1), the Associated ID of the AI / ML being used by the terminal at the source base station (110-1), the ID of the AI / ML model being used by the terminal (120-1) at the source base station (110-1), the validity period of the AI / ML model being used by the terminal (120-1) at the source base station (110-1), whether the AI / ML model being used by the terminal (120-1) at the source base station (110-1) is a cell-specific model or a generalized model, the data collection resolution (granularity or quantization level) supported by the terminal (120-1), information related to the configuration or weights of the learned AI / ML model, a device (CPU, NPU, GPU, etc.) that can be assigned for the AI / ML model, and a device that can be assigned for the AI / ML model. It may include at least one of the computing performance categories of the terminal (120-1).

[0111] In step S104, the source base station (110-1) may determine a handover based on the measurement control report of the terminal (120-1). The source base station (110-1) may determine a handover based on at least one of the downlink channel measurement result of the terminal (120-1), the uplink channel measurement result based on the SRS transmitted by the terminal, the SBFD related information of the terminal (120-1), the SBFD related information of neighboring base stations (110-2, 110-3), the traffic load of the source base station (110-1), and the traffic load of neighboring base stations (110-2, 110-3).

[0112] In the handover decision process, the source base station (110-1) can determine at least one of the neighboring base stations (110-2, 110-3) as the target base station based on AI / ML related information of the terminal (120-1) and AI / ML related information of the neighboring base stations (110-2, 110-3).

[0113] For example, if the terminal (120-1) does not support AI / ML, the source base station (110-1) can determine the target base station in the same way as in a legacy system.

[0114] If the terminal (120-1) supports AI / ML, the source base station (110-1) can adjust the priority of the base station that supports AI / ML among the neighboring base stations (110-2, 110-3).

[0115] For example, the source base station (110-1) may adjust the priority so that a base station among neighboring base stations (110-2, 110-3) that supports at least one of the AI / ML functions, AI / ML associated IDs, and AI / ML models supported by the terminal (120-1) is prioritized and determined as the target base station. If there is no significant difference in other handover determination conditions, the source base station (110-1) may determine a base station among neighboring base stations (110-2, 110-3) that supports at least one of the AI / ML functions, AI / ML associated IDs, and AI / ML models like the terminal (120-1) as the target base station.

[0116] In the above description, at least one of the AI / ML functions, AI / ML associated IDs, and AI / ML models supported by the terminal (120-1) may be determined by considering the resources available to the terminal (120-1) for AI / ML (devices that can be allocated for AI / ML models, number of devices). The AI / ML models, AI / ML associated IDs, AI / ML functions, etc. that the terminal (120-1) can support may be limited according to predefined minimum requirements per AI / ML model, number of parameters, complexity of the AI / ML model, etc.

[0117] In the process of determining the target base station, the source base station (110-1) may prioritize a base station that supports the maximum data collection resolution (granularity or quantization level) supported by the terminal (120-1) and determine it as the target base station.

[0118] In the above description, an example was described in which the source base station (110-1) considers AI / ML-related information of the terminal (120-1) and AI / ML-related information of surrounding base stations (110-2, 110-3) when determining the target base station. The source base station (110-1) may determine the target base station by considering various factors in addition to AI / ML-related information. For example, even if the terminal (120-1) supports AI / ML, the source base station (110-1) may determine the target base station as a base station that does not support AI / ML by considering other factors. For example, if a delay in the handover of the terminal (120-1) is allowed by considering the service quality required for the terminal (120-1) and the traffic load for the terminal (120-1), the source base station (110-1) may determine the target base station as a base station that does not support AI / ML.

[0119] In the handover determination process, the source base station (110-1) can determine the handover margin by considering whether the terminal (120-1) supports AI / ML, whether the source base station (110-1) supports AI / ML, and whether the target base station (110-2) supports AI / ML. The method for determining the handover margin will be explained in more detail with reference to FIGS. 12 to 14, which will be described later.

[0120] In step S106, the source base station (110-1) may send a handover request to the target base station (110-2). The handover request may include the cell ID of the target base station (110-2), the C-RNTI of the terminal (120-1), the RRM configuration, and a rule that maps the current QoS flow to the DRB.

[0121] In the current 5G standard specifications, HandoverPreparationInformationIE is used for handover preparation. HandoverPreparationInformationIE is an information element used for handover preparation, and the handover request in step S106 may include the information element (HandoverPreparationInformationIE) used for handover preparation. The information element used for handover preparation may include AI / ML related information of the terminal (120-1).

[0122] HandoverPreparationInformationIE is an IE transmitted from the source base station (110-1) to the target base station (110-2) during the handover preparation or terminal context retrieval (UE context retrieval) process. For example, HandoverPreparationInformationIE may be transmitted from the source base station (110-1) to the target base station (110-2) in step S106 of FIG. 6. However, the embodiment is not limited thereto. HandoverPreparationInformationIE may also be transmitted from the source base station (110-1) to the target base station (110-2) during other handover preparation procedures or terminal context retrieval processes.

[0123] The source base station (110-1) can include AI / ML related information of the terminal (120-1) in the HandoverPreparationInformationIE. For example, the source base station (110-1) can transmit the AI / ML related information of the terminal (120-1) to the target base station (110-2) through the UE-NR-CapabilityIE corresponding to the ue-CapabilityRAT-List included in the HandoverPreparationInformationIE. The source base station (110-1) can add the AI / ML related information of the terminal (120-1) to the featureSets or featureCombinations of the UE-NR-Capability. As another example, the source base station (110-1) can create a separate IE within the HandoverPreparationInformationIE or create a separate IE within the UE-NR-Capability and transmit the AI / ML related information of the terminal (120-1) to the target base station (110-2) through the created IE.

[0124] The HandoverPreparationInformationIE included in the handover request may include Terminal Assistance Information (UEAssistanceInformation). Terminal Assistance Information may be used to inform the network of the terminal's assistance information. The source base station (110-1) may include the terminal's (120-1) AI / ML assistance information in the terminal assistance information. The terminal's (120-1) AI / ML assistance information may include at least one of the following: the current state of the terminal (120-1) regarding the implementation of AI / ML functions, power saving conditions based on the terminal's (120-1) request, and information regarding the options preferred by the terminal (120-1) regarding the implementation of AI / ML functions. Here, examples of options preferred by the terminal (120-1) regarding the implementation of AI / ML functions may include whether the terminal (120-1) prefers low-complexity AI / ML functions due to low battery, information regarding the number of beams used or beam configuration preferred by the terminal (120-1), but the embodiments are not limited thereto.

[0125] UEAssistanceInformation can be included in the as-Context IE or as-Context message of HandoverPreparationInformationIE.

[0126] As described above, the terminal assistance information (UEAssistanceInformation) may include AI / ML assistance information of the terminal (120-1). The target base station (110-2) can verify the AI / ML assistance information of the terminal (120-1) from the terminal assistance information. Based on the AI / ML assistance information of the terminal (120-1), the target base station (110-2) can identify the constraints of the terminal (120-1) regarding the implementation of AI / ML functions or the options preferred by the terminal (120-1). Through this, the operation of AI / ML functions between the target base station (110-2) and the terminal (120-1) can be efficiently carried out without a separate signaling procedure.

[0127] AI / ML related information of the terminal (120-1) may be transmitted through the terminal assistance information (UEAssistanceInformation) included in the HandoverPreparationInformationIE. To this end, AI / ML related information of the terminal (120-1) may be added to the terminal assistance information (UEAssistanceInformation) included in the as-Context IE of the HandoverPreparationInformationIE.

[0128] In step S108, the target base station (110-2) can perform admission control for a handover request. The target base station (110-2) can decide whether to accept the handover request. The target base station (110-2) can perform admission control by determining whether the terminal (120-1) can be accepted based on its available resources, traffic load, etc.

[0129] In step S110, the target base station (110-2) can transmit a handover request acknowledgment. The source base station (110-1) can receive the handover request acknowledgment. The handover request acknowledgment may include a new C-RNTI of the terminal (120-1) used in the cell of the target base station (110-2), the cell ID of the target base station (110-2), security algorithm identifiers of the target base station (110-2), a dedicated RACH resource set of the target base station (110-2), connections between RACH resources, terminal-specific CSI-RS configuration information, common RACH resources, system information of the target base station (110-2), etc.

[0130] According to an exemplary embodiment, the handover request response may further include AI / ML-related information of the target base station (110-2). For example, if there are no updates to the AI / ML-related information of the target base station (110-2) shared in step S100, the AI / ML-related information of the target base station (110-2) may be omitted from the handover request response.

[0131] In step S112, the source base station (110-1) can trigger a handover by transmitting an RRC Reconfiguration message to the terminal (120-1). The RRC Reconfiguration message may include the cell ID of the target base station (110-2), a new C-RNTI, security algorithm identifiers of the target base station (110-2), a dedicated set of RACH resources of the target base station (110-2), connections between RACH resources, terminal-specific CSI-RS configuration information, common RACH resources, system information of the target base station (110-2), etc.

[0132] The RRC Reconfiguration message may further include AI / ML-related information of the target base station (110-2). The RRC Reconfiguration message may include information on at least one of the following: an ID of an AI / ML model supported by the target base station (110-2), an AI / ML association ID supported by the target base station (110-2), a type of AI / ML function supported by the target base station (110-2), information on whether the AI / ML model supported by the target base station (110-2) is a cell-specific model or a generalized model, a data collection resolution (granularity or quantization level) for training the AI / ML model of the target base station (110-2), a configuration or weight of a trained AI / ML model supported by the target base station (110-2), a device (CPU, NPU, GPU, etc.) that can be assigned for the AI / ML model, and the number of devices that the target base station (110-2) can assign for the AI / ML model. The terminal (120-1) can check AI / ML related information of the target base station (110-2) from the RRC reset (RRCRecofniguration) message. Based on the AI / ML related information of the target base station (110-2), the terminal (120-1) can utilize AI / ML in future communication with the target base station (110-2).

[0133] When a handover is triggered, the terminal (120-1) can perform a random access procedure to the target base station (110-2) based on an RRC reset message.

[0134] In step S114, the source base station (110-1) can transmit an SN status transfer message to the target base station (110-2). In the case of a DAPS handover, the source base station (110-1) may transmit an early status transfer message instead of an SN status transfer message.

[0135] In step S116, the target base station (110-2) can buffer user data received from the source base station (110-1).

[0136] In step S118, the terminal (120-1) can complete the RRC handover procedure by synchronizing with the target base station (110-2) and sending an RRCReconfigurationComplete message to the target base station (110-2). In the case of a DAPS handover, the terminal (120-1) can maintain a connection with the source base station (110-1) even after receiving the RRCReconfiguration message in step S112. In the case of a DAPS handover, the terminal (120-1) can disconnect from the source base station (110-1) after receiving an explicit release instruction from the target base station (110-2). After step S118, further procedures such as path switching of user data may be performed.

[0137] In step S118, the terminal (120-1) can perform a random access procedure for the target base station (110-2). During this process, the terminal (120-1) can transmit a random access message to the target base station (110-2). The terminal (120-1) can receive a random access response message from the target base station (110-2). The terminal (120-1) can transmit a random access completion message to the target base station (110-2).

[0138] As described above, in the handover procedure, AI / ML-related information of the terminal (120-1) is transmitted to the target base station (110-2) through the message used in the existing handover procedure, and AI / ML-related information of the target base station (110-2) can be transmitted to the terminal (120-1). Through this, information for using AI / ML can be shared between the terminal (120-1) and the target base station (110-2) while minimizing the increase in signaling overhead.

[0139] FIG. 7 is a conceptual diagram exemplarily illustrating the method by which a source base station (110-1) selects a target base station in step S104 of FIG. 6.

[0140] In FIG. 7, it can be assumed that the AI / ML model is a terminal-only model. Additionally, it can be assumed that the source base station (110-1) has obtained AI / ML-related information of neighboring base stations (110-2, 110-3) in advance through the Xn interface or AMF. The source base station (110-1) can obtain information regarding the AI / ML association ID of each of the neighboring base stations (110-2, 110-3). The source base station (110-1) can determine the target base station by considering the association ID supported by each of the neighboring base stations (110-2, 110-3) and the association ID used by the terminal (120-1) within the coverage of the source base station (110-1).

[0141] Referring to FIG. 7, the terminal (120-1) is within the coverage of the source base station (110-1). A terminal-only model may be operated using the set of associated IDs. If there is no significant difference in other handover-related conditions, the source base station (110-1) among neighboring base stations (110-2, 110-3) A base station capable of utilizing the associated ID set can be preferentially selected as the target base station. If neighboring base stations (110-2, 110-3) If there is no base station that can utilize all of the associated ID sets, the base station that can utilize the most associated IDs among the associated IDs included in the associated ID set can be selected as the target base station first.

[0142] FIG. 8 is a conceptual diagram exemplarily illustrating the method by which a source base station (110-1) selects a target base station in step S104 of FIG. 6.

[0143] In FIG. 7, it can be assumed that the AI / ML model is a joint model. Referring to FIG. 8, the terminal (120-1) is within the coverage of the source base station (110-1). A joint model may be operated using the set of associated IDs. If there is no significant difference in other handover-related conditions, the source base station (110-1) among neighboring base stations (110-2, 110-3) A base station capable of utilizing the associated ID set can be preferentially selected as the target base station. If neighboring base stations (110-2, 110-3) If there is no base station that can utilize all of the associated ID sets, the base station that can utilize the most associated IDs among the associated IDs included in the associated ID set can be selected as the target base station first.

[0144] FIG. 9 is a flowchart illustrating the process of performing a conditional handover (CHO) according to an exemplary embodiment. In describing the embodiment of FIG. 9, descriptions of parts that overlap with or are similar to FIG. 6 are omitted. For example, the description of steps S200, S201, and S204 of FIG. 9 is replaced with the description of steps S100, S101, and S104 of FIG. 6.

[0145] Referring to FIG. 9, in step S202, the terminal (120-1) can transmit a measurement control report to the source base station (110-1). The measurement control report may include AI / ML related information of the terminal (120-1).

[0146] In steps S206 and S207, the source base station (110-1) can send a handover request to at least one candidate target base station (110-2, 110-3).

[0147] The method by which the source base station (110-1) determines at least one candidate target base station may be similar to the method by which the source base station (110-1) determines the target base station described with reference to FIG. 6.

[0148] According to an exemplary embodiment, the handover request may further include AI / ML-related information of the terminal (110-2). The AI / ML-related information of the terminal (110-2) may be included in the handover request as part of the terminal capability information. As described above, the source base station (110-1) may transmit the AI / ML-related information of the terminal (120-1) to the target base station (110-2) through the UE-NR-CapabilityIE corresponding to the UE-CapabilityRAT-List included in the HandoverPreparationInformationIE. As another example, the source base station (110-1) may create a separate IE within the HandoverPreparationInformationIE or create a separate IE within the UE-NR-CapabilityIE and transmit the AI / ML-related information of the terminal (120-1) to the target base station (110-2) through the created IE.

[0149] In steps S208 and S209, at least one of the candidate target base stations (110-2, 110-3) can perform admission control for a handover request.

[0150] In steps S210 and S211, the source base station (110-1) may receive a handover request response from at least one candidate target base station (110-2, 110-3). According to an exemplary embodiment, the handover request response may further include AI / ML-related information from at least one candidate target base station (110-2, 110-3).

[0151] In step S212, the source base station (110-1) can trigger a handover by sending an RRC Reconfiguration message to the terminal (120-1). The RRC Reconfiguration message may further include AI / ML-related information of at least one candidate target base station (110-2, 110-3).

[0152] In step S214, the terminal (120-1) can transmit an RRC reset complete message to the source base station (110-1).

[0153] In step S216, the terminal (120-1) can evaluate whether a predetermined CHO condition is satisfied. If the CHO condition is satisfied, in step S218, the terminal (120-1) can disconnect from the source base station (110-1) and synchronize with the target base station (110-2). The terminal (120-1) can perform random access on the target base station (110-2) and, in step S220, send an RRC reset completion message to the target base station (110-2) to complete the RRC handover procedure.

[0154] Figure 10 is a diagram showing a variation of the handover procedure shown in Figure 6.

[0155] In describing the embodiment of FIG. 10, the description of steps that overlap with FIG. 6 is omitted.

[0156] Referring to FIG. 10, the associated ID of the AI / ML being used by the terminal (120-1) (or at least one of the model ID and function ID) may not match the associated ID of the AI / ML supported by the target base station (110-2) (or at least one of the model ID and function ID). In this case, the source base station (110-1) may acquire the AI / ML related information of the target base station (110-2) and then transmit the AI / ML related information of the target base station (110-2) to the terminal (120-1) in step S312. The AI / ML related information of the target base station (110-2) may include a dataset corresponding to the associated ID of the AI / ML supported by the base station (110-2). Here, the dataset corresponding to the associated ID may include at least one of a training dataset corresponding to the associated ID and configuration information of an AI / ML model corresponding to the associated ID.

[0157] To explain in more detail, at step S300, the source base station (110-1) can obtain AI / ML-related information from each of the neighboring base stations (110-2, 110-3) via the Xn interface or AMF. If necessary, the source base station (110-1) can obtain a data set corresponding to the associated ID supported by the target base station (110-2) from the target base station (110-2). As another example, the source base station (110-1) can obtain a data set corresponding to the associated ID supported by the target base station (110-2) from the AMF.

[0158] In step S304, the source base station (110-1) can determine the handover. The source base station (304) can determine the target base station (110-2) among the neighboring base stations (110-2, 110-3). For example, the relevant AI / ML information of the neighboring base stations (110-2, 110-3) transmitted in step S300 may not include the specific association ID (or at least one of the AI / ML model ID and AI / ML function ID) supported by each of the neighboring base stations (110-2, 110-3). In this case, the source base station (110-1) can determine the target base station (110-2) by considering only whether each of the neighboring base stations (110-2, 110-3) supports AI / ML. If the terminal (120-1) supports AI / ML and there is no significant difference in other conditions between neighboring base stations (110-2, 110-3), the source base station (110-1) can determine the base station that supports AI / ML as the target base station. This allows the benefit effect from using AI / ML to be maximized.

[0159] In the relevant AI / ML information of neighboring base stations (110-2, 110-3) transmitted in step S300, specific association IDs (or at least one of an AI / ML model ID and an AI / ML function ID) supported by each of the neighboring base stations (110-2, 110-3) may be included. And, the association IDs (or at least one of an AI / ML model ID and an AI / ML function ID) supported by the neighboring base stations (110-2, 110-3) may be different from the association ID (or at least one of an AI / ML model ID and an AI / ML function ID) used by the terminal (120-1). Even if the associated ID used by the terminal (120-1) and the associated ID used by neighboring base stations (110-2, 110-3) are different, the source base station (110-1) can determine the base station among the neighboring base stations (110-2, 110-3) that supports AI / ML as the target base station by considering the possibility that the data set will be shared through step S312 as described below.

[0160] In step S306, the target base station (110-2) can verify AI / ML related information of the terminal (120-1) included in the handover request message. From the AI / ML related information of the terminal (120-1), the target base station (110-2) can determine whether the associated ID (or at least one of the function ID and AI / ML model ID) supported by the target base station (110-2) matches the associated ID (or at least one of the function ID and AI / ML model ID) used by the terminal (120-1).

[0161] In step S310, the target base station (110-2) may transmit a handover request response message to the source base station (110-1). If the associated ID supported by the target base station (110-2) does not match the associated ID used by the terminal (120-1), information indicating the discrepancy may be included in the handover request response message. The source base station (110-1) may decide to initiate step S312 based on the information indicating the discrepancy. As another example, the source base station (110-1) may compare the AI / ML related information of the terminal (120-1) obtained independently with the AI / ML related information of the target base station (110-2). If it is determined that the associated ID used by the terminal (120-1) and the associated ID used by the target base station (110-2) are different, the source base station (110-1) may decide to initiate step S312.

[0162] In step S312, the source base station (110-1) may transmit a data set corresponding to an association ID supported by the target base station (110-2) to the terminal (120-1). The data set corresponding to the association ID may include at least one of training data corresponding to the association ID and configuration information of an AI / ML model corresponding to the association ID. When the source base station (110-1) transmits the data set corresponding to the association ID to the terminal (120-1) through step S312, the terminal (120-1) may determine that there is an implicit instruction to use the AI / ML model corresponding to the association ID in communication with the target base station (110-1).

[0163] The terminal (120-1) can perform data configuration corresponding to the associated ID based on the associated ID of the target base station (110-2) obtained from the source base station (110-1). The source base station (110-1) can change the data collection required by the terminal (120-1) for AI / ML model training, updating, monitoring, inference, etc. into a configuration agreed upon by the associated ID and transmit it to the terminal (120-1).

[0164] As described above, the source base station (110-1) can transmit a data set corresponding to an associated ID supported by the target base station (110-2) to the terminal (120-1) through step S312. Accordingly, the terminal (120-1) can prepare in advance for using an AI / ML model by using the data set.

[0165] That is, even if the associated ID previously used by the terminal (120-1) and the associated ID supported by the target base station (110-2) are different, the terminal (120-1) can use the associated ID supported by the target base station (110-2) after the handover by using the data set obtained in step S312.

[0166] Figure 11 is a diagram showing a variation of the handover procedure shown in Figure 6.

[0167] In describing the embodiment of FIG. 11, the description of parts that overlap with FIG. 6 and FIG. 10 is omitted.

[0168] Referring to FIG. 11, the associated ID of the AI / ML being used by the terminal (120-1) (or at least one of the model ID and function ID) may not match the associated ID of the AI / ML being supported by the target base station (110-2) (or at least one of the model ID and function ID).

[0169] The target base station (110-2) can verify the AI / ML-related information of the terminal based on the handover request message of step S406. The target base station (110-2) can verify that the associated ID of the AI / ML being used by the terminal (120-1) (or at least one of the model ID and function ID) does not match the associated ID of the AI / ML supported by the target base station (110-2) (or at least one of the model ID and function ID). The source base station (110-1) may not possess a dataset corresponding to the associated ID supported by the target base station (110-2). In this case, the target base station (110-2) can transmit the dataset corresponding to the associated ID supported by the target base station (110-2) to the source base station (110-1) through step S412. Here, the dataset may include at least one of the configuration information of the AI / ML model corresponding to the associated ID and a training dataset.

[0170] The source base station (110-1) can obtain a data set corresponding to an associated ID supported by the target base station (110-2) in step S412. The source base station (110-1) can transmit the data set corresponding to the associated ID supported by the target base station (110-2) to the terminal (120-1) in step S414. Through this, the terminal (120-1) can obtain the data set corresponding to the associated ID supported by the target base station (110-2) in advance before the handover procedure is completed.

[0171] In step S104, which was previously described with reference to FIG. 6, the source base station (110-1) may consider a handover margin when determining the handover. The source base station (110-1) may adjust the handover margin based on at least one of the AI / ML related information of the terminal (120-1), the AI / ML related information of the source base station (110-1), and the AI / ML related information of the target base station (110-2).

[0172] FIG. 12 is a conceptual diagram exemplarily illustrating a source base station (110-1) adjusting a handover margin according to an exemplary embodiment. In FIGS. 12 to 14 below, the L1 graph represents the signal strength of a terminal (120-1) received by the source base station (110-1), and the L2 graph represents the signal strength of a terminal (120-1) received by the target base station (110-2). Additionally, in FIGS. 12 to 14, the horizontal axis represents time, and the vertical axis represents signal strength.

[0173] Referring to FIG. 12, the source base station (110-1) may support AI / ML. Additionally, the terminal (120-1) may use AI / ML in communication with the source base station (110-1). If the target base station (110-2) does not support AI / ML, the terminal (120-1) may not be able to use AI / ML after the handover and thus may not be able to enjoy the resulting improvement in communication performance. Considering this point, the source base station (110-1) may set the handover margin (m2) to be larger than the handover margin (m1) in the legacy system. For example, the source base station (110-1) may set the handover margin (m2) to a gain correction value (G) for the use of AI / ML larger than the handover margin (m1) in the legacy system. AI It can be set to be larger by ) This allows the terminal (120-1) to use AI / ML at the source base station (110-1) for a longer period of time, thereby maximizing the performance optimization effect resulting from the use of AI / ML.

[0174] FIG. 13 is a conceptual diagram exemplarily illustrating a source base station (110-1) adjusting a handover margin according to an exemplary embodiment.

[0175] Referring to FIG. 13, the source base station (110-1) may not support AI / ML, while the target base station (110-2) may support AI / ML. Additionally, it can be assumed that the terminal (120-1) can use the AI / ML supported by the target base station (110-2). In this case, if the handover of the terminal (120-1) is faster, the terminal (120-1) can utilize AI / ML earlier to enjoy performance optimization effects. Therefore, the source base station (110-1) may set the handover margin (m4) to be smaller than the handover margin (m3) of the legacy system. For example, the source base station (110-1) may set the handover margin (m4) to a gain correction value (G) for the use of AI / ML smaller than the handover margin (m3) in the legacy system. AI It can be set to be smaller by ) and the terminal (120-1) can be induced to use AI / ML at the target base station (110-2) to maximize the performance optimization effect resulting from the use of AI / ML.

[0176] FIG. 14 is a conceptual diagram exemplarily illustrating a source base station (110-1) adjusting a handover margin according to an exemplary embodiment.

[0177] Referring to FIG. 14, the source base station (110-1) may support AI / ML, and the target base station (110-2) may also support AI / ML. The associated ID (A) of the AI / ML supported by the source base station (110-1) and the associated ID (B) of the AI / ML supported by the target base station (110-2) may be different. In this case, the source base station (110-1) may set the handover margin (m2) to be larger than the handover margin (m1) in the legacy system to maintain the quality of service of the terminal (120-1). For example, the source base station (110-1) may set the handover margin (m2) to a gain correction value (G) for the use of AI / ML than the handover margin (m1) in the legacy system. AI It can be set larger by )

[0178] Figure 15 is a flowchart showing the handover procedure according to the comparative example.

[0179] In the embodiment of FIG. 15, the AI / ML related information of the terminal (120-1) may not be included in the handover request of step S606. Additionally, the AI / ML related information of the target base station (110-2) may not be included in the handover request response of step S610. Furthermore, the AI / ML related information of the target base station (110-2) may not be included in the RRC reset message used in step S612. Therefore, there may be no exchange of AI / ML related information between the terminal (120-1) and the target base station (110-2) while the handover procedure is in progress.

[0180] Therefore, after the handover is completed, a separate signaling procedure is required for the terminal (120-1) to use the AI / ML supported by the target base station (120-1). For example, through step S620, the terminal (120-1) can transmit its AI / ML-related information to the target base station (110-2). The target base station (110-2) can determine at least one of an AI / ML model, an AI / ML function, and an associated ID that can issue an activation instruction to the terminal (120-1) through the AI / ML-related information of the terminal (120-1). Additionally, the target base station (110-2) can transmit information necessary for the terminal (120-1) to operate the AI / ML model.

[0181] FIG. 16 is a flowchart illustrating the process of performing step S620 of FIG. 15.

[0182] Referring to FIG. 16, in step S120-1, the target base station (110-2) can query the terminal (120-1) for AI / ML capabilities that the terminal can support via a UECapabilityEnquiry message. When the terminal (120-1) receives the message, in step S120-2, it transmits UECapabilityInformation to the target base station (110-2), which includes the existing UECapability along with the AI / ML capabilities it can support. The target base station (110-2) receives the UECapabilityInformation transmitted by the terminal (120-1) and, in step S120-3, transmits it via an RRCReconfiguration message containing additional network conditions (or corresponding associated IDs) for the AI / ML capabilities that the terminal (120-1) can support. When the terminal (120-1) receives the corresponding RRCReconfiguration message, it determines available AI / ML functions by comparing its own capabilities, its own additional conditions (or corresponding associated IDs), and the network additional conditions (or corresponding associated IDs) transmitted by the base station. Then, in step S120-4, the terminal (120-1) reports its available AI / ML functions to the base station via UE Assistance Information. After receiving the message, the target base station (110-2) transmits the configuration that the terminal (120-1) will use for inference in step S120-5.

[0183] As described above, more than five additional signaling steps may be required to exchange AI / ML-related information after the handover. Receiving the RRC reset messages in steps S620-3 and S620-5 of Fig. 16 may incur significant overhead.

[0184] Similar to the comparative example in Fig. 15, a procedure similar to step S620 may be added in the comparative example related to the conditional handover procedure.

[0185] On the other hand, according to embodiments with reference to FIGS. 6, 9, 10, and 11, AI / ML-related information of the terminal (120-1) and AI / ML-related information of the target base station (110-2) can be shared in advance through a handover request, a handover request response, an RRC reset message, etc. Therefore, the terminal (120-1) can optimize the service by utilizing the activation / deactivation / inference / monitoring functions of AI / ML supported by the target base station (110-2) after the handover without performing the S620 step shown in FIGS. 15 and 16. Through this, signaling overhead for AI / ML implementation after the handover can be reduced.

[0186] Figure 17 is a flowchart illustrating a monitoring procedure for reusing an existing trained AI / ML model after the handover procedure is completed.

[0187] Referring to FIG. 17, since the handover is completed, the terminal (120-1) can receive services from the target base station (110-2). The terminal (120-1) can communicate with the target base station (110-2). The terminal (120-1) can reuse the AI / ML model used at the source base station (110-1) after the handover.

[0188] During the handover process, the terminal (120-1) can obtain AI / ML-related information of the target base station (110-2) by receiving an RRC reset message. If the target base station (110-2) supports the AI / ML model that the terminal (120-1) is using at the source base station (110-1), the terminal (120-1) can maintain the previously used AI / ML model without discarding it even after the handover is completed. For example, if the associated ID of the AI / ML used by the terminal (120-1) at the source base station (110-1) is the same as the associated ID of the AI / ML supported by the target base station (110-2), the terminal (120-1) can reuse the existing AI / ML model. If the associated ID of the AI / ML used by the terminal (120-1) at the source base station (110-1) is the same as the associated ID of the AI / ML supported by the target base station (110-2), the terminal (120-1) can reuse the existing AI / ML model even if the model ID of the AI / ML used by the terminal (120-1) and the model ID of the AI / ML supported by the target base station (110-2) are different.

[0189] In step S712 of FIG. 17, the target base station (110-2) may instruct the terminal (120-1) to retain the AI / ML model that was used at the source base station (110-1). However, the embodiment is not limited thereto. The terminal (120-1) may decide on its own to retain the AI / ML model that was used at the source base station (110-1).

[0190] To verify the operation of the reused AI / ML model, the target base station (110-2) can monitor the performance of the AI / ML model. To do this, the target base station (110-2) can command the terminal (120-1) to report the inference value of the AI / ML model. In step S714, the terminal (120-1) can report the inference value of the reused AI / ML model.

[0191] The target base station (110-2) can monitor the performance of the reused AI / ML model based on the inference value report of the reused AI / ML model obtained in step S714. In step S718, the target base station (110-2) can transmit instructions for managing the reused AI / ML model to the terminal (120-1) based on the monitoring results. For example, if the performance of the reused AI / ML model is sufficiently good as a result of monitoring, the management instructions may include instructions to maintain the application of the reused AI / ML model. The management instructions may include instructions for the terminal (120-1) to monitor and / or retrain the reused AI / ML model based on data measured from the target base station (110-2) according to the AI / ML full-cycle process of FIG. 4. The management instructions may include the monitored AI / ML model ID, association ID, function or function ID, and monitoring results. If it is decided to maintain the AI / ML model, the management instructions may include the validity period of the AI / ML model. When it is decided to discard an AI / ML model, the management instruction may include at least one piece of information among a new AI / ML model ID, an association ID, a function or function ID, and a resource allocation for training. In step S719, the terminal (120-1) may perform management on the reused AI / ML model according to the management instruction.

[0192] FIG. 18 is a flowchart illustrating a monitoring procedure for reusing a previously trained AI / ML model after the handover procedure is completed. In describing the embodiment of FIG. 18, content that overlaps with FIG. 17 is omitted.

[0193] Referring to FIG. 18, monitoring of the AI / ML model can be performed by the terminal (120-1). Therefore, unlike FIG. 17, the procedure for the terminal (120-1) to report the AI / ML model inference value can be omitted. To verify the operation of the reused AI / ML model, the terminal (120-1) can monitor the performance of the AI / ML model on its own.

[0194] In step S726, the terminal (120-1) may report the AI / ML model monitoring results to the target base station (110-2), and the target base station (110-2) may accept or reject the monitoring results of the terminal (120-1) in consideration of available resources and performance improvement. In step S728, the target base station (110-2) may transmit a management instruction to the terminal (120-1). The management instruction may include a decision on whether the target base station (110-2) accepts the monitoring results. In step S728, the terminal (120-1) may perform management on the reused AI / ML model according to the management instruction.

[0195] With reference to FIGS. 17 and 18, the explanation assumes that the AI / ML association ID previously used by the terminal (120-1) is the same as the AI / ML association ID supported by the target base station (110-2). Even if the AI / ML association ID previously used by the terminal (120-1) is different from the AI / ML association ID supported by the target base station (110-2), the terminal (120-1) can reuse the existing AI / ML model. In this case, the terminal (120-1) can search for the most suitable association ID among the association IDs supported by the target base station (110-2).

[0196] FIG. 19 is a flowchart illustrating a monitoring procedure for reusing a previously trained AI / ML model after the handover procedure is completed. The associated ID of the AI / ML used by the terminal (120-1) at the source base station (110-1) and the associated ID of the AI / ML supported by the target base station (110-2) may be different.

[0197] In step S810, the target base station (110-2) may instruct the terminal (120-1) to maintain the AI / ML model that was used at the source base station (110-1). However, the embodiments are not limited thereto. The terminal (120-1) may independently decide to maintain the AI / ML model that was used at the source base station (110-1).

[0198] In step S820, the terminal (120-1) and the target base station (110-2) can search for an associated ID suitable for the AI / ML model of the terminal (120-1) among the associated IDs supported by the target base station (110-2). The target base station (110-2) can configure and transmit a reference signal to the terminal (120-1) with a configuration corresponding to a plurality of associated IDs. The target base station (110-2) can include associated ID information in the reference signal resource configuration and the measurement report resource configuration for the terminal (120-1). The target base station (110-2) can transmit reference signals (e.g., CSI-RS) using configurations corresponding to a plurality of associated IDs. The terminal (120-1) can measure the reference signals transmitted by the target base station (110-2) and utilize the measurement results as input values ​​for the reused AI / ML model. After receiving reference signals corresponding to multiple associated IDs (e.g., associated ID #1 to associated ID #N), the terminal can perform a monitoring result report. In the monitoring result report, the terminal (120-1) can report the associated ID with the best performance of the AI / ML model. As another example, the terminal (120-1) may report multiple associated IDs that guarantee the performance of the AI / ML model higher than a predetermined standard.

[0199] In step S840, the target base station (110-2) can transmit management instructions to the terminal (120-1). In step S850, the terminal (120-1) can perform management on the reused AI / ML model according to the management instructions.

[0200] FIG. 20 is a flowchart illustrating a monitoring procedure for reusing a previously trained AI / ML model after the handover procedure is completed. In describing the embodiment of FIG. 20, descriptions of parts that overlap with or are similar to FIG. 19 are omitted. The associated ID of the AI / ML used by the terminal (120-1) at the source base station (110-1) and the associated ID of the AI / ML supported by the target base station (110-2) may be different.

[0201] In step S920, the terminal (120-1) and the target base station (110-2) can search for an associated ID suitable for the AI / ML model of the terminal (120-1) among the associated IDs supported by the target base station (110-2). Unlike step S820 of FIG. 19, in step S920 of FIG. 19, the target base station (110-2) can discontinuously transmit reference signals corresponding to multiple associated IDs. After receiving reference signals corresponding to multiple associated IDs (e.g., associated ID #1 to associated ID #N), the terminal can perform a monitoring result report. In the monitoring result report, the terminal (120-1) can report the associated ID that has the best performance of the AI / ML model. As another example, the terminal (120-1) may report multiple associated IDs that guarantee the performance of the AI / ML model higher than a predetermined standard.

[0202] According to the embodiments described with reference to FIGS. 17 to 20, the terminal (120-1) can monitor the performance of the recycled AI / ML model while recycling the AI / ML model that has already been trained even after the handover. This allows the time required for training the AI / ML model of the terminal (120-1) to be shortened, and enables the terminal (120-1) to improve mobile communication performance using AI / ML even before completing retraining by collecting training data in a new base station environment.

[0203] With reference to FIGS. 1 to 14 and FIGS. 17 to 20, a method and apparatus for performing a handover procedure in a mobile communication system according to exemplary embodiments have been described above. According to at least one embodiment, AI / ML-related information can be shared between a terminal (120-1) and a target base station (110-2) while minimizing the increase in signaling overhead during the handover procedure. According to at least one embodiment, the source base station (110-1) can increase the benefits from AI / ML utilization by determining the target base station (110-2) by considering at least one of the AI / ML-related information of the terminal (120-1), the AI / ML-related information of the target base station (110-2), and the AI / ML-related information of the source base station (110-1). According to at least one embodiment, the effect of AI / ML utilization can be maximized by adjusting the handover margin by considering the AI / ML-related information during the handover process, either the source base station (110-1) or the terminal (120-1). According to at least one embodiment, in the existing handover procedure, the terminal (120-1) must go through four additional steps, such as reconfiguring the RRC for using the AI / ML model after handover to the target base station (110-2); however, by exchanging AI / ML-related information in advance during the handover process, the configuration of the RRC for utilizing the AI / ML function can be processed in advance. According to at least one embodiment, after handover to the target base station (110-2) that supports the AI / ML function, the terminal (120-1) can monitor the performance of the base station and the AI / ML model by reusing the AI / ML model that was previously used. If the performance of the reused AI / ML model is better than the performance of a legacy mobile communication system that does not use the AI / ML model, the time required for training the AI / ML model can be shortened by continuing to reuse the AI / ML model. In addition, the terminal (120-1) can utilize AI / ML even before building a new AI / ML model through the reuse of the existing AI / ML model.

[0204]

[0205] Based on the foregoing description of various embodiments of this disclosure, a person skilled in the art will clearly understand that the methods and / or processes of the present invention and the steps thereof may be realized in hardware, software, or any combination of hardware and software suitable for a particular use case. The hardware may include a general-purpose computer and / or a dedicated computing device or a specific computing device or a particular form or component of a specific computing device. The processes may be realized by one or more processors having internal and / or external memory, such as a microprocessor, a controller, such as a microcontroller, an embedded microcontroller, a microcomputer, an arithmetic logic unit (ALU), a digital signal processor, such as a programmable digital signal processor, or other programmable device. In addition, or as an alternative, the above processes may be carried out by an application-specific integrated circuit (ASIC), a programmable gate array, such as a field programmable gate array (FPGA), a programmable logic unit (PLU), or a programmable array logic (PAL), or any other device capable of executing and responding to instructions, any other device or combination of devices that may be configured to process electronic signals. The processing unit may execute an operating system (OS) and one or more software applications executed on said operating system. Additionally, the processing unit may access, store, manipulate, process, and generate data in response to the execution of the software.For ease of understanding, the processing unit may be described as being used as a single unit, but a person of ordinary skill in the art will understand that the processing unit may include multiple processing elements and / or multiple types of processing elements. For example, the processing unit may include multiple processors or one processor and one controller. In addition, other processing configurations, such as parallel processors, are also possible.

[0206] Software may include computer programs, code, instructions, or a combination of one or more of these, and may configure a processing unit to operate as desired or command the processing unit independently or collectively. Software and / or data may be permanently or temporarily embodied in any type of machine, component, physical device, virtual equipment, computer storage medium or device, or transmitted signal wave in order to be interpreted by the processing unit or to provide instructions or data to the processing unit. Software may be distributed over networked computer systems and may be stored or executed in a distributed manner. Software and data may be stored on one or more machine-readable recording media.

[0207] Furthermore, parts contributing to the objects of the technical solution of the present invention or to the prior art may be implemented in the form of program instructions that can be executed through various computer components and recorded on a machine-readable medium. The machine-readable medium may include program instructions, data files, data structures, etc., either alone or in combination. The program instructions recorded on the machine-readable recording medium may be those specifically designed and configured for the embodiments, or they may be those known and available to a person skilled in the art of computer software. Examples of machine-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes; optical recording media such as CD-ROMs, DVDs, and Blu-rays; magneto-optical media such as floptical disks; and hardware devices specifically configured to store and execute program instructions, such as ROM, RAM, and flash memory. Examples of program instructions include machine code, bytecode, as well as high-level language code that can be executed by a computer using an interpreter, etc., which can be created using a structured programming language such as C, an object-oriented programming language such as C++, or high-level or low-level programming languages ​​(assembly language, hardware description languages, and database programming languages ​​and technologies), which can be stored and compiled or interpreted to be executed on a machine capable of executing any of the aforementioned devices, as well as a processor, a processor architecture, or a heterogeneous combination of different hardware and software combinations.

[0208] Accordingly, in one embodiment according to the present invention, when the methods and combinations thereof described above are performed by one or more computing devices, the methods and combinations thereof may be implemented as executable code that performs each step. In another embodiment, the methods may be implemented as systems that perform the steps, and the methods may be distributed in various ways across devices, or all functions may be integrated into a single dedicated, standalone device or other hardware. In yet another embodiment, the means for performing the steps associated with the processes described above may include any of the hardware and / or software described above. All such sequential combinations and combinations are intended to fall within the scope of this disclosure.

[0209] For example, the above-described hardware device may be configured to operate as one or more software modules to perform the operation of the embodiment, and vice versa. The hardware device may include a processor such as an MPU, CPU, GPU, or TPU that is combined with memory such as ROM / RAM for storing program instructions and configured to execute instructions stored in said memory, and may include a communication unit capable of exchanging signals with an external device. Additionally, the hardware device may include a keyboard, mouse, or other external input device for receiving instructions written by developers.

[0210] Although the present invention has been described above with specific details such as specific components, limited embodiments, and drawings, this is provided only to aid in a more comprehensive understanding of the invention, and the invention is not limited to the above embodiments, and a person skilled in the art to which the invention belongs can make various modifications and variations from this description.

[0211] Accordingly, the scope of the present invention is not limited to the embodiments described above, and all modifications equivalent to or equivalent to the claims attached to this disclosure, as well as the claims attached to this disclosure, shall be considered to be within the scope of the scope of the concept of the present invention. For example, appropriate results may be achieved even if the described techniques are performed in a different order than the described method, and / or the components of the described system, structure, device, circuit, etc. are combined or assembled in a form different from the described method, or are replaced or substituted by other components or equivalents.

[0212] Such equivalent or equivalent modifications may include, for example, logically equivalent methods capable of producing the same result as carrying out the method according to the present invention; the true meaning and scope of the present invention shall not be limited by the examples described above, but shall be understood in the broadest sense permissible by law.

Claims

1. In a handover method performed by a base station, Step of receiving a measurement control report of a terminal; A step of determining a handover based on the measurement control report of the above terminal; and It includes the step of transmitting a handover request to a target base station, The above handover request is a handover method including AI / ML-related information of the terminal.

2. In Paragraph 1, A handover method wherein the above handover request includes an information element used for handover preparation, and the information element used for handover preparation includes AI / ML-related information of the terminal.

3. In Paragraph 1, The above handover request includes an information element used for handover preparation, and the information element used for handover preparation includes auxiliary information of the terminal, and The auxiliary information of the above terminal includes AI / ML auxiliary information of the above terminal for assisting the implementation of AI / ML functions of the above terminal, and AI / ML auxiliary information of the above terminal includes at least one of the current state of the above terminal related to the implementation of AI / ML functions, power saving according to the request of the terminal, and an option preferred by the above terminal related to the implementation of AI / ML functions.

4. In Paragraph 1, A handover method comprising at least one of the following: AI / ML-related information of the terminal, whether the terminal supports AI / ML functions, the type of AI / ML function supported by the terminal, the ID of the AI / ML model supported by the terminal, the Associated ID of the AI / ML being used by the terminal, the ID of the AI / ML model being used by the terminal, the validity period of the AI / ML model being used by the terminal, whether the AI / ML model being used by the terminal is a cell-specific model or a generalized model, the data collection resolution supported by the terminal, information related to the configuration or weight of the AI / ML model being used by the terminal, a device that the terminal can allocate for the AI / ML model, the number of devices that the terminal can allocate for the AI / ML model, and a computing performance category of the terminal.

5. In Paragraph 1, A handover method further comprising the step of obtaining capability information of the terminal from the terminal, wherein the capability information of the terminal includes AI / ML-related information of the terminal.

6. In Paragraph 4, The method further includes the step of acquiring AI / ML-related information from at least one neighboring base station, A handover method for determining a target base station based on AI / ML related information of the terminal and AI / ML related information of at least one neighboring base station in the step of determining the handover above.

7. In Paragraph 6, A handover method in which, when the above terminal supports AI / ML, the priority of the neighboring base station that supports AI / ML among the at least one neighboring base station is determined to be higher.

8. In Paragraph 1, In the step of determining a handover based on the measurement control report of the above terminal, if the handover margin is satisfied, a handover is determined. A handover method in which the handover margin is determined based on at least one of the AI / ML related information of the terminal, the AI / ML related information of the base station, and the AI / ML related information of the target base station.

9. In Paragraph 8, A handover method in which, when the terminal and the base station support AI / ML and the target base station does not support AI / ML, the handover margin is determined to be higher than the handover margin of the legacy system.

10. In Paragraph 8, A handover method in which, when the terminal and the target base station support AI / ML and the base station does not support AI / ML, the handover margin is determined to be smaller than the handover margin of the legacy system.

11. In Paragraph 1, The method further includes the step of receiving a handover request response from the target base station, The above handover request response includes AI / ML-related information of the target base station, and A handover method comprising at least one of the following: AI / ML related information of the target base station, the ID of the AI / ML model supported by the target base station, the AI / ML association ID supported by the target base station, the type of AI / ML function supported by the target base station, information on whether the AI / ML model supported by the target base station is a cell-specific model or a generalized model, the data collection resolution for learning the AI / ML model of the target base station, the configuration or weight of the learned AI / ML model supported by the target base station, the device (CPU, NPU, GPU, etc.) that the target base station can allocate for the AI / ML model, and the number of devices that the target base station (110-2) can allocate for the AI / ML model.

12. In Paragraph 11, The method includes the step of transmitting a message to the terminal that triggers a handover, A handover method in which a message triggering the handover includes AI / ML-related information of the target base station.

13. In Paragraph 12, The handover method is a message that triggers the above handover, which is an RRC Reconfiguration message.

14. Communications Department; and In a base station including a processor, The processor performs the steps of: receiving a measurement control report of a terminal; determining a handover based on the measurement control report of the terminal; and transmitting a handover request to a target base station. The above handover request is a base station containing AI / ML-related information of the terminal.

15. In a handover method performed by a target base station, A step of receiving a handover request from a source base station; and It includes the step of transmitting a handover request response to the source base station, The above handover request is a handover method that includes AI / ML-related information of a terminal performing the handover.

16. In Paragraph 15, The above handover request response is a handover method including AI / ML-related information of the target base station.

17. In Paragraph 15, A handover method further comprising: a step of receiving a report of inference values ​​of the reused AI / ML model from the terminal when the terminal reuses the AI / ML model used at the source base station after the handover is completed; and a step of monitoring the performance of the reused AI / ML model.

18. In Paragraph 15, A handover method further comprising the step of obtaining performance monitoring results of the reused AI / ML model from the terminal when the terminal reuses the AI / ML model used at the source base station after the handover is completed.

19. In Paragraph 15, After the handover is completed, if the terminal reuses the AI / ML model used at the source base station, but the AI / ML association ID used by the terminal does not match the association ID supported by the target base station, A handover method comprising: transmitting reference signals corresponding to a plurality of associated IDs to the terminal; obtaining a monitoring result from the terminal based on the measurement result of the reference signals corresponding to the plurality of associated IDs; and generating a management instruction for a reused AI / ML model based on the monitoring result and transmitting the management instruction to the terminal.

20. In a handover method performed by a terminal, A step of transmitting a measurement control report to the source base station; A step of receiving a message that triggers a handover from the source base station; and The method includes the step of performing random access to a target base station identified by a message triggering the handover, and A handover method in which a message triggering the handover includes AI / ML-related information of the target base station.

21. In Paragraph 20, A handover method further comprising the step of providing capability information of the terminal to the source base station, wherein the capability information of the terminal includes AI / ML-related information of the terminal.

22. In Paragraph 20, The above measurement control report is a handover method including AI / ML-related information of the terminal.

23. In Paragraph 20, A handover method further comprising: a step of transmitting a report of the inference value of the reused AI / ML model to the target base station when the terminal reuses the AI / ML model used at the source base station after the handover is completed; and a step of receiving a management instruction for the reused AI / ML model from the target base station.

24. In Paragraph 20, A handover method further comprising: a step of reporting performance monitoring results of a reused AI / ML model; and a step of receiving management instructions for the reused AI / ML model from the target base station.

25. In Paragraph 20, After the handover is completed, the terminal reuses the AI / ML model used at the source base station, but if the AI / ML association ID used by the terminal does not match the association ID supported by the target base station, A handover method comprising: receiving reference signals corresponding to a plurality of associated IDs supported by the target base station from the target base station; and providing a monitoring result based on the measurement result of the reference signals corresponding to the plurality of associated IDs to the target base station.

26. Communications Department; and In a terminal including a processor, The processor performs the steps of: receiving a measurement control report of a terminal; determining a handover based on the measurement control report of the terminal; and transmitting a handover request to a target base station. The above handover request is a base station containing AI / ML-related information of the terminal.

Images